Arkebe Oqubay, Christopher Cramer, Ha-Joon Chang, and Richard Kozul-Wright - The Oxford Handbook of Industrial Policy-Oxford University Press (2020)

Copyright Page
Copyright Page
Edited by Arkebe Oqubay, Christopher Cramer, Ha-Joon Chang, and Richard
Kozul-Wright
The Oxford Handbook of Industrial Policy
Edited by Arkebe Oqubay, Christopher Cramer, Ha-Joon Chang, and Richard Kozul-Wright
Print Publication Date: Oct 2020 Subject: Economics and Finance

Online Publication Date: Nov 2020
(p. iv) Copyright Page
Great Clarendon Street, Oxford, OX2 6DP,

United Kingdom
Oxford University Press is a department of the University of Oxford.

It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide. Oxford is a registered trade mark of
Oxford University Press in the UK and in certain other countries
© Oxford University Press 2020
The moral rights of the authors have been asserted
First Edition published in 2020
Impression: 1
All rights reserved. No part of this publication may be reproduced, stored in

a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by licence or under terms agreed with the appropriate reprographics
rights organization. Enquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above
You must not circulate this work in any other form

and you must impose this same condition on any acquirer
Published in the United States of America by Oxford University Press
Page 1 of 2
PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).
Copyright Page
198 Madison Avenue, New York, NY 10016, United States of America
British Library Cataloguing in Publication Data

Data available
Library of Congress Control Number: 2020937433
ISBN 978–0–19–886242–0
Printed and bound by

CPI Group (UK) Ltd, Croydon, CR0 4YY
Links to third party websites are provided by Oxford in good faith and
for information only. Oxford disclaims any responsibility for the materials
contained in any third party website referenced in this work.
Page 2 of 2
Acknowledgements
Acknowledgements
Kozul-Wright

(p. v) Acknowledgements
Working on this exciting yet challenging scholarly project has involved close to fifty au
thors. We have enjoyed the full support of contributors who put significant time and effort
into this volume. The Handbook has benefited from constructive comments provided by
internal and anonymous external reviewers. In particular, the editors are very grateful to
all the contributors of this Handbook and thank them sincerely for their exceptional ef
forts and their personal commitment to ensuring the success of this volume.
Special thanks are due to our commissioning editor, Adam Swallow, and his team at Ox
ford University Press, as well as to the entire production team for their advice, support,
and encouragement. We also wish to thank the anonymous reviewers and delegates of
OUP who approved our book proposal. We are grateful to Eva Paus, Mount Holyoke Col
lege; Edlam Yemeru, United Nations Economic Commission for Africa (UNECA), Addis
Ababa; Fantu Cheru, Leiden University; and Dirk Willem te Velde, ODI; as well as all peer
reviewers, who provided invaluable guidance on the structure and content of the book.
This gratitude is also extended to all the participants at both the inception and chapter
review workshops held in Addis Ababa in April 2019 and September 2019 for their in
sightful presentations, discussions, feedback, and collaboration.
Our sincere appreciation goes to Mr Tewolde Gebremariam, Group CEO Ethiopian Air
lines, and his team, and the Ethiopian Airlines Skylight Hotel, for their support of this
project. Special thanks are also due to Keith Povey Editorial Services Ltd for the pre-pub
lication copy-editing of the volume. We also thank Nigisty Gebrechristos and Ayalew
Mamo for their continued support of this project. And finally, our appreciation and grati
tude goes to our project and research coordinator, Deborah M. Kefale, and the team of
Helena Alemu, Tsion Kifle, Meron Tilahun, for their dedicated support in the preparation
of the manuscript and assistance throughout the project.
Page 1 of 2
Acknowledgements
Every effort has been made to trace and contact copyright holders to reproduce quota
tions in this volume prior to going to press however if notified the publisher will under
take to rectify any errors or omissions at the earliest opportunity.
Page 2 of 2
List of Figures
List of Figures
Kozul-Wright

(p. xi) List of Figures

3.1 Productivity and GDP dynamics 74
4.1 Empirical density (Pr, vertical axis) of labour productivities, whole manufacturing
of China, France, and Italy, 1998, 2002, and 2006 100
4.2 Scatterplot of log (VA/L) versus log (K/L) for corn milling sector (CIC 131), 1998
104
4.3 Empirical density of (log) labour productivity of exporters and non-exporters of
transport equipment (CIC 37) and electrical machinery and equipment (CIC 39) sec
tors in selected years (1998, 2003, and 2007) 110
9.1 Stylized manufacturing value chain smile curve 247
9.2 Evolution of GVC positioning in selected developing economies (participation and
upstreamness), 2005–11 248
9.3 Evolution of GVC positioning in selected developing economies (downstreamness
and upstreamness), 2005–11 249
9.4 Sectoral positioning of BRIC economies, 2005–11 250
10.1 Learning curve associated with solar PV power 268
10.2 India’s solar PV installed capacity, 2010–19 273
10.3 China: Rising proportion of electric power sourced from water, wind, and sun,
1990–2018 275
12.1 A mission-oriented approach to cleaning the oceans 324
14.1 Industrial ecosystem framework 380
16.1 Change in women’s relative concentration in industrial-sector employment,
1991–2017 435
17.1 An export-led growth model with cumulative causation 457
17.2 Successful industrial policy raises equilibrium output growth and productivity
growth but reduces employment growth 458
17.3 Trade-off between labour productivity growth and employment growth in twen
ty-four late industrializing countries, 1950–2003 459
Page 1 of 3
List of Figures
17.4 An export-led growth model with cumulative causation and a balance-of-pay

ments constraint 464
(p. xii) 17.5 Successful industrial policy raises equilibrium output growth above the
BoP-constrained growth rate 466

17.6 The macroeconomic impacts of a commodity-price boom 470
17.7 Escaping the primary-commodity specialization trap 472
17.8 The (+/–) impacts of a higher real wage on the profit rate (ρ) 478
18.1 Industry 4.0+ 488
18.2 The sad curve 489
19.1 The frequency of ‘David Ricardo’ (in English) during the first 100 years after
the 1817 publication of his main work, Principles of Economics, compared to that of
two other, then much more famous, English economists 509
19.2 Frequency of the term ‘comparative advantage’ (in English) from 1817 until to
day 509
19.3 The cover of the first edition of the book that laid the earliest foundations for a
theory of industrial policy: Botero’s 1589 volume 515
19.4 An example of the practical consequences of ‘the cult of value added’ 518
19.5 von Thünen’s map of a modern state, with the industrial city at its core 519
19.6 How industries differ 526
19.7 Productivity explosion: cotton spinning 528
19.8 US learning curve in men’s shoes, 1850‒1936 528
19.9 The quality index of economic activities 531
19.10 Increasing diversity and specialization over time (= ‘tid’) 543
19.11 Economic growth since 1989 (fall of the Berlin Wall), selected countries: per
centiles of population with income growth above/below the 1989 level / the G7 aver
age level 550
21.1 Manufacturing value added as a percentage of GDP, selected countries, 1991–
2018 610
22.1 The industrial policy structure of the EU 629
22.2 Spending on industrial policy in the European Union, average 2014–17 635
22.3 Funding of industrial policy in the European Union by themes, average 2014–17
638
22.4 Spending on industrial policy from the EU and from member states, by country
groups, average 2014–17 641
22.5 Spending on industrial policy from the EU budget, by country groups and
themes, average 2014–17 643
(p. xiii) 22.6 Share of Horizon 2020 applications per EU member state, average
2014–16 646
24.1 Innovation-driven industrial policy 685
24.2 Manufacturing value added share in GDP, selected economies, 1960–2017 695
24.3 Manufactured exports share in total merchandise exports, selected economies,
1962–2017 696
24.4 Electric and electronics exports, selected economies, 2001–18 698
24.5 Motor vehicle exports, selected economies, 2001–18 698
Page 2 of 3
List of Figures
24.6 Trade balance, electric and electronics, selected economies, 2001–18 699
24.7 Trade balance, motor vehicles, selected economies, 2001–18 700
24.8 GERD in GDP, South East Asian market economies, 1996–2014 (%) 706
24.9 Intellectual property trade balance, selected economies, 1981–2018 710
26.1 BRICS’ growing but heterogeneous contribution to global economic dynamics
758
26.2 BRICS: structural change and catching up, 1990‒2015 760
27.1 GDP change in economies of the former Soviet Union, 1989 = 100% 782
27.2 Share of manufacturing value added in GDP in post-communist transition coun
tries between 1985 and 2017, % 784
27.3 Share of industry value added in GDP of post-communist countries, 1985–2017,
% 785
27.4 Share of service value added in GDP in selected post-communist countries,
1985–2017, % 786
27.5 Manufacturing exports, % of merchandise export, in some post-communist
countries 787
27.6 R&D (research and development) Expenditure in selected countries, % of GDP
798
27.7 The share of private sector in GDP in some former Soviet republics, 1989–2009,
% 801
27.8 Public, private, and gross investment in developing countries as a % of GDP,
2012 802
28.1 Investment ratio and capital account openness in Latin America 821
28.2 Real exchange rate (2005 = 100) 822
28.3 Relative productivity vs. United States 830
28.4 Latin America: Share of manufacturing in GDP, 1950–2017 832
(p. xiv) 28.5 Share of manufacturing in GDP: Largest Latin American countries vs.
Korea 833
28.6 Index of the share of engineering industries in manufacturing value added vs.
United States 834
28.7 Research and development as % of GDP 834
28.8 Strength of the National Systems of Innovation (NIS5) 835
Page 3 of 3
List of Tables
List of Tables
Kozul-Wright

(p. xv) List of Tables

4.1 Annual growth rate of labour productivity amongst firms over 1998–2007, and
subperiods 1998–2002 and 2002–2007 amongst ‘continuing’ firms (i.e firms remain
ing in the same two-digit sector over the relevant period) 99
4.2 Median capital intensity (capital/output ratios) by sector, China, Italy, and
France, 1998, 2002, and 2006 102
4.3 Ratio of the average labour productivity of the second highest decile over the
second lowest decile, China, 1998, 2002, and 2007 105
4.4 Contribution of each two-digit sector to total manufacturing value added, China,
1998, 2002, and 2007 (percentages) 106
4.5 Within-sector learning vs. structural change in productivity growth 107
5.1 Policy hierarchy: industry support by alternative measures 128
9.1 Uneven returns in the iPod value chain 254
15.1 Thirty-year global GDP growth trajectory, 2021–50 409
15.2 Global energy demand and energy-efficiency cost projections for 2050 410
15.3 Global clean renewable energy expansion and cost projection for 2050 410
15.4 Costs of thirty-year clean energy investment project as share of average GDP,
2021–50 411
A15.1 Energy intensity ratios, global average and selected countries 423
A15.2 Estimates of cost savings from energy-efficiency investments 424
A15.3 Average global levellized costs of electricity from utility-scale renewable ener
gy sources vs. fossil-fuel sources, 2010–17 424
A15.4 Capital expenditure costs for building renewable electricity productive equip
ment, present values of total lump-sum capital costs per Q-BTU of electricity 424
A15.5 Major funding sources for global clean energy investments 425
A15.6 Change in overall domestic content of clean energy investment activities after
20 per cent import increase with tradable activities 425
Page 1 of 2
List of Tables
A15.7 Reliance on fossil fuels and imports as energy sources in selected countries,
2014 426
(p. xvi) 16.1 Gender regimes in BRICS and selected developing and developed coun
tries 441
17.1 Late industrializing economies: stylized facts and parameter values 454
17.2 Income elasticity of exports (ε) and imports (μ): selected late industrializing
countries, 1962–2006/2014 462
17.3 Impact on economic growth of a one percentage point increase in real wage
growth, selected late industrializing countries 475
17.4 Share of wages in the gross output price, selected regions and countries 475
19.1 How the game changed 527
19.2 Characteristics of the collusive and classical modes of diffusion of productivity
improvements 530
19.3 ‘Good’ and ‘bad’ economic activities 532
19.4 Average wage per cluster category in Europe, based on 255 European regions
532
22.1 The three phases of industrial policy in the European Union 625
22.2 EU and member states? spending on industrial policy by policy field, 2014–17
(annual averages) 637
A22.1 List of member states, country codes, and country groups 654
A22.2 Industrial policy item within the MFF 2014?20 655
A22.3 Categorization of the thematic objectives in the ESIF into industrial policies
themes 656
25.1 China’s National Champions in Fortune Global 500, 2018 736
25.2 National Champions in the top 100 companies ranked by R&D in mainland Chi
na 740
26.1 Composition of gross manufacturing exports by country of origin of value
added, 2005–15 763
26.2 UNCTAD FDI estimates by ultimate investor, share in inward FDI stock, 2017
764
27.1 R&D expenditure in post-communist countries as a percentage of GDP, 2015
799
30.1 Manufacturing performance during early ISI, 1960–80 871
30.2 Manufacturing performance during SAPs, 1981–2000 879
30.3 Manufacturing performance post SAPs, 2001‒17 888
Page 2 of 2
List of Contributors
Kozul-Wright

(p. xvii) List of Contributors
Arkebe Oqubay
(PhD) is a senior minister and special adviser to the prime minister of Ethiopia and
has been at the centre of policymaking for over twenty-five years. He is the former
mayor of Addis Ababa, winner of the Best African Mayor of 2005 and finalist in the
World Mayor Award 2005, for his work transforming the city. He is a recipient of the
Order of the Rising Sun, Gold and Silver Star, presented by the Emperor of Japan. He
currently serves as board chair of several leading public organizations and interna
tional advisory boards. He is an ODI Distinguished Fellow and a research associate at
the Centre of African Studies in the University of London, and holds a PhD in develop
ment studies from SOAS, University of London. His recent works include the path-
breaking Made in Africa: Industrial Policy in Ethiopia (OUP, 2015); The Oxford Hand
book of the Ethiopian Economy (OUP, 2019); How Nations Learn: Technological
Learning, Industrial Policy, and Catch-up (OUP, 2019); China‒Africa and an Economic
Transformation (OUP, 2019); African Economic Development: Evidence, Theory, and
Policy (OUP, 2020); and The Oxford Handbook of Industrial Hubs and Economic Devel
opment (OUP, 2020). He was recognized as one of the 100 Most Influential Africans of
2016, and a ‘leading thinker on Africa’s strategic development’ by the New African,
for his work, both theoretical and practical, on industrial policies.
Christopher Cramer
Page 1 of 16
is professor of the political economy of development at SOAS, University of London.

He is a vice-chair of the Royal Africa Society and chair of the Scientific Committee of
the African Programme on Rethinking Development Economics (APORDE). His publi
cations include Civil War Is Not A Stupid Thing: Accounting for Violence in Develop
ing Countries (Hurst Publishers, 2006), African Economic Development: Evidence,
Theory, and Policy (OUP, 2020) and The Oxford Handbook of the Ethiopian Economy
(OUP, 2019, co-edited with Cheru and Oqubay). He led the research project Fairtrade,
Employment, and Poverty Reduction in Ethiopia and Uganda.
Ha-Joon Chang
teaches economics at the University of Cambridge. His main books include Kicking
Away the Ladder (2002), Bad Samaritans (2007), 23 Things They Don’t Tell You about
Capitalism (2011) and Economics: The User’s Guide (2014). His writing has been
translated into forty-one languages in forty-four countries. Worldwide, his books have
sold over 2 million copies. He is the winner of the 2003 Gunnar Myrdal Prize and the
2005 Wassily Leontief Prize.
(p. xviii) Richard Kozul-Wright
is director of the Globalisation and Development Strategies Division in UNCTAD. He

has worked at the United Nations in both New York and Geneva. He holds a PhD in
economics from the University of Cambridge. He has published widely on economic
issues including, inter alia, in The Economic Journal, Cambridge Journal of
Economics, The Journal of Development Studies, and Oxford Review of Economic Poli
cy. He is the author of many books, including The Rise and Fall of Global Microcredit:
Development, Debt and Disillusion (2018, with S. Blankenburg and M. Bateman), Se
curing Peace: State-Building and Economic Development in Post-Conflict Countries
(2011, with P. Fortunato), Climate Protection and Development (2012, with Frank Ack
erman) and The Resistible Rise of Market Fundamentalism (2008, with Paul Ray
ment). His edited volumes include Transnational Corporations and the Global Econo
my, Economic Insecurity and Development, Securing Peace, Climate Protection and
Development, and Industrial Policy. He is a frequent contributor to newspapers world
wide on economic issues, including the Financial Times, The Guardian, and Project
Syndicate.
Page 2 of 16
Antonio Andreoni
is associate professor in industrial economics at University College London (UCL),

and research head of UCL Institute for Innovation and Public Purpose. He is also visit
ing associate professor SARChI Industrial Development, University of Johannesburg.
He has acted as an adviser for the World Bank, OECD, DFID, UNIDO, UNCTAD, ILO,
UNDP, UNECA, UNU-WIDER, EU Growth and Innovation, and to governments in
South Africa, Tanzania, and the United Kingdom. His work on structural change, pro
duction, technology, innovation, financialization, corruption, political economy, and in
dustrial policy has appeared in Cambridge Journal of Economics, Development and
Change, Cambridge Journal of Regions, Economy and Society, Structural Change and
Economic Dynamics, Oxford Review of Economic Policy, European Journal of Develop
ment Research, and Energy Policy.
Sam Ashman
is associate professor in the School of Economics at the University of Johannesburg

where she is director of the UJ-IDEP MPhil in industrial policy programme. Her re
search focuses on the financialization of the world economy and its various impacts—
on firm behaviour, patterns of development, the shaping of economic and social poli
cy, and the worsening of inequality. She also researches industrial policy and develop
ment issues, including the impact of global value chains, the evolving political econo
my of South Africa, and the history and philosophy of economics and political econo
my.
David Bailey
is professor of business economics at the University of Birmingham and senior fellow

of the ESRC’s ‘UK in the Changing Europe’ programme, exploring the impacts of
Brexit on the UK automotive and manufacturing industries. He has written extensive
ly on industrial and regional policy, especially in relation to manufacturing and the au
to industry.
Page 3 of 16
Michael H. Best
has been actively engaged in industrial restructuring policy in London, Massachu

setts, Slovenia, Cyprus, Jamaica, Honduras, Malaysia, Northern (p. xix) Ireland, and
the Republic of Ireland, and has written production audits based on extensive factory
visits in many other countries. His three books on industrial policy are The New Com
petition: Institution of Industrial Restructuring (HUP, 1990), The New Competitive Ad
vantage: The Renewal of American Industry (Oxford University Press, 2001), and How
Growth Really Happens: The Making of Economic Miracles through Production, Gov
ernance, and Skills (Princeton University Press, 2018) which won the 2018 Schum
peter Prize.
Patrizio Bianchi
is professor of economics at the University of Ferrara, Italy, and has been, since 2010,
minister for education and research in the regional government of the Emilia-Ro
magna region. Besides his academic activity, he has been an adviser to Italian institu
tions such as the Prime Minister’s Office, the Minister of Industry’s Cabinet and the
regional government of Emilia-Romagna, and to international institutions such as the
European Commission, the BID, and UNIDO. Patrizio has more than 200 publications,
including books and articles in scientific journals. Publications on industrial policy in
clude Industrial Policy after the Crisis: Seizing the Future (Edward Elgar, 2011) and
Industrial Policy for the Manufacturing Revolution: Perspectives on Digital Globalisa
tion (Edward Elgar, 2018) with Sandrine Labory.
Muyang Chen
is an assistant professor at the School of International Studies, Peking University. Her

research interests include infrastructure finance, development banking, international
development, and state‒market relations. She has been a visiting scholar at the Na
tional Graduate Institute for Policy Studies (Japan) and a pre-doctoral fellow at the
Global Development Policy Center of Boston University. She received her PhD from
the University of Washington, an MA from University of California, Berkeley, and BAs
from Peking University and Waseda University.
Page 4 of 16
Horman Chitonge
is a professor at the Centre for African Studies, University of Cape Town (UCT). His
research interests include agrarian political economy, social welfare, and alternative
strategies for economic growth in Africa. His most recent books include: Industrialis
ing Africa: Unlocking the Economic Potential of the Continent (Peter Lang, 2019), So
cial Welfare Policy in South Africa: From the Poor White Problem to a Digitised Social
Contract (Peter Lang, 2018), Contemporary Customary Land Issues in Africa: Navigat
ing the Contours of Change (Cambridge Scholars Publishing, 2017) and Economic
Growth and Development in Africa: Understanding Trends and Prospects (Routledge,
2015).
Mario Cimoli
is deputy executive secretary of the United Nations Economic Commission for Latin
America and the Caribbean (ECLAC). In 2004, he was appointed co-director (with
Giovanni Dosi and Joseph Stiglitz) of two task forces: Industrial Policy and Intellectual
Property Rights Regimes for Development (Initiative for Policy Dialogue, Columbia
University, New York). Amongst his broad interests are economic development and its
relationship to production structure, productivity growth, international trade, and
structural change. His work analyses the linkages between industrial policy, technolo
gy development and innovation, and their role in shaping development trajectories.
(p. xx) Lisa De Propris
is professor of regional economic development, University of Birmingham. She has

written extensively on competitiveness in clusters and regions; technological change
and innovation; clusters and globalization; and regional and industrial policy.
Jonathan Di John
is a senior lecturer in the political economy of development at SOAS, University of

London. His main areas of expertise are development economics, institutional eco
Page 5 of 16
nomics, and the political economy of growth and development in Latin America, espe
cially in Brazil and Venezuela. His research focuses on the political economy of indus
trial strategy, taxation, development strategies in oil economies, and social service de
livery in violent contexts. He has undertaken consultancy work for the World Bank,
United Nations Development Programme (UNDP), the UK Department for Interna
tional Development (DFID), the Organisation of Economic Cooperation and Develop
ment (OECD) and the International Monetary Fund (IMF).
Budi A. Djafar
is a PhD candidate in economics at the New School for Social Research. His disserta
tion investigates the implications of international trade and the rise of global produc
tion networks for national income and gender-based employment in the manufactur
ing sector in Indonesia from 1970 to 2010. Budi previously served in the Ministry of
Foreign Affairs of the Republic of Indonesia and the Presidential Advisor’s Council of
the Republic of Indonesia, 2010‒14.
Giovanni Dosi
is professor of economics at Sant’Anna School of Advanced Studies, Pisa, Italy, and

continental Europe editor of the journal Industrial and Corporate Change. He is in
cluded in the ISI Highly Cited Research list of scholars who have made fundamental
contributions to the advancement of science and technology, and is a corresponding
member of the Accademia Nazionale dei Lincei, the first academy of sciences in Italy.
In 2016, he received the Wiley TIM Distinguished Scholar Award from the Technology
and Innovation Management Division of the American Academy of Management. His
major research areas, in which he is author and editor of several works, include the
economics of innovation and technological change, industrial economics, evolutionary
theory, economic growth and development, and organizational studies.
Daniel E. Esser
is an associate professor in the American University’s School of International Service

in Washington, DC. He holds MSc and PhD degrees from the London School of Eco
nomics and Political Science. His current book project focuses on sociopolitical
Page 6 of 16
processes forging perceptions of legitimacy where democratic institutions are frac

tured or absent. He also continues to pursue scholarly interests in the politics of sub-
national governance and communities’ responses to violence and terror. His research
frequently takes him to Latin America and South Asia, and its findings have been pub
lished in leading international journals and two Oxford Handbooks.
Piergiuseppe Fortunato
is an economist at the United Nations Conference for Trade and Development. He

previously worked in the Department of Economic and Social (p. xxi) Affairs of the
United Nations in New York and held positions at the University of Bologna and Uni
versité Paris I. His research interests include political economics, economic develop
ment, and industrial policy. In these fields he has published in journals including Jour
nal of Economic Growth, The Economic Journal, and European Economic Review. He
is also heavily involved in policy-oriented research, has edited a volume on post-con
flict recovery, published in policy journals such as Global Policy and contributes to
several blogs.
Rainer Kattel
is an Estonian academic and science administrator. He is deputy director and profes

sor of innovation and public governance at the Institute for Innovation and Public Pur
pose, University College London.
Sandrine Labory
is associate professor of economics at the University of Ferrara, Italy. She has a MSc
in economics from University College London, and a PhD from the European Universi
ty Institute in Florence, Italy. Her research is focused on industrial economics and
policy, including comparative analysis of national and regional industrial policies, in
novative and productive processes at firm and territorial levels, structural changes
and industrial development. She has published numerous articles in Italian and inter
national journals, as well as books such as Industrial Policy after the Crisis: Seizing
Page 7 of 16
the Future (Edward Elgar, 2011) and Industrial Policy for the Manufacturing Revolu
tion: Perspectives on Digital Globalisation (2018, Edward Elgar) with Patrizio Bianchi.
Michael A. Landesmann
was scientific director of the Vienna Institute for International Economic Studies (wi
iw) from 1996 to 2016 and is professor of economics at Johannes Kepler University,
where he is head of the Department of Economic Theory and Quantitative Research.
His research focuses on international economic relations, European economic inte
gration, globalization and labour markets, and migration. He has a DPhil from the
University of Oxford, was a lecturer and fellow (Jesus College) at the University of
Cambridge, and has held visiting professorships at Harvard University (Pierre Keller,
Schumpeter) and a range of other universities (Brandeis, Jerusalem, Central Euro
pean University, Padova, Basel, Osaka, Mumbai).
Peter Lawrence
is emeritus professor of economics at Keele University, United Kingdom. He holds a

BA and MA from the University of Sussex and a PhD from the University of Leeds. He
has taught and researched in several African countries and published variously on
rural and industrial development, macroeconomic policy, and development strategy.
Keun Lee
is a professor of economics at the Seoul National University, fellow of the CIFAR pro
gramme on Innovation, Equity and Prosperity, and founding director of the Center for
Economic Catch-up. He is an editor of Research Policy and an associate editor of In
dustrial and Corporate Change. He served as president of the International Schum
peter Society (2016‒18), a member of the Committee for Development Policy of UN
(2014‒18), and a council member of the World Economic Forum (2016‒19). He is
(p. xxii) the winner of the 2014 Schumpeter Prize for his monograph Schumpeterian
Analysis of Economic Catch-up (Cambridge University Press, 2013), as well as the

2019 Kapp Prize from the EAEPE for his article on national innovation systems. He
obtained his PhD in economics from the University of California, Berkeley. One of his
most cited articles is the paper ‘Korea’s Technological Catch-up’ published in Re
Page 8 of 16
search Policy, with 1,100 citations (Google Scholar). His H-index is 40, with about 100
papers with more than ten citations. His latest book is The Art of Economic Catch-up:
Barriers, Detours, and Leapfrogging (Cambridge University Press, 2019).
Chen Li
is an assistant professor at the Centre for China Studies, The Chinese University of
Hong Kong (CUHK). He is also an assistant professor (by courtesy) at CUHK’s Lau
Chor Tak Institute of Global Economics and Finance (IGEF). He has researched and
written extensively on issues related to China’s economic development, business envi
ronment, and public policies, such as China’s state-owned enterprise reform, financial
regulatory reform, government‒business relations, and industrial and regional devel
opment. He received his PhD and MPhil in development studies from the University of
Cambridge and dual bachelor’s degrees in law and economics from Peking University.
John A. Mathews
is professor emeritus in the Faculty of Business and Economics at Macquarie Univer

sity, Sydney. He was a professor of strategy at Macquarie Graduate School of Manage
ment for twenty years, retiring from active teaching in 2018. From 2009–12 he held
concurrently Enr Chair of Competitive Dynamics and Global Strategy at LUISS Gardo
Carli University in Rome. For the past several years he has focused on the greening of
industry with an emphasis on the role of China. The year 2017 saw the publication of
his Global Green Shift by Anthem Press, London. In July 2018 he was awarded the bi-
annual Schumpeter Prize in recognition of his work and most recent book.
Mariana Mazzucato
is professor of the economics of innovation and public value at University College

London (UCL), where she is founding director of the UCL Institute for Innovation &
Public Purpose (IIPP). She is winner of the 2018 Leontief Prize for Advancing the
Frontiers of Economic Thought and the 2019 All European Academies Madame de
Staël Prize for Cultural Values. She is author of many publications including The En
trepreneurial State: Debunking Public vs. Private Sector Myths (2013) and The Value
Page 9 of 16
of Everything: Making and Taking in the Global Economy (2018). She advises global
policymakers on innovation-led growth.
William Milberg
is dean and professor of economics at The New School for Social Research and direc
tor of the Heilbroner Center for Capitalism Studies at The New School. His research
focuses on the relation between globalization, income distribution, and economic
growth, and the history and philosophy of economics. His most recent book is Out
sourcing Economics: Global Value Chains in Capitalist Development, co-authored with
Deborah Winkler. He has served as a consultant to UNCTAD, ILO, (p. xxiii) WTO, and
the World Bank. He is a member of the Advisory Board for the Center for Business
and Human Rights at New York University.
James H. Mittelman
is distinguished research professor and university professor emeritus at the American

University. Previously, he was professor and dean at Queens College, City University
of New York; professor and dean, Graduate School of International Studies (today the
Korbel School), University of Denver; and Pok Rafeah chair, National University of
Malaysia. He received the International Studies Association’s Distinguished Scholar
award in international political economy, and is an honorary fellow at the Helsinki
Collegium for Advanced Studies. His books include The Globalization Syndrome:
Transformation and Resistance (Princeton University Press, 2000) and Implausible
Dream: The World-Class University and Repurposing Higher Education (Princeton
University Press, 2017).
Susan Newman
is professor of economics at the Open University, United Kingdom. She has published
in areas including finance and industrial policy, the political economy of industrial de
velopment in South Africa, and financialization and the restructuring of production.
She is particularly interested in the dynamics of wealth and income distribution along
global value chains. She is a member of Reteaching Economics and the International
Page 10 of 16
Initiative for Promoting Political Economy and is committed to the promotion of politi
cal economy in economics education.
José Antonio Ocampo
is professor at Columbia University’s School of International and Public Affairs, and

chair of the Committee for Development Policy of the United Nations Economic and
Social Council (ECOSOC). He has occupied numerous positions at the United Nations
and his native Colombia, including UN under-secretary-general for economic and so
cial affairs, executive secretary of the UN Economic Commission for Latin America
and the Caribbean (ECLAC), and member of the board of directors of Banco de la
República (Colombia’s central bank), minister of finance, minister of agriculture and
director of the National Planning Office of Colombia.
Robert Pollin
is distinguished university professor of economics and co-director of the Political

Economy Research Institute (PERI) at the University of Massachusetts-Amherst. He is
also the founder and president of PEAR (Pollin Energy and Retrofits), an Amherst,
MA-based green energy company operating throughout the United States. His books
include The Living Wage: Building a Fair Economy (co-authored, 1998), Contours of
Descent: U.S. Economic Fractures and the Landscape of Global Austerity (2003), An
Employment-Targeted Economic Program for South Africa (co-authored, 2007), Back
to Full Employment (2012), Greening the Global Economy (2015), and The Political
Economy of Climate Change and the Global Green New Deal (co-authored, forthcom
ing 2020).
Vladimir Popov
is a principal researcher in the Central Economics and Mathematics Institute of the

Russian Academy of Sciences. He is also professor emeritus at the New Economic
School in Moscow, and an adjunct research professor at the Institute of (p. xxiv) Euro
pean and Russian Studies at Carleton University in Ottawa. He has published exten
sively on world economy and development issues; he is editor of four books, and au
thor of hundreds of articles and ten books including Mixed Fortunes: An Economic
Page 11 of 16
History of China, Russia, and the West (Oxford University Press, 2014; also published
in Chinese).
Gabriel Porcile
is an economic affairs officer at ECLAC and professor of economics at the Federal

University of Parana, Brazil. He holds a PhD in economic history from the London
School of Economics. He has been a visiting scholar at the University of Sao Paulo,
Brazil, University of the Republic, Uruguay and Columbia University, United States.
His main research topics are economic growth, distribution, structural change, and
the political economy of development. His latest publications are ‘New Structuralism
and the Balance-of-Payments Constraint’, Review of Keynesian Economics, 7, 2019
(with Giuliano Yajima) and ‘A Technology Gap Interpretation of Growth Paths in Latin
America and Asia’, Research Policy, 48, 2019 (with Mario Cimoli and Joao Basilio
Pereima).
Rajah Rasiah
currently holds the position of distinguished professor of economics at the University

of Malaya. His previous positions were professor of industrial organization at Nation
al University of Malaysia, and senior research fellow at UNU-INTECH, Maastricht. He
obtained his doctorate in economics from the University of Cambridge in 1992. He
was awarded the Celso Furtado Prize by the World Academy of Sciences in 2014 for
advancing social science thought (economics).
Erik S. Reinert
a Norwegian citizen, is a professor at Tallinn University of Technology and guest re

searcher at the Centre for Sciences and the Humanities at the University of Bergen.
His research area is the theory of uneven growth, that is, the factors which—contrary
to the predictions of standard economic theory—cause world economic development
to be such an uneven process. He holds a BA from Hochschule St. Gallen, Switzer
land, an MBA from Harvard University and a PhD in economics from Cornell Universi
Page 12 of 16
ty. His book How Rich Countries Got Rich and Why Poor Countries Stay Poor has been
published in more than twenty languages.
Simon Roberts
is a professor of economics at the University of Johannesburg and an economics direc

tor at the UK Competition and Markets Authority. He founded and directed the Cen
tre for Competition, Regulation and Economic Development (CCRED) at the Universi
ty of Johannesburg. Prior to this he was chief economist at Competition Commission
South Africa. He has worked extensively on issues of industrial development, trade,
regional value chains, competition and economic regulation in Southern and East
Africa, advising governments, competition authorities, and regulators. He has also
testified as an expert witness in a number of major competition cases.
Fernando Santiago
is an industrial policy officer at the United Nations Industrial Development Organiza

tion (UNIDO) where he specializes in the design, monitoring, (p. xxv) and evaluation of
industrial policies. He is also a member of the scientific board of the Latin American
Network for the Study of Learning Systems, Innovation and Competency Building
(LALICS). He has published on topics related to innovation and industrial policy. He is
an economist from the National Autonomous University of Mexico and holds a
master’s degree from SPRU-University of Sussex and a PhD in economics and policy
studies of technical change from UNU-MERIT/University of Maastricht.
Stephanie Seguino
is professor of economics, University of Vermont, United States and fellow, Gund In
stitute for the Environment. Her research explores the relationship between inter
group inequality by class, race, and gender on the one hand, and economic growth
and development on the other. She is past president of the International Association
for Feminist Economics and an associate editor of Feminist Economics, Journal of Hu
man Development and Capabilities and Review of Keynesian Economics. Her work has
appeared in Cambridge Journal of Economics, Development and Change, Structural
Change and Economic Dynamics, and World Development, amongst others. She has
Page 13 of 16
consulted with numerous international organizations, including the UNDP, UNRISD,

UNCTAD, and the World Bank.
Roman Stöllinger
is a staff economist at the Vienna Institute for International Economic Studies (wiiw).
In his research he focuses on issues related to international trade and global value
chains, industrial policy and structural change. Recent publications include a paper
on the impact of global value chains on structural change and the need for appropri
ate industrial policies in Europe and its neighbourhood. He is a regular contributor to
research reports by international institutions such as the European Commission and
UNIDO. He is strongly involved in the Research Centre International Economics
(FIW), an Austrian think-tank and infrastructure platform in the field of international
economics. Since 2012 he has also been a lecturer at the Vienna University of Eco
nomics and Business, teaching international macroeconomics and industrial policy.
Before joining wiiw in 2008, he worked at OeKB, the Austrian Export Credit Agency,
where he gained real-world experience in implementing state support policies. He
holds a master’s degree in international economics from the University of Innsbruck
and a PhD in economics from the University of Vienna.
Servaas Storm
is a senior lecturer at Delft University of Technology. He obtained a PhD in economics

from Erasmus University Rotterdam. His work has appeared in scholarly journals in
cluding Cambridge Journal of Economics, Development and Change, Eastern Econom
ic Review, Industrial Relations, International Journal of Political Economy, Journal of
Post-Keynesian Economics, and Structural Change and Economic Dynamics. His latest
book, co-authored with C.W.M. Naastepad, is Macroeconomics beyond the NAIRU
(Harvard University Press, 2012), winner of the 2013 Myrdal Prize of the European
Association for Evolutionary Political Economy. He is one of the editors of Develop
ment and Change.
(p. xxvi) Fiona Tregenna
Page 14 of 16
holds the DST/NRF South African Research Chair in Industrial Development and is al
so a professor of economics at the University of Johannesburg. She is appointed to
policy bodies including the Competition Tribunal and the Presidential Economic Advi
sory Council, sits on a number of boards, and consults for various international orga
nizations. She has a PhD in economics from the University of Cambridge, a master’s
degree in economics from the University of Massachusetts, and earlier degrees from
the Universities of the Witwatersrand and Natal (South Africa). Her primary research
interest is in issues of structural change, deindustrialization, and industrial develop
ment.
John Weiss
is emeritus professor of development economics at the University of Bradford, United

Kingdom. He has written extensively on issues of cost‒benefit analysis, industrializa
tion, and industrial policy, publishing several textbooks, including Project Analysis in
Developing Countries (with Steve Curry, MacMillan, 2000), Industry in Developing
Countries (Routledge, 1990) and The Economics of Industrial Development
(Routledge, 2011). He is the joint editor of the Routledge Handbook on Industry and
Development (Routledge, 2016). He has been a consultant to, among others, the Unit
ed Nations Industrial Development Organization (UNIDO), the World Bank, and the
Asian Development Bank.
Lindsay Whitfield
is professor (with special responsibilities) in Global Studies and leader of the Centre
of African Economies in the Department of Social Sciences and Business, Roskilde
University, Denmark. She is author of several books on African politics and
economies, including The Politics of African Industrial Policy: A Comparative Perspec
tive (Cambridge University Press, 2015) and Economies after Colonialism: Ghana and
the Struggle for Power (Cambridge University Press, 2018).
Xiaodan Yu
Page 15 of 16
is an assistant professor in the Department of Entrepreneurship and Innovation at

Nottingham University Business School, China. Prior to joining NUBS, she held re
search positions at the National Research Council and Scuola Superiore Sant’Anna in
Italy. She received her BEng from Beihang University in China, MSc from University
of Cincinnati, and PhD in economics from Scuola Superiore Sant’Anna in 2014. Her
research interests include the economics of innovation, industrial dynamics, firm
growth, and entrepreneurship.
Nimrod Zalk
is industrial development adviser at the South African Department of Trade and In
dustry (DTI). Prior to this he was deputy director-general of the Industrial Develop
ment Division (IDD) of the DTI. He sits on the board of the South African Industrial
Development Corporation (IDC) and chairs the steering committee of the Industrial
Development Think Tank (ITT) at the University of Johannesburg.
Page 16 of 16
Introduction to Industrial Policy and Development

Arkebe Oqubay, Christopher Cramer, Ha-Joon Chang, and Richard Kozul-
Wright
Print Publication Date: Oct 2020 Subject: Economics and Finance, Industrial Organization
Online Publication Date: Nov 2020 DOI: 10.1093/oxfordhb/9780198862420.013.1
Abstract and Keywords
Industrial policy has a long history both in practice and in theory. It became a leading fo
cus of development research and policy analysis after the end of the Second World War,
although its meaning, scope, and instruments have varied significantly, and it has been
the subject of sustained criticism and debate, especially during the 1980s and 1990s. Not
ing that industrial policy in fact never went away, and at a moment when it has returned
to centre stage, this volume offers a comprehensive reference work that presents differ
ent schools of thought regarding industrial policy and reflects the evolution in contempo
rary thinking, alongside empirical evidence from advanced, emerging, and developing
economies. This volume also makes the connection between industrial policy and other
policies. The volume reviews the theoretical perspectives and methodological aspects of
the study of industrial policy, and uses case studies of policies and practices to offer new
insights for policymakers, practitioners, and policy researchers. Contributors identify and
assess evolving challenges to industrial policy and the shifting terrain of the industrial.
They emphasize a political economy approach rather than reducing industrial policy to a
technical exercise. The Handbook is forward looking, while also presenting a comprehen
sive review of the evolving context and trajectories of industrial policy.
Keywords: industrial policy, structural transformation, catch-up, industrialization, innovation and technological ca
pability, development economics
1.1 An Overview of Industrial Policy and Devel

opment
1.1.1 History and Debate in Industrial Policy
INDUSTRIAL policy has a long history both in practice and in theory. Early practices of
what would come to be regarded as industrial policy stretch back at least as far as Me
dieval Europe and can be found in Italian city-states or late medieval England. There is al
Page 1 of 16
so a rich history of thinking and analysis about such policies, including sixteenth-century
Italian texts, long before the more frequently referenced classic late eighteenth- and ear
ly nineteenth-century texts of Alexander Hamilton (who argued that Adam Smith’s theo
ries were ‘geometrically true’ but ‘practically false’) and Friedrich List.1
Industrial policy discussions resurfaced during the inter-war period, particularly against
the backdrop of the decline of traditional industries in some advanced economies under
threat from a mixture of emerging competitors and new technologies (Chick, 2018) and
often connected to growing economic problems at the regional level. With the breakdown
of the gold standard and the protectionist turn in the 1930s, industrial policy was closely
linked to a wider set of measures aimed at expanding local markets, particularly in poor
er countries, where discussions around import-substitution (p. 4) industrialization began
to take root. Industrial policy became a staple of development research and policy analy
sis from the end of the Second World War.2
Nonetheless, the most extensive theoretical and empirical perspectives on industrial poli
cy date from the late twentieth century, often linked to a close examination of the policies
adopted by the successful industrializing economies of East Asia (Johnson, 1982; Amsden,
1989, 2001; Wade, 1990; Chang, 2002, among others). These approaches highlighted the
significance of ‘a country’s structure of production and trade, because different goods
have different production and demand characteristics’ (Thirlwall, 2019: 556), and the role
of state institutions in shifting those structures to support and sustain economic develop
ment (Haggard, 2018).
But industrial policy has also been subject to sustained criticism. Especially from the late
1970s onwards, in mainstream economic analysis, in the pronouncements of international
development agencies, and in explicit policy statements by some (especially Anglo-Saxon)
governments, industrial policy fell out of favour.3 Neoclassical growth economists tended
to favour one-sector growth models and to argue that there were no special properties for
any sector. They also pointed to the obvious potential for industrial policy to create ‘rent-
seeking’ opportunities.4 The debates between proponents and opponents of industrial pol
icy as a strategy of economic development, sustained growth, and competitiveness be
came sharply ideological and were not always supported by comprehensive, evidence-
based analyses.
Despite these criticisms, industrial policy never went away. Even in the most committed
‘neo-liberal’ policy environments—the United Kingdom and the United States, for exam
ple—industrial policies were implemented (often especially in the military-industrial poli
cy arena), though they were not always advertised loudly as such (see Chapter 20). In the
United States, for example, the Department of Energy ‘has a program to provide low-in
terest loans to companies to encourage risky corporate innovation in alternative energy
and energy efficiency…as a whole, since its inception in 2009, the program has turned a
profit…Its loans to early-stage solar energy companies launched the industry’ (Lewis,
2018: 33–4). Still, lingering hostility to industrial policy meant that these policies were of
Page 2 of 16
ten presented as ‘energy policy’ or ‘defence policy’, though they promoted some frontier
technologies through the public funding of R&D and procurement preferences.
This volume comes at a moment when industrial policy has returned to centre stage. Ad
vanced-economy governments have been designing, implementing, and announcing a new
generation of industrial policies with more confidence as part of their efforts to benefit
from the Fourth Industrial Revolution—even to the point where there are concerns about
‘industrial policy wars’ alongside trade wars.5 International (p. 5) organizations have be
come more open to discussing industrial policy in lower- and middle-income countries, in
Africa for example, in part thanks to China’s remarkable transformation over the last
forty years (UNCTAD, 2016) but also in the context of a climate emergency that requires
a rapid shift from high- to low-carbon growth paths in both developed and developing
countries (UNIDO, 2017). Two IMF economists even attracted considerable attention for
hailing the ‘return of the policy that shall not be named’ (Cherif and Hasanov, 2019).
UNCTAD’s (2018) World Investment Report identified eighty-four countries (accounting
for 90 per cent of global GDP) that had adopted formal industrial policies in the five years
to 2018. Further, in recent years industrial policy has come to rely on an ‘expanded range
of support measures and instruments that aim to improve infrastructure, education and
training, enterprise development, the building of clusters and linkages, entrepreneurship,
innovation, access to finance, and social policies’ (UNCTAD, 2018: 126, citing Salazar-
Xirinachs et al., 2014).
At such a moment, this volume offers a comprehensive reference work that presents dif
ferent schools of thought and reflects evolution in contemporary thinking on industrial
policy, alongside theoretical and empirical evidence from both advanced and developing
economies. It also makes the connection between industrial policy and other policies.
When economists (and others) across a broad spectrum seem to agree on the relevance,
desirability, and practicability of industrial policy, it is especially important to assess sug
gestions that there is a ‘consensus’ around what this means and involves. For some, in
dustrial policy should still be seen as subordinate to other goals, such as competitiveness
or structural reform, while others argue that this ‘mainstreaming’ of industrial policy
amounts to a dilution of the idea. Hence the relevance of this book, which provides the
historical background and conceptual underpinnings, and presents a range of perspec
tives on industrial policy, together with regional and individual country case studies.
1.1.2 Evolving Challenges to Industrial Policy
This volume captures a range of perspectives on industrial policy at a very specific mo
ment in the history of such policies. Another reason why this book is particularly relevant
is that, just as industrial policy has become more popular again among both academic
economists and policymakers, there are also new challenges to the viability of industrial
policy, especially in developing countries. The upshot, for some observers, is that low-in
come countries, for example those in Africa, should not even try to industrialize any more,
given the advent of a post-industrial age (IMF, 2018), or should simply limit their ambi
Page 3 of 16
tion to providing links in global value chains, given how impenetrable the global industri
al hierarchy has become (Baldwin, 2016).
The most significant of these threats is the spectre of ‘premature de-industrialization’:

many countries have experienced a decline in the share of manufacturing in GDP—de-in
dustrialization—at far lower levels of per capita GDP than has historically been (p. 6) the
case. This trend, and some of the factors that may lie behind it, suggests to some econo
mists that manufacturing (and industrial policy) cannot be the ‘engine of growth’ that it
historically has been (World Bank, 2017). But this is contested, and debates around the
ineluctability of premature de-industrialization and about the continued role for manufac
turing and industrial policy are taken up across different chapters in this book.6
Premature de-industrialization is just one of a number of features of the contemporary

global political economy that affect the scope for industrial policy. Related arguments
about automation and the Fourth Industrial Revolution suggest to some that it will be
even more difficult, particularly in lower- and middle-income countries, successfully to
pursue industrial policies geared towards competitiveness, productivity growth, and em
ployment (Cramer and Tregenna, 2020). Others argue that there remains plenty of scope
for industrial policies to harness the dynamics of new waves of technical change and that
these may even in some respects make it easier to ‘catch up’.7 Meanwhile, there are argu
ments that recent developments in global political economy and governance, for example
around the increased market power of large international firms, restrict policy space still
further for developing countries in particular. And there are counter-arguments pointing
to the opportunities for policy space that are created by a more multi-polar, fragmented
global political order (Grabel, 2015). If the ‘rules of the game’ have become increasingly
unstable, that may make the global economy both more hostile and more ‘permissive’ for
developing countries.
Finally, much production nowadays is organized through global value chains commanded
by ‘systems integrators’ controlling a cascade of ‘tiers’ of suppliers, with very little scope
for lower-income countries to occupy anything higher than the lower levels. There are
complex debates about what scope there is for these lower-income countries to ‘move up’
within GVCs, about the ways in which GVCs have reshaped much industrial policy around
foreign direct investment policy,8 and indeed about the implications of evidence of some
‘reshoring’ of production (bringing production and innovation physically closer together
again) by firms based in advanced economies.9
1.1.3 The Shifting Terrain of Industrial Policy
The spatial dimensions of firm decisions within a globalized economy are an important in
fluence on the evolution of industrial policies. Another influence is the blurring of bound
aries between ‘sectors’ traditionally treated as discrete categories of economic activity.
One feature of modern capitalism is the way in which firms, especially those based in ad
vanced economies, concentrate and control knowledge expertise in ways (p. 7) that add to
the advantages of first movers and economies of scale and scope, etc. This represents one
Page 4 of 16
facet of ‘servicification’, the increasing share of the value of final output of goods ac
counted for by service-sector activities—branding, design, R&D, logistics, etc., whether
carried out in-house or contracted out. But economic data—for example, US production
and employment statistics—have not adjusted to reflect accurately the relative weight of
different activities. US statistics, for example, ‘do not include [in manufacturing statis
tics] pre-production services to manufacturing such as research and development or de
sign or post-production services such as repair and maintenance or sales. Yet manufactur
ing firms invest heavily in these services’ (Whitefoot and Valdivia, 2015: 1). The basis of
the North American Industry Classification System (NAICS) is the assumption that extrac
tive industries, manufacturing, and service industries ‘rely on essentially different pro
duction processes and establishments in each of these three categories have similar
processes’ (Whitefoot and Valdivia, 2015: 3). As a result, much that is fundamentally man
ufacturing-dependent economic activity is classified as non-manufacturing in service-sec
tor data. Another way in which activities traditionally assigned to discrete ‘sectors’ have
become increasingly imbricated is the ‘industrialization of freshness’ (Cramer, Di John,
and Sender, 2018). Much production of ‘fresh’ and apparently unprocessed high-value
agricultural commodities (e.g. avocados, blueberries, poinsettia, or pelargonium
seedlings) is very different from traditionally conceived ‘primary commodity production’
and in its ‘roundabout’ production processes has many of the attributes of industrial ac
tivity.10
1.1.4 Political Economy and Industrial Policy
The recent return of industrial policy to international policy discussions is not the result
of new analytical insights into the process of structural transformation. Arguably, it has
more to do with the weak outcomes of policies pursued by many developing countries un
der the guidance of the Washington Consensus,11 the shock to market-friendly policies in
the global financial crisis of 2008 and its aftermath, the challenges to the welfare state
and ‘normal’ politics in many advanced, de-industrialized economies, growing regional
disparities, new sources and forms of global competition and market demand (in China
above all),12 and the questions posed by technological changes and climate change.
Meanwhile, governments in developing countries are once again asking (p. 8) how they
can achieve their development goals by creating new sources of growth and dynamism,
rather than simply trying to do the best with what they currently have.13
In discussing these and other issues, the present volume takes a political economy ap
proach. For industrial policy cannot be reduced to a technical or ‘purely’ economic argu
ment or set of instruments. Both more and less orthodox economists agree on this. But
there are very different ways of taking account of the political economy of industrial poli
cy. Public-choice theory may be one way of thinking about the political economy but it is
not the same as more historically minded, less methodologically individualist frameworks,
such as the political settlements literature or older (and to some extent more explicitly
class-based) analyses like Hirschman’s discussion of the political economy of import-sub
stitution industrialization (ISI) in Latin America (Hirschman, 1968). Arguably, the richest
vein of the political economy of industrial policy is found in domestic class-based analysis
Page 5 of 16
and in analyses of particular countries’ positions in international alliances and hierar

chies. From that perspective, domestic and international security has often been an im
portant theme in accounting for governments resorting to and pushing industrial policy.
For example, Doner, Ritchie, and Slater (2005) analysed industrial policy and the ‘devel
opmental state’ in North East Asia as responses to internal and external ‘threats’. In this
volume, another kind of example is given in Mathews’ argument about the role of energy
security in provoking innovative industrial policy in China which has created new, lower-
cost technologies that over time can in turn help lower-income economies avoid reinvent
ing a sustainable wheel.14
1.1.5 The Cabinet of Policy Errors
One further feature distinguishing this volume is its commitment to highlighting learning
from policy mistakes. While we are wary of simplistic labels of ‘failure’ and ‘success’,
whose evaluative criteria are often not laid out clearly or are subjective, arbitrary, or un
duly narrow, the issue of policy failure is a key theme in debates on industrial policy.
Much of the hostile literature on industrial policy simply points to failures and uses these
to argue against its adoption. Much of the literature more favourable to industrial policy
picks out and explores the conditions behind obvious ‘successes’. Thus, the literature can
become bifurcated and static, reduced to an unhelpful either/or of market failure or state
failure. We argue for a more dynamic focus on ‘failure’. This stance is inspired not only by
Hirschman’s (1967) subtle investigation of the dynamics of the ‘Hiding Hand’ in large
projects, but also by the experience of Enzo Ferrari. Ferrari often held board meetings in
a room whose walls were lined with cabinets displaying the parts of racing cars that had
malfunctioned (p. 9) during races, to focus the board members’ minds on the obsessive
need to learn from what had gone wrong and to improve. We advocate a wider institution
al adoption of the policy official’s equivalent of the ‘cabinet of errors’.
1.2 Aims and Approaches of the Handbook

1.2.1 Aims and Approaches
No country has made the arduous journey from widespread rural poverty to post-industri
al prosperity without employing targeted and selective government policies to shift the
production structure towards activities and sectors with higher productivity, better-paid
jobs, and greater technological potential.15 Alfred Marshall’s ‘principle of continuity’, that
is, development through incremental changes (Marshall, 1920) has never been sufficient
to bring about fundamental structural changes and rising productivity and has always
been supplemented by the visible hand of policy intervention. Such policies are conven
tionally called industrial policies, although they are also termed ‘production/productive
transformation policies’. Much of the discussion on industrial policy in recent years, how
ever, has revolved around sterile debates on government failures or whether govern
Page 6 of 16
ments can pick winners. This volume focuses on theoretical foundations, on institutional
conditions, and on emerging challenges.
The Oxford Handbook of Industrial Policy introduces specialist research by a wide range
of scholars and seeks to make a significant contribution to our understanding of industri
al policies and their role in economic growth, development, and structural change, both
historically and in today’s changing policy landscape. The volume reviews the conceptual
underpinnings, theoretical perspectives, and methodologies of the study of industrial poli
cies. It uses case studies of policies and practices to offer new insights for policymakers,
practitioners, and policy researchers. The Handbook is forward looking while also draw
ing on long historical perspectives, and aims to serve as a single, comprehensive, and au
thentic reference on industrial policies. The book also highlights major themes and policy
perspectives rather than speculating on recommendations that may not stand the test of
time.
The evolving debates on industrial policy and the perspectives of various schools of
thought are presented, from the early Industrial Revolution to the adaptive industrial pol
icy practices of the present day.16 Theoretical perspectives are integrated with empirical
evidence from both advanced economies and developing and emerging economies, bridg
ing the gap between the theory and practice of industrial policy. (p. 10) The volume’s criti
cal examination of the progress and direction of debates lays the foundation for future re
search. It attempts to provide new perspectives for scholars and graduate students, as
well as a collection of authoritative and original research.
1.2.2 Process
The first step in ensuring the high standard and timely publication of this volume was the
careful selection of topics and contributing authors. At an inception workshop in Addis
Ababa, editors and contributors met to discuss the themes and chapters chosen, to re
view abstracts, and to discuss new ideas and suggestions. Contributors were encouraged
to focus on mistakes and failures as well as successes, to consider connections and con
texts, to be cognizant of unevenness in policy design and outcomes, and to highlight polit
ical economy issues.
External and internal reviewers have carefully reviewed each chapter, providing a high
level of peer review engagement (especially for an edited volume). A chapter review
workshop reviewed all first drafts to improve chapters and ensure complementarity. A
book project coordination team provided the necessary support service and a pre-publish
er copy-editing service improved readability and ensured uniformity of style.
1.3 Organization and Structure

The thirty chapters of the Handbook are organized in five sections: introduction; theoreti
cal perspectives; context and connections; experiences in advanced economies; emerging
economies and developing countries.
Page 7 of 16
1.3.1 Part I: Introduction
The background, purpose, approaches, and organization of the Handbook are introduced
in Chapter 1, together with an outline of the evolving landscape of industrial policy and
emerging research questions. The volume’s wide conceptualization of industrial policy in
cludes the transformation of production, the development of technological and innovation
capabilities, and the constant struggle to occupy more advantageous positions in interna
tional markets. In addition to industrial policy within national and sub-national struc
tures, regional application of industrial policy can be seen, for example, in the European
Union. Oqubay’s Chapter 2, ‘The Theory and Practice of Industrial Policy’, presents indus
trial policy as a driver of structural change and a conduit of technological catch-up, un
derlining the strategic (p. 11) role of exports and of sectors with higher dynamic efficien
cy.17 The dynamic and adaptive aspects of industrial policies are examined, together with
their origins and the unevenness of policy design and outcomes. The chapter provides the
groundwork for subsequent chapters that present theoretical perspectives, contexts and
connections, and experiences from advanced, emerging, and developing countries.
1.3.2 Part II: Theoretical Perspectives
The five chapters in this section present a comprehensive conceptual framework by re
viewing the industrial policy perspectives of various schools and traditions. Ocampo
(Chapter 3) examines the macroeconomic dimension of industrial policy, arguing that
structural change and active industrial policy or productive transformation policies must
be at the centre of economic development strategy. The chapter underlines the centrality
of state intervention in promoting the dynamic efficiency of economic structures and their
capacity to generate new waves of structural change. Cimoli, Dosi, and Yu (Chapter 4)
discuss the role of industrial policy in the evolutionary view of innovation and learning as
drivers of economic development. Technological paradigms and trajectories are consid
ered, as well as the link between catch-up processes and the dynamics of capability accu
mulation within and across firms, and their embeddedness in national systems of innova
tion.18
Weiss (Chapter 5) reviews neoclassical perspectives on industrial policy. The chapter dis
cusses market imperfections and ‘distortions’, the idea of a ‘policy hierarchy’ that might
lead to a case for industrial policy, and the neoclassical assumptions underlying recent
work on ‘discovery process’, ‘export space’, and ‘facilitative’ interventions to exploit com
parative advantage. Roberts (Chapter 6) presents a firm-based perspective linking large
firm strategies to industrial policies, and reviews relevant theories of the firm and capa
bility development. Ashman, Newman, and Tregenna (Chapter 7) look at various ‘radical’
economic traditions relating to industrial policy and industrialization and consider how
these views differ from structuralist or evolutionary economics or other heterodox ap
proaches. Their discussion gives greater attention, for example, to arguments about the
‘inherent contradiction of labour and capital’ and the ‘contradictory’ role of the state.
Page 8 of 16
1.3.3 Part III: Context and Connections
This section comprises eleven critical chapters on the context of industrial policy and its
connections with other policies. Djafar and Milberg (Chapter 8) discuss global value
chains and the regionalization of industrial policy, based on the case of ASEAN.
(p. 12)
The changing nature of global production networks and their concentration is examined,
together with regional coordination of industrial policy, which is strengthening produc
tion integration and promoting competitiveness. Kozul-Wright and Fortunato (Chapter 9)
examine industrial policy in an interdependent world and review debates around trade,
participation in global value chains, and industrial policy. The chapter discusses the diver
sification of the export basket, the sophistication of export products, and the upgrading of
productive capacities. Mathews (Chapter 10) examines the greening of industrial policy,
the shift to renewables, and the solutions offered by the circular economy.19 Case studies
from advanced, emerging, and developing countries are presented, encompassing energy,
materials and finance, and other green options.
Esser and Mittelman (Chapter 11) address the blind spot in the evolving and competing
narratives of globalization and industrial policy, showing the importance of understanding
these representations as they are objectified as policy. Mazzucato and Kattel (Chapter
12), examining the ‘mission-oriented’ approach to industrial policy, which focuses on pro
viding the direction for growth, argue for the importance of a new analytical framework
to understand how policies can shape market forces. Di John (Chapter 13) reviews the
evolution and political economy of development banking as a catalyst of industrial policy
in the context of late development and structural change (with particular attention to the
case study of Brazil).
Andreoni (Chapter 14) reviews the role of technical change as a driver of the shifting ter
rain of industrial sectors and the industrial ecosystem. Pollin (Chapter 15) identifies the
prime contributors to climate change and examines viable routes to achieving agreed
global emissions goals. The chapter also looks at the role of clean energy industrial poli
cies and global investment, and how industrial policy will promote technical innovation
and the expansion of investment. Seguino (Chapter 16) explores the intersection of gen
der relations and industrialization and industrial policy, and argues that gender-inclusive
industrial policies can ensure equitable and sustainable development. Storm (Chapter 17)
shows the importance of macroeconomic policy management for the effectiveness of in
dustrial policy and argues that demand-side policies are crucial. He explores the ways in
which food policy and wages and labour protection legislation can underpin industrial
policy and shows how overvalued exchange rates and restrictive monetary policy have of
ten undermined industrial policy efforts. Bailey and De Propris (Chapter 18) review tech
nological change in global value chains, the origins of internationalization strategies and
global production networks, and the importance of appreciating the spatial dimensions of
industrial policy. (p. 13)
Page 9 of 16
1.3.4 Part IV: Experiences in Advanced Economies
The five chapters in Part IV focus on empirical evidence of industrial policies in selected
advanced economies. Reinert (Chapter 19) reviews historical perspectives on the origins
of industrial policy and industrialization, the role of the state and key arguments ranging
from before the Industrial Revolution up to the early twentieth century. Best (Chapter 20)
reviews the success of US industrial policy during the Second World War, and the
country’s post-war de-industrialization, arguing that industrial policy has been integral to
core production processes and innovation capability. Bianchi and Labory (Chapter 21) ex
amine the evolution of post-Second World War European industrial policy through a com
parative review of three phases (strong and selective intervention, market-led approach
es, and more recently, more interventionist), illustrating how the EU’s multilevel gover
nance adapted to new technological disruptions. Landesmann and Stöllinger (Chapter 22)
review EU industrial policy, with a focus on contemporary Europe, and examine the key
challenges of keeping pace with the frontiers of technology, responding to the fast catch
ing-up by emerging economies, convergence and cohesion within the EU, and the climate
challenge. Lee (Chapter 23) reviews East Asian experiences with a focus on South Korea’s
economic catch-up from the 1960s to the 2010s, showing how industrial policy was used
to build innovation capability and upgrade the private sector during the four phases of
transformation.
1.3.5 Part V: Experiences in Emerging and Developing Countries
Part V focuses on industrial policy experiences and outcomes in emerging and developing
countries, including South East Asia, post-communist transition economies, Latin Ameri
ca, and Africa. First, Rasiah (Chapter 24) reviews industrial policy and industrialization in
South East Asia, with a comparative analysis that shows the persistence of premature de-
industrialization and the weakness of industrial policies in contrast to those of Japan,
South Korea, and Taiwan. Li and Chen (Chapter 25) examine the development of national
champions as a key element of industrial policy, showing how China transformed and con
solidated its big state-owned enterprises into world-class firms.
Santiago (Chapter 26) reviews industrial policy in the BRICS (Brazil, Russia, India, China,
South Africa), exposing the asymmetry of these countries’ economic performances and di
vergent industrial policies, despite the cooperation between them. Popov (Chapter 27)
considers the lessons from the evolution of industrialization and industrial policy out
comes among (post-communist) transition economies of Europe and Asia. Ocampo and
Porcile (Chapter 28) present a comparative review of macroeconomic and industrial poli
cies of Latin American countries and the outcomes in terms of catch-up and eco
(p. 14)
nomic transformation. Whitfield and Zalk’s (Chapter 29) comparative review of industrial
policy in four African countries (Ethiopia, Nigeria, Rwanda, and South Africa) highlights
their mixed outcomes and the limited economic development achieved. Chitonge and
Lawrence (Chapter 30) examine the different phases of industrial policy in post-indepen
Page 10 of 16
dence Africa, reviewing the history through the lens of power relations and the state as
the focus of political economy.
References
Amsden, Alice H. (1989) Asia’s Next Giant: South Korea and Late Industrialization. Ox
ford University Press on Demand.
Amsden, Alice (2001) The Rise of ‘the Rest’: Challenges to the West from Late-industrial
izing Economies. New York: Oxford University Press.
Baldwin, Richard (2016) The Great Convergence: Information Technology and the New
Globalization. Cambridge, MA: Harvard University Press.
Best, Michael H. (2018) How Growth Really Happens: The Making of Economic Miracles
through Production, Governance, and Skills. Princeton, NJ: Princeton University Press.
Chang, Ha-Joon (2002) Kicking away the Ladder: Development Strategy in Historical Per
spective. London: Anthem Press.
Cherif, Reda and Fuad Hasanov (2019) ‘The Return of the Policy That Shall Not Be
Named: Principles of Industrial Policy’. International Monetary Fund. Available at https://
www.imf.org/en/Publications/WP/Issues/2019/03/26/The-Return-of-the-Policy-
That-Shall-Not-Be-Named-Principles-of-Industrial-Policy-46710.
Chick, Victoria (2018) ‘Industrial Policy, Then and Now’. Real-world Economics Review
84: 178–88. Available at http://www.paecon.net/PAEReview/issue84/Chick84.pdf.
Cimoli, Mario, Giovanni Dosi, and Joseph E. Stiglitz (eds) (2009) Industrial Policy and De
velopment: The Political Economy of Capabilities Accumulation. The Initiative for Policy
Dialogue Series. Oxford: Oxford University Press.
Cramer, Christopher and Fiona Tregenna (2020) ‘Heterodox Approaches to Industrial Pol
icy and Implications for Industrial Parks’, in Justin Yifu Lin and Arkebe Oqubay (eds) The
Oxford Handbook of Industrial Hubs and Economic Development. Oxford: Oxford Univer
sity Press.
Cramer, Christopher, Jonathan Di John, and John Sender (2018) ‘Poinsettia Assembly and
Selling Emotion: High Value Agricultural Exports in Ethiopia’. AFD Research Papers
Series No. 2018–79, September. Paris: Agence Française du Développement.
Cramer, Christopher, John Sender, and Arkebe Oqubay (2020) African Economic Develop
ment: Evidence, Theory, Policy. Oxford: Oxford University Press.
Doner, Richard, Bryan Ritchie, and Dan Slater (2005) ‘Systemic Vulnerability and the Ori
gins of Developmental States: Northeast and Southeast Asia in Comparative Perspective’,
International Organization 59(2): 327–61.
Page 11 of 16
Gereffi, Gary (2018) Global Value Chains and Development: Redefining the Contours of
21st-century Capitalism. Cambridge: Cambridge University Press.
Grabel, Ilene (2015) ‘The Rebranding of Capital Controls in an Era of Productive Incoher
ence’, Review of International Political Economy 22(1): 7–43.
Haggard, Stephan (2018) Developmental States. Cambridge: Cambridge University Press.

(p. 15)
Hirschman, Albert O. (1958) The Strategy of Economic Development. New Haven, CT:
Yale University Press.
Hirschman, Albert O. (1967) Development Projects Observed. Washington, DC: Brookings

Institution.
Hirschman, Albert O. (1968) ‘The Political Economy of Import-substituting Industrializa

tion in Latin America’, The Quarterly Journal of Economics 82(1): 1–32.
IMF (2018) World Economic Outlook: Cyclical Upswing, Structural Change. Washington,
DC: IMF.
Johnson, Chalmers (1982) MITI and the Japanese Miracle: The Growth of Industrial Poli
cy, 1925–1975. Stanford, CA: Stanford University Press.
Kaldor, Nicolas (1966) Causes of the Slow Rate of Economic Growth of the United King
dom: An Inaugural Lecture. Cambridge: Cambridge University Press.
Lee, Keun (2019) The Art of Economic Catch-up: Barriers, Detours and Leapfrogging in
Innovation Systems. Cambridge: Cambridge University Press.
Lewis, Michael (2018) The Fifth Risk. New York: W. W. Norton & Company.
Lin, Justin Yifu and Ha-Joon Chang (2009) ‘Should Industrial Policy in Developing Coun
tries Conform to Comparative Advantage or Defy It? A Debate between Justin Lin and Ha-
Joon Chang’, Development Policy Review, 27(5): 483–502.
Marshall, Alfred (1920) Principles of Economics. London: Palgrave Macmillan.
Mathews, John (2014) Greening of Capitalism: How Asia Is Driving the Next Great Trans
formation. Stanford, CA: Stanford University Press.
Mazzucato, Mariana (2013) The Entrepreneurial State: Debunking Public vs. Private Sec
tor Myths, Vol. 1. London: Anthem Press.
Mittelman, James H. (2017) Implausible Dream: The World-class University and Repur
posing Higher Education. Princeton, NJ: Princeton University Press.
Myrdal, Gunnar (1968). Asian Drama: An Inquiry into the Poverty of Nations. New York:
Pantheon.
Page 12 of 16
Naudé, Wim (2019) ‘African Countries Can’t Industrialise? Yes, They Can’. The Conversa
tion, 4 November. Available at https://theconversation.com/african-countries-cant-
industrialise-yes-they-can-125516.
Nayyar, Deepak (2019) Asian Transformations: An Inquiry into the Development of Na
tions. Oxford: Oxford University Press.
Nolan, Peter (2001) China and the Global Economy: National Champions, Industrial Policy
and the Big Business Revolution. Basingstoke: Palgrave.
Oqubay, Arkebe (2015) Made in Africa: Industrial Policy in Ethiopia. New York: Oxford
University Press.
Oqubay, Arkebe and Kenichi Ohno (2019) How Nations Learn: Technological Learning, In
dustrial Policy, and Catch-up. New York: Oxford University Press.
Prebisch, Raúl (1950) The Economic Development of Latin America and its Principal
Problems. New York: UN Department of Economic Affairs.
Salazar-Xirinachs, José M., Irmgard Nübler, and Richard Kozul-Wright (eds) (2014) Trans
forming Economies: Making Industrial Policy Work for Growth, Jobs and Development.
ILO/UNCTAD.
Schumpeter, Joseph (1934) Theory of Economic Development. Cambridge, MA: Harvard

University Press.
Thirlwall, Anthony (2019) ‘Thoughts on Balance-of-payments-constrained Growth after 40

Years’, Review of Keynesian Economics 7(4): 554–67.
UNCTAD (2016) Trade and Development Report 2016: Structural Transformation

(p. 16)
for Sustained and Inclusive Growth. Geneva: UNCTAD.
UNCTAD (2017) ‘Beyond Austerity: Towards a Global New Deal’, in Trade and Develop
ment Report 2017. Geneva: UNCTAD.
UNCTAD (2018) World Investment Report 2018. New York and Geneva: UNCTAD.
UNIDO (2017) Structural Change for Inclusive and Sustainable Development. Vienna:
UNIDO.
Wade, Robert (1990) Governing the Market: Economic Theory and the Role of Govern
ment in East Asian Industrialization. Princeton, NJ: Princeton University Press.
Whitefoot, Kate and Walter Valdivia (2015) Innovation and Manufacturing Labor: A Value-
chain Perspective. Washington, DC: Centre for Technology Innovation, Brookings.
World Bank (2017) Trouble in the Making? The Future of Manufacturing-led

Development. Washington, DC: World Bank.
Page 13 of 16
Notes:
(1) See Chapter 19.
(2) See Prebisch (1950), Myrdal (1968).
(3) See Chapter 20.
(4) See Chapter 5.
(5) ‘Industrial Policy War—Capitalism with Chinese Characteristics’, Financial Times, 21

September 2019.
(6) For instance, see For instance, see Chapters 2, 3, 14, and 23.
(7) For example, Naudé (2019), UNCTAD (2017).
(8) See also Mittelman (2017), Lee (2019), UNCTAD (2018).
(9) See Gereffi (2018). See also Chapters 8 and 9.
(10) On the enduring difficulties with the classification of economic activities, see also
Marshall (1920) and Schumpeter (1934). See also Chapter 14, which discusses this issue
of the shifting terrain of industrial policy.
(11) Oqubay (2015), Cramer, Sender, and Oqubay (2020).
(12) Nolan (2001), Nayyar (2019).
(13) See Cimoli, Dosi, and Stiglitz (2009), Best (2018), and Lee (2019).
(14) See Chapter 10.
(15) See Chapters 2 and 3.
(16) See Chapters 5, 7, and 11. See also Lin and Chang (2009).
(17) See Hirschman (1958), Kaldor (1966), Amsden (1989), and Mazzucato (2013).
(18) See Cimoli, Dosi, and Stigltiz (2009), Best (2018), and Oqubay and Ohno (2019).
(19) See Mathews (2014).
Arkebe Oqubay
Arkebe Oqubay (PhD) is a senior minister and special adviser to the prime minister of
Ethiopia and has been at the centre of policymaking for over twenty-five years. He is
the former mayor of Addis Ababa, winner of the Best African Mayor of 2005 and fi
nalist in the World Mayor Award 2005, for his work transforming the city. He is a re
cipient of the Order of the Rising Sun, Gold and Silver Star, presented by the Emper
or of Japan. He currently serves as board chair of several leading public organiza
Page 14 of 16
tions and international advisory boards. He is an ODI Distinguished Fellow and a re
search associate at the Centre of African Studies in the University of London, and
holds a PhD in development studies from SOAS, University of London. His recent
works include the path-breaking Made in Africa: Industrial Policy in Ethiopia (OUP,
2015); The Oxford Handbook of the Ethiopian Economy (OUP, 2019); How Nations
Learn: Technological Learning, Industrial Policy, and Catch-up (OUP, 2019); China‒
Africa and an Economic Transformation (OUP, 2019); African Economic Develop
ment: Evidence, Theory, and Policy (OUP, 2020); and The Oxford Handbook of Indus
trial Hubs and Economic Development (OUP, 2020). He was recognized as one of the
100 Most Influential Africans of 2016, and a ‘leading thinker on Africa’s strategic de
velopment’ by the New African, for his work, both theoretical and practical, on indus
trial policies.
Christopher Cramer
Christopher Cramer is professor of the political economy of development at SOAS,

University of London. He is a vice-chair of the Royal Africa Society and chair of the
Scientific Committee of the African Programme on Rethinking Development Econom
ics (APORDE). His publications include Civil War Is Not A Stupid Thing: Accounting
for Violence in Developing Countries (Hurst Publishers, 2006), African Economic De
velopment: Evidence, Theory, and Policy (OUP, 2020) and The Oxford Handbook of
the Ethiopian Economy (OUP, 2019, co-edited with Cheru and Oqubay). He led the re
search project Fairtrade, Employment, and Poverty Reduction in Ethiopia and Ugan
da.
Ha-Joon Chang
Ha-Joon Chang teaches economics at the University of Cambridge. His main books
include Kicking Away the Ladder (2002), Bad Samaritans (2007), 23 Things They
Don’t Tell You about Capitalism (2011) and Economics: The User’s Guide (2014). His
writing has been translated into forty-one languages in forty-four countries. World
wide, his books have sold over 2 million copies. He is the winner of the 2003 Gunnar
Myrdal Prize and the 2005 Wassily Leontief Prize.
Richard Kozul-Wright
Richard Kozul-Wright is director of the Globalisation and Development Strategies Di

vision in UNCTAD. He has worked at the United Nations in both New York and Gene
va. He holds a PhD in economics from the University of Cambridge. He has published
widely on economic issues including, inter alia, in The Economic Journal, Cambridge
Journal of Economics, The Journal of Development Studies, and Oxford Review of
Economic Policy. He is the author of many books, including The Rise and Fall of Glob
al Microcredit: Development, Debt and Disillusion (2018, with S. Blankenburg and M.
Bateman), Securing Peace: State-Building and Economic Development in Post-Con
flict Countries (2011, with P. Fortunato), Climate Protection and Development (2012,
with Frank Ackerman) and The Resistible Rise of Market Fundamentalism (2008,
with Paul Rayment). His edited volumes include Transnational Corporations and the
Page 15 of 16
Global Economy, Economic Insecurity and Development, Securing Peace, Climate

Protection and Development, and Industrial Policy. He is a frequent contributor to
newspapers worldwide on economic issues, including the Financial Times, The
Guardian, and Project Syndicate.
Page 16 of 16
The Theory and Practice of Industrial Policy

Arkebe Oqubay
This chapter reviews the theory and practice of industrial policy, focusing on three key
themes. First, it presents industrial policy as a vehicle by which manufacturing (and other
dynamic and high-productivity services and agricultural activities that share technologi
cal and linkage spillovers) drives structural change. It also underlines the strategic role of
exports, especially as a source of international learning and relaxing balance-of-payment
constraints. Second, it examines industrial policy as a conduit of technological catch-up,
and emphasizes the centrality of technological learning and the development of techno
logical and innovation capabilities for economic catch-up and sustained economic growth.
Finally, it reviews the origins, the dynamic and adaptive nature of industrial policies, and
the unevenness and variations in policy design and outcomes. The chapter draws primari
ly from structural development economics and evolutionary economics, and includes a
wider scan of the literature. It provides the groundwork for subsequent chapters that ex
amine theoretical perspectives and connections, and experiences from advanced, emerg
ing, and developing countries.
Keywords: industrial policy, structural transformation, catch-up, technological learning, manufacturing, exports,
state and politics, literature review
2.1 Introduction to the Theory and Practice of

Industrial Policy
DISCOURSE on industrial policy and scholarly debates echoes global economic shifts,
changes in the international power structure, and the ideological fervour of the period.
Arguably, many governments have continued to use some form of industrial policy, al
though it appeared unfashionable in the 1970s‒1990s with the rise of new liberal waves.
Nonetheless, industrial policy has recently attracted growing attention from policymakers
and researchers, and is now part of mainstream economic debate. A recent working pa
per by IMF researchers (Cherif and Hasanov, 2019) was entitled ‘The Return of the Policy
That Shall Not Be Named: Principles of Industrial Policy’. Nonetheless, there is little
Page 1 of 48
agreement on the underlying conceptual perspectives and basic essence of industrial poli
cies and empirical interpretation (Andreoni and Chang, 2019).1
Pioneering theoretical work on industrial policy, particularly the theory of infant industry
protection, is mainly associated with the classical political economy of the early capitalist
economy in eighteenth-century England and the economic catch-up of continental Eu
rope, the United States, and Japan in the nineteenth century (p. 18) (Hamilton, [1791]
1934; List, 1856).2 Industrial policy as a viable path to capitalist development has been
practised by both forerunners and latecomers (Chang, 2003; Reinert, 2008; Reinert,
Gosh, and Kattel, 2016), with sustained growth and economic transformation associated
with industrial policy and the transformation of productive capability.3 However, the expe
rience of industrial policies has been uneven across countries, sectors, and historical peri
ods. Debate on industrial policy has tended to be influenced more by ideological con
tentions than by empirical evidence and economic history. The recent proliferation of lit
erature brings the divergent underlying perspectives to the fore.
From a long-term, strategic perspective, industrial policy underpins structural transfor

mation and catch-up, sustained economic growth, productivity gains, technological ad
vances, and broader socio-economic transformation. The approach taken in this chapter
is based on a broader conceptualization of industrial policy incorporating production, in
ternational trade, technology, and innovation, rather than a narrower definition confined
to government interventions or policy instruments. This means that the purpose and pat
terns of industrial policy and the perspectives underlying it are of paramount importance
(Oqubay, 2015; Best, 2018; Andreoni and Chang, 2019).
This chapter reviews the theory and practice of industrial policy and advances three sets
of arguments. First, it presents industrial policy as a vehicle of structural transformation
and wealth creation, in which manufacturing (and other dynamic, high-productivity, and
knowledge-intensive activities) exhibits increasing returns (both static and dynamic) and
drives structural change.4 It underlines the strategic role of exports, especially as a
source of international learning and relaxing balance-of-payment constraints. Second, it
underlines the dynamics of industrial policy as a conduit for technological learning and
the development of innovation capability (Best, 2018; Dosi and Nelson, 2018).5 Finally, it
shows that unevenness and adaptability have been key aspects of the practice of industri
al policies over the last two centuries, suggesting their continuing relevance for structur
al change, structural transformation, and catch-up, even in the radically transformed
global economic landscape of the twenty-first century, where policy space is more re
stricted than in the past.
The chapter is divided into six sections. Following the introduction, section 2.2 draws on
classical political economy and structuralist development economics to examine structur
al transformation as the prime foundation of industrial policy, and reviews related con
cepts such as the infant industry protection theory. In section 2.3 we explore evolving
thinking on industrial policy in the twentieth century with a focus on its role in technolog
ical learning and economic catch-up. The unprecedented rate of latecomer economy in
Page 2 of 48
dustrialization, together with rapid changes in technology, the (p. 19) economic land
scape, and firms’ domestic and international competition strategies, has influenced our
understanding of industrial policy. Section 2.4 reviews the state‒market relationship, and
the politics of industrial policy. Section 2.5 focuses on principles and practice, including
linkage effects and industrial ecosystems. Section 2.6 presents a summary of observations
and emerging issues in the changing landscape of industrial policy.
2.2 Theoretical Perspectives on Industrial Poli

cy
2.2.1 The Concept of Industrial Policy
An active industrial policy underpinning structural transformation and catch-up compris

es measures to develop productive and technological capabilities and selective interven
tions to build new industries and activities with dynamic efficiency (UNCTAD-UNIDO,
2011; Chang, 1994; Johnson, 1982; Amsden, 1989; Wade, 2004; Oqubay, 2015).6 An indus
trial policy is ‘a strategy that includes a range of implicit or explicit policy actions and in
struments selectively focused on specific industrial sectors and new activities for the pur
pose of shaping structural change and promoting catch-up in line with a broader national
vision and development strategy…Successful catching up has depended on active indus
trial policy. What distinguishes industrial policy is its “soul”, that is, the purpose and un
derlying pattern of industrial policy’ (Oqubay, 2015: 19, 294). This definition emphasizes
economic development strategies as the prime point of departure for inducing productive
investment and changing the political economy accordingly.
This conception assists a long-term policy approach that emphasizes synergies and con
nections with macroeconomic policies (such as exchange rate regimes, saving and invest
ment rates, financing systems), human capital development policies (including the orien
tation and quality of tertiary and vocational schools, the links between universities and in
dustry, and the focus on engineering and technology), and the development of physical in
frastructure (such as energy and modern transport). It also follows a tradition of
strengthening industrial competitiveness and institutions.
A key feature of industrial policy is that it constantly adapts to shifts in the international
and domestic environment, including constraints posed by international governing institu
tions such as the WTO.7 However, while multilateral trade rules have (p. 20) restricted the
policy space of countries, with implications for industrial policymaking, it has been ar
gued that many industrial policy instruments remain available to latecomer countries
wishing to pursue an industrialization agenda (UNECA, 2016). For some countries, in
cluding least developed countries (LDCs), multilateral rules either do not apply at all or
do so rather leniently. Environmental sustainability and the greening of industrial policy
has increasingly gained importance in policy debates in the late twentieth century (Math
ews, 2016a, 2017).8
Page 3 of 48
A pragmatic rather than dogmatic, doctrinaire approach to the design and execution of
industrial policy is a key requirement. New ideas can be tested, and old policies discard
ed if the measures deviate from facts on the ground. Successful outcomes imply the sub
stitution of new instruments for failed ones. Policy learning is rooted in learning by doing
and learning by experimentation, augmented by the search for new solutions and learn
ing from international experiences (Oqubay and Ohno, 2019). Pragmatism combined with
long-term vision serves as a compass while allowing the trial-and-error approach and has
been typical of industrial policy in East Asian economies.9 The post-1978 Chinese ap
proach to industrial policy and reform has been characterized by pragmatism, learning by
doing, and experimentation.10
2.2.2 Structural Transformation and Industrial Policy
2.2.2.1 Perspectives on Structural Transformation

Economic history shows that growth alone is not sufficient to sustain economic develop
ment. Structural change is characterized by sectoral shifts, sustained productivity
growth, and technological spillover, accompanied by shifts in demand, occupations, and
income levels, as well as institutional and socio-economic changes (Kuznets, 1973; Ocam
po, Rada, and Taylor, 2009). Structural transformation involves shifts from low- to high-
productivity activities and sectors, diversification into new activities and industries, and
sector-specific industrial deepening and upgrading. A process of fundamental structural
change necessitates the development and expansion of a country’s productive capacity,
namely, accumulation of various types of capabilities.11 Nonetheless, technical progress
does not necessarily occur uniformly across sectors or during (p. 21) different stages of
industrialization (Ocampo, Rada, and Taylor, 2009; Andreoni and Chang, 2017).12
Pasinetti states that structural transformation is not about temporary or short-term

changes but rather about fundamental shifts, ‘those changes in the composition that are
permanent and irreversible’ (Pasinetti, 1993: 1). Structural transformation implies the
transformation of production and hence the need for a production-centred framework.
Ocampo et al. (2009: 7) highlight ‘the composition of production activities, the associated
patterns of specialization in international trade, the technological capabilities of the econ
omy, including the education level of the labour force, the structure of ownership of fac
tors of production, the nature and development of basic state institutions, and the degree
of development and constraints under which certain markets operate (the absence of cer
tain segments of the financial market or the presence of a large underemployed labour
force)’ (own emphasis).
Thinkers and development economists have multiple views on how structural transforma
tion takes shape. The Prebisch‒Singer hypothesis focuses on what prevents structural
change in developing countries and leads to strategic implications for how to overcome
this obstacle (Prebisch, 1950). The trade imbalances between developed and developing
countries that are inherent in the economic structure and the core‒periphery relationship
are articulated in economic dependency theory.13 This hypothesis suggests that in the
long term the price of primary commodities declines relative to the price of manufactured
Page 4 of 48
goods, due to the greater income elasticity of demand observed relative to primary prod
ucts, and emphasizes the importance of import-substitution industrialization to kick-start
the process of structural change.
2.2.2.2 Manufacturing as an Engine of Growth and Structural Change

A proposition widely argued by the structuralist perspective but not universally accepted
is that the special properties of manufacturing, which are critical for the long-term dy
namism of both industrial and broader economic systems, include the ability to generate
linkage effects, increasing return to scale and productivity gains both in manufacturing it
self and in the broader economy through linkages, technological spillovers, and job cre
ation (both direct and indirect). The relative significance of these special features varies
across manufacturing industries and phases of development. Kaldor’s pioneering work on
growth laws underlines manufacturing’s special contribution to growth and productivity:
The first law is that there exists a strong causal relation between the growth of
manufacturing output and the growth of GDP. The second law states that there ex
ists a strong positive causal relation between the growth of manufacturing output
and the growth of productivity in manufacturing as a result of static and dynamic
returns to scale. This is also known as Verdoorn’s law. The third law (p. 22) states
that there exists a strong positive causal relation between the rates at which the
manufacturing sector expands and the growth of productivity outside the manu
facturing sector because of diminishing returns in agriculture and many petty ser
vice activities which supply labour to the industrial sector.
(Thirlwall, 2002: 42; own emphasis)
From Kaldor’s perspective, manufacturing is associated with an enormous increase in the

‘technical dynamism’ of the economic system, with the introduction of new knowledge in
ducing productive investment (Andreoni and Chang, 2017).14 This inextricable interaction
between technical change and investment impels the accumulation of technology and
capital, together with sustained productivity growth and dynamism. Capital intensity and
economic viability are, however, determined by the size and nature of demand as well as
technological advances. Kaldor highlights how ‘both technology and demand contribute
to determining the degree of capital intensity of the economic system as a whole, and
both the evolution of technology and the evolution of demand govern its movement
through time. But no relevant role on these matters can be attributed to the rate of prof
it…The most important characteristic of capitalist business enterprises is the continuous
change and improvement in the methods of production’ (1980: 293). Hence, knowledge
intensity and technology remain the key measures of development.15
As per Young (1928), manufacturing offers countries the opportunity to capture increas
ing returns as a result of manufacturing specialization.16 Historically, manufacturing has
been viewed to have stronger elasticity of demand than primary commodities—though
this distinction may be changing—and increasing returns to scale (Young, 1928; Prebisch,
1950). Pasinetti states that ‘The physical quantity of each commodity to be produced is
Page 5 of 48
determined by demand…the nature of human needs and preferences gives rise to entirely
different compositions of demand, and, therefore, different structures of production and
employment, at the various level of real per capita income’ (Pasinetti, 1981: 251).
Young (1928) highlights that the notion of increasing returns comprises static increasing
returns, captured in firms, and dynamic increasing returns, generated in networks or
connections of firms and clustering, driving specialization and differentiation. Kaldor
(1967) adds an important dimension, namely, the ‘scope for learning-by-doing’ at the indi
vidual, firm-level, sector-level, and industrial workforce, which is evident in the ‘special
properties’ of manufacturing activities. However, it is noteworthy that sectoral classifica
tion has evolved to reflect the changing (p. 23) characteristics of production and technolo
gy, and the growing heterogeneity in the nature and characteristics of activities within
sectors.
Moreover, the nature of inter-sectoral linkages is evolving as the boundaries between sec
tors are changing; examples include the ‘servification’ featured in the linkage between
services and manufacturing, and the ‘industrialization of freshness’ that shapes the link
age between agriculture and manufacturing (Cramer and Sender, 2019).17 While activi
ties with dynamic efficiency and arguably a sectoral approach are a critical aspect of in
dustrial policy, a systematic survey and analysis is essential to understand the nature of
such activities (Cramer and Tregenna, 2020).18 These changes, including the blurred
boundaries between sectors and the increasing breadth and heterogeneity of production
activities, may arguably offer widening scope for industrial policies while calling for in
dustrial policy to be adapted to the new environment.
2.2.2.2.1 Linkages and complementarities

It has been argued that manufacturing industries also complement agriculture and ser
vices, and are associated with stronger inter-sectoral forward and backward production
linkage effects (see Hirschman, 1958, among others). Historically, early industrialization
is accompanied by the transformation of agriculture, both as the initial source of demand
and as a provider of food surplus to urban populations.19 Agricultural productivity is in
creased through both research and widely applied technological advances, with high-pro
ductivity agriculture exhibiting the stronger dynamism of manufacturing. Manufacturing
has similarly stimulated the growth of services, with the transfer of activities such as dis
tribution, logistics, and marketing to the service sector and the outsourcing of other ac
tivities such as accounting and human resource management (Andreoni and Chang,
2017). Knowledge intensity in the service sector varies, ranging from the informal sector
to high-productivity industries such as ICT and logistics, which are critical for the com
petitiveness of manufacturing. The potential and market of services is partly shaped by
the scale and sophistication level of manufacturing.
2.2.2.3 The Strategic Role of Exports

There are conflicting views on the role of exports. A market-friendly or neo-liberal view
regards opening to international markets and liberalization as a route to international
competitiveness.20 The other view argues that what matters is not trade liberalization or
Page 6 of 48
integration with the global market but how each country inserts itself into interna
(p. 24)
tional trade (Ocampo, Rada, and Taylor, 2009). Hirschman ([1968] 2013: 125) argues that
export is critical for late latecomers because of its implications for economic transforma
tion:
First, through exports they would overcome whatever obstacles of market size lim
it their growth or prevent their establishment. Second, through exports they
would loosen the balance-of-payments constraint which may otherwise prevent ca
pacity operation of existing industries as well as establishment of new industries.
Finally, by competing in world markets, industries would be forced to attain and
maintain high standards of efficiency and product quality and would thereby ac
quire defence against oligopolistic collusion and decay to which they often suc
cumb in highly protected, small local markets.
Exports and positioning in international trade have been recognized as pivotal for growth,
and exports have been associated with structural transformation, especially with industri
alization and manufactured exports (Thirlwall, 2013). Policymakers may pursue a variety
of export promotion strategies resulting in different outcomes. However, from a struc
turalist perspective, there are at least three fundamental reasons why exports play a
strategic role in economic growth and structural change.
First, exports are sources and propellers of international learning, as positioning in inter
national trade requires competitiveness in terms of quality, delivery time, and cost. Ex
porters also serve as conduits for indirect exporters and domestic suppliers, and generate
spillover effects in local firms through management skills, workforce transfer, and pur
poseful emulation. Although international learning is the principal gain from exports,
knowledge transfer is less mobile than traded goods. Pasinetti highlights that:
When firms in one country are challenged by lower priced products from abroad,
they will either learn how to cut down costs or close down…a widespread failure
to realise that the primary source of international gains is not mobility of goods
but mobility of knowledge…International learning must therefore remain, for any
country, the major and primary aim. This principle of economic policy is one of
general and unconditional validity…it generally comes from its coincidence with
the primary aim (learning) of any international policy.
(Pasinetti, 1981: 259, 271‒2)21
Second, exports represent the most sustainable response to the balance-of-payments con
straints that can slow down economic growth and impede rapid industrialization, which
ultimately slows down the process of fundamental structural change and can trigger
macroeconomic crises. An economy that is unable to use exports to generate (p. 25) for
eign exchange is likely to be dependent on less reliable sources such as overseas aid and
external loans.22
Page 7 of 48
Third, where a country has limited capability to add value, exports can stimulate the local
economy by creating markets for surplus domestic production as well as natural re
sources for which there is insufficient demand locally. This underlines the complementari
ty of export-led industrialization (ELI) and import-substitution industrialization (ISI)
strategies. The ELI strategy, carefully synchronized with ISI, has offered an accelerated
industrialization path to twentieth-century latecomers such as Japan, the East Asian new
ly industrializing economies (South Korea, Taiwan, Singapore), and China. In many in
stances, ISI is the precursor to ELI (Amsden, 1989).23 Stiffening international competition
and a crowded marketplace have led some observers to predict that ELI will be a difficult
strategy to pursue successfully in the twenty-first century (UNCTAD, 2016).24 However,
this makes exports a more relevant and pressing source of learning than ever.
Foreign direct investment (FDI), which has shown exponential growth since the 1970s, is
arguably associated with export growth and international learning.25 However, the long-
term contribution to the host economy is neither automatic nor inherent in FDI. Akyüz
(2017: 169‒70) asserts that ‘the contribution of FDI to balance of payments and external
financial stability, and growth and industrialization is highly contentious…none of these
are intrinsic qualities of FDI. Rather policy in host countries plays a key role in determin
ing the impacts of FDI in these areas.’ It is worth noting first, that countries like South
Korea have placed little reliance on FDI, and second, that new research shows that a
huge share of FDI is really ‘phantom’, being merely transfers of profits across countries
but within firms.
With a deliberate and effective industrial policy in the host economy, FDI can serve as a
source of international learning and may have significant spillover effects (technological,
management skills, and market capability), especially at earlier stages of industrializa
tion, and these may be more important for economic catch-up than FDI’s contribution to
capital formation and balance of payments (Lee, 2019a; Amsden, 1989; Akyüz, 2017).
However, maximizing the benefits from FDI requires (p. 26) local absorptive capacity both
at the firm level and in the economy as a whole. Some economies, such as China and Sin
gapore, have been more effective than others in their management of FDI.
Governments of host economies have used policy innovation and strategies to promote
and manage FDI, which reflects specific local conditions. These include careful identifica
tion of foreign firm capabilities, targeting specific industries, careful management of
know-how transfer channels (such as joint ventures, licensing, management services),
and enforcing performance requirements, incentives, and restrictions. Akyüz (2017)
states that competition, imitation, demonstration effects, labour turnover, forward and
backward linkages, and R&D activities are all drivers of technological spillover.
2.2.2.4 Structural Transformation and its Implications for Industrial Policy

Perspectives on structural transformation have fundamental implications for the design
and practice of industrial policy in three key areas. First, structural transformation rein
forces the purpose of industrial policy and the strategic importance of export-led industri
alization not only for small economies with a limited domestic market but also for those
Page 8 of 48
that enjoy a large domestic market. With manufacturing-led industrialization and exports,
critical drivers of sustained growth and economic transformation, productive investment,
industrial deepening, and upgrading within specific industries and across industrial sec
tors become critical.
Second, a sectoral approach is essential for targeting industrial sectors and economic ac
tivities (products, production processes, technology, linkages) as foundations of industrial
policy. Individual economic and industrial activities have different characteristics and
growth dynamics. It is important to ensure that industrial sectors and activities are priori
tized based on three characteristics: technological intensity and the scope for technologi
cal learning; the dynamics of linkage effects (such as forward and backward linkages);
and demand elasticity.
Third, supporting and nurturing industrial sectors requires appropriately aligned instru
ments that promote the move towards high-productivity activities and highly productive
investments. Matching incentives with performance (Amsden (1989) calls it a ‘reciprocal
monitoring system’) is critical for industrial policy and exports are perfectly positioned to
apply pressure for learning. Hobday (2000: 156) highlights that ‘exports, whether with
TNCs or from local firms, acted as a focusing device for technological learning and invest
ment’. Understanding the structure of the industry is critical for designing industrial poli
cy instruments that can adapt to the changing industrial structure. Firms can promote in
ternational learning as well as exploit increasing economies of scale and scope by imple
menting policies that both support big businesses and foster start-ups. As Hirschman
(1958) argues, industrial policy expedites structural change by anticipating developments
and patterns, by removing binding constraints, and by championing emerging and new
activities.
(p. 27) 2.2.3 The Infant Industry Theory
The three pillars shaping the development and practice of industrial policy are, arguably,
the theory of infant industry, the comparative advantage perspective, and the concept of
dynamic and interdependent linkage effects (List, 1856; Hamilton, 1934; Hirschman,
1958; Andreoni and Chang, 2019). The aim of protecting infant industries is to provide
time for the learning and accumulation of the experience that will enable them to reach
international levels of competitiveness (Pasinetti, 1981: 272). The theory of infant indus
try originates in the classical political economy of the eighteenth century and the pioneer
ing work of List and Hamilton. Alexander Hamilton (1755‒1804), the first secretary of the
US Treasury, was the architect of US industrial policy who wrote a famous ‘Report on
Manufacturers’ in 1791.26 In his comprehensive four-volume National System, Friedrich
List (1789‒1846), a political economist and forefather of the German historical school of
economics, set out an industrial policy that focused on national systems of production ca
pacity, protection of domestic industry, and innovation. The key perspectives in these clas
sical works refute David Ricardo’s notions of free trade and comparative advantage. List
(1856: 297) showed that England rose to a leading industrial and military power by ad
hering to three strategies: ‘First, to prefer constantly the importation of productive power
Page 9 of 48
to that of commodities; second, to maintain and carefully protect the development of pro
ductive power; third, to import only raw materials and agricultural products, and to ex
port only manufactured articles.’27
The central arguments relating to infant industry theory can be summarized as follows:
a) Manufacturing as the key to wealth creation and building economic and military
might. The analysis was based on the importance of economic diversification, mod
ernization, and economic returns. Hamilton based his argument on full employment
and labour productivity, increased demand and domestic supply of industrial prod
ucts, and access to a secure domestic market.
b) The notion that nations should develop national production systems rather than
international or sub-national economies and should focus on production capacity in
preference to exchanges and trade.
c) The importance of selecting sectors with higher value addition and knowledge in
tensity, attracting talent, and providing incentives to support selected sectors.
d) The necessity of protecting domestic industry until it is able to compete locally
and then with advanced economies, the superiority of domestic manufacturing over
the import of industrial goods, and the export of manufactured goods as a true indi
cator of economic power.
Two centuries later, the infant industry hypothesis remains part of the debate
(p. 28)
around industrial policy (Chang, 2003). This notion is related to the issue of comparative
advantage and its strategic implications for industrial policy.28 Historical and empirical
evidence shows there are countries that have achieved significant progress in climbing
the development ladder by building competitive advantage to sustain catch-up and struc
tural transformation without relying on their latent competitive advantage (Chang, 2003).
This does not imply deviating from the external and internal environment to follow an ir
rational and unrealistic path.
A plausible view is that conforming with and defying comparative advantages are not mu
tually exclusive, often occurring under the same national strategy but across different
sectors and at different stages. Schwartz (2010) argues that advanced economies and
newly industrializing economies of the late twentieth century had to rely on both the Ri
cardian strategy of using latent comparative advantage and the Kaldorian strategy of cre
ating new comparative advantage.29 Oqubay (2015: 295; emphasis added) highlights:
An active industrial policy, while initially dependent on and overlapping with a Ri
cardian strategy (relying on comparative advantage in agricultural exports and
low-value light manufacturing), will eventually shift its focus to a more Kaldorian
strategy. A Ricardian strategy on its own can neither bring structural change to
the economy nor achieve catch-up. Ultimately, it is the Kaldorian strategy (which
partly ignores factor disadvantages and advantages, focuses on manufacturing ex
ports, and is investment-driven) that can address the challenges of catch up in
Page 10 of 48
terms of investment concentration, learning, and innovation…This is what climb

ing the ladder means.
While it is necessary to fully exploit revealed latent comparative advantage, what matters
most for economic catch-up and deepening of structural transformation is the capacity to
create and shape comparative advantages. In a nutshell, structural transformation should
underpin industrial policy. Nonetheless, this will be incomplete without an equivalent per
spective on technological learning and catch-up which has become more marked in the
twentieth century—the focus of the discussion in section 2.3.
2.3 Evolving Thinking on Industrial Policy:

Technological Learning and Catch-up
Thinking on industrial policy evolved profoundly in the twentieth century, with an increas
ing number of latecomers recording accelerated industrialization and developing (p. 29)
technological capabilities at an unprecedented rate (see Nayyar, 2016, 2019, among oth
ers). Key elements in this evolution are the rapidly changing global economic landscape,
the unmatched pace and shift of technological advancements, and how firms and coun
tries compete. This has encouraged empirical research and theoretical work on the un
even development of capitalism, the diversity of development paths and economic catch-
up, the process of technological learning, and the centrality of productive and technologi
cal capabilities. The proliferation of debate has underpinned an evolving view of industri
al policy as the right strategy to ensure technological learning and economic catch-up,
which is the focus of this section. The rarity of middle-income countries escaping from the
‘middle-income trap’ has become an additional impetus to the study of the nexus between
industrial policy and technological learning and catch-up.
2.3.1 Technical Change and Economic Development
An evolutionary economics or Schumpeterian evolutionary tradition offers an important

view of industrial policy as well as new concepts. Evolutionary theory builds on the cen
trality of technical change or technological change as the driver of capitalism and stress
es the importance of learning and capability development as a key driver of competitive
ness of firms (Dosi and Nelson, 2018; Nelson and Winter, 1982; Mowery and Rosenberg,
1989; Rosenberg, 1994). Classical political economy (Smith, 1776; Babbage, 2005; List,
1856) emphasizes the importance of technological and technical change and domestic
technological capability for building an innovation-driven knowledge economy.30
Although technological advancement accelerated in the early days of the first industrial
revolution in England, it had already been dependent on science for many centuries, for
example in China. Kuznets (1966: 9) states that ‘the epochal innovation that distinguished
the modern economic epoch is the intended application of science to problems of econom
ic production’. According to Kaldor, innovation implies the introduction of new knowledge
in which the eagerness to absorb technological change and the preparedness to invest in
Page 11 of 48
firms are central. Kaldor states that ‘the most important characteristic of capitalist busi
ness enterprise is the continuous change and improvement in the methods of
production’ (Kaldor, 1980: 293). Pasinetti (1981: 274) underscores the centrality of tech
nical progress: ‘it is only by absorbing technical knowledge that the poor countries will be
able to permanently increase their wealth.’
Technical change is the process by which new technologies are being constantly intro
duced into economic activities by incremental improvements and innovation. It usually
comes in various forms and is path dependent. Rosenberg (1994: 9) highlights that ‘the
growth of technological knowledge is fundamental to the improvement of (p. 30) econom
ic performance…technological changes are often “path dependent” in the sense that their
form or direction tends to be influenced strongly by the particular sequence of earlier
events, out of which a new technology has emerged’. The duality of diversity and path de
pendence is a process imbued with uncertainty.31 Kaldor (1980: 233) highlights that tech
nological progress and productivity increase, not in isolation, but alongside the growth of
population and accumulation of capital.
In his pioneering work on creative destruction, Schumpeter ([1934] 2017, 1950) held that
industrial capitalism constantly rejuvenates itself in a cycle of technical and economic
progress involving heroic entrepreneurs or inventors, imitation by other firms, an invest
ment boom, an ensuing depression, and new ‘innovating’ investment.32 Innovation is both
incremental and radical, and includes products, processes, markets, and organizational
set-ups that are new to the country or the firm (OECD, 2005).33 Thus, innovation is de
fined as a process by which firms master and implement the design and production of
goods and services that are new to them, irrespective of whether they are new to their
competitors, their countries, or the world (see, for example, Mytelka and Tesfachew,
1998; Mytelka, 2000).
Dosi and Nelson (2018), among others, highlight that tacit knowledge plays an important
role in know-how transfer, technology diffusion, and innovation, and is nested in organiza
tions; and also that innovation is not a technical exercise that can be detached from politi
cal economy.34 They further emphasize that the focus on capability development is more
valuable than financial incentives. Latecomer economies may focus on learning by doing
or imitation, incremental innovations, and ultimately new innovations that require domes
tic R&D capability and the development of science and technology infrastructure (Ams
den, 1989; Kim, 2000; Kim and Nelson, 2000). Learning (p. 31) and technological catch-
up, and building domestic capability, which also includes growing investment in R&D,
were all essential components of successful cases of economic catch-up (Cimoli, Dosi, and
Stiglitz, 2009; Lee, 2019a).
Absorptive capacity is an important component of innovation and technological capability.

Cohen and Levinthal (1990) define absorptive capacity as the ability to identify, assimi
late, and apply knowledge; they make a distinction between learning capability and prob
lem-solving capability. They also emphasize the importance of multi-functional teams, di
versity of knowledge and experience within firms, and openness to the external environ
Page 12 of 48
ment of buyers, suppliers, and research centres. Contributions on innovation systems by

Nelson and Winter (1982) and Kim and Nelson (2000) focus on national innovation sys
tems as a proxy for building absorptive capacity, including the education system, the R&D
system, government funding of research, universities, research centres, government
agencies, and interactions between these, also captured by Best’s (2018) production-cen
tric ‘capability triad’, namely production systems, business governance, skill formation,
and their interconnectedness. Chung and Lee (2015) show the origins of absorptive ca
pacity in latecomer economies such as South Korea. Breschi and Malerba (2005) have
emphasized the importance of addressing innovation systems at both national and sec
toral levels.
The market failure paradigm (derived from the neoclassical view of innovation and capa
bility development) is based on information asymmetry and an assumption that knowl
edge is freely available to all firms, and ignores the dimension of ‘issues of
utilization’ (Rosenberg, 1994; Mowery and Rosenberg, 1989).35 The role of the state is
confined to facilitation and financial rewards and incentives for innovation. This view
point ignores the inescapable truth that firms cannot capture all the benefits and risks of
financing innovations, or the lags intrinsic in adopting technology. It views innovation as
an exogenous factor, ignoring its multiple sources, complexity, path dependency, and un
clear boundaries. It views innovation as dissociated from issues of practical application.
Finally, it focuses on incentives rather than firms’ capabilities or innovation systems at
both sectoral and national levels.36
2.3.2 Theoretical Underpinning of Catch-up and Late Development
Unevenness among economies and regions is an essential characteristic of capitalism and

industrialization (Gerschenkron, 1962; Chang, 2003; Schwartz, 2010).37 Economic (p. 32)
history shows that countries do not have uniform or linear development paths and there
is no standard prescription. Late latecomers are compelled to rely on their abundant
‘backward’ resources and pursue a strategy that induces disequilibrium and imbalances
(Hirschman, 1958). Gerschenkron’s (1962: 45) pioneering work on catch-up and late de
velopment shows that latecomers, despite the disadvantage of being followers, can ex
ploit the advantages of backwardness or late development: ‘the more backward countries
are, the great spurt of industrial development occurred despite the lack of these prerequi
sites…[and] the higher the degree of backwardness, the more discontinuous the develop
ment is likely to be.’ This hypothesis confirms that catch-up requires a deliberate strategy
and policy, and active state intervention. Forerunners provide compelling learning for
latecomers, and contact with forerunners is an essential condition for late development.
Page 13 of 48
Lee and Malerba (2018) define economic catch-up as narrowing the distance between
latecomers and frontrunners or leaders. Abramovitz, writing on catch-up and the advan
tages of backwardness, reviews three key aspects of the catch-up hypothesis:
1) Its prime concern is a single aspect of economic relations among countries: tech
nological borrowing by followers, meaning the level of technology embodied in a
country’s capital stock (Abramovitz, 1989: 232).
2) Being a latecomer carries with it the potential for faster productivity gains: ‘The
catch-up hypothesis asserts that being backward in level of productivity carries a po
tential for rapid advance…Followers tend to catch up faster if they are initially more
backward…Backwardness carries an opportunity for modernization in disembodied
as well as embodied technology.’ Abramovitz further states: ‘those who are behind,
however, have the potential to make a larger leap. New capital can embody the fron
tier of knowledge, but the capital it replaced was technologically
superannuated’ (Abramovitz, 1989: 221).
3) However, as the distance between latecomer and frontrunner narrows, productivi
ty gains slow down. Abramovitz (1989: 221, 232‒3) highlights: ‘a follower’s potential
for growth weakens as its productivity level converges toward that of the leader’;
and ‘countries in the course of catching up, however, exploit the possibility of ad
vance scale dependent technologies by import substitution and expansion of exports’.
Learning has become the hallmark of economic catch-up and late development. Amsden
(1989) states that ‘diversity notwithstanding, all late industrializers have in common in
dustrialization on the basis of learning, which has conditioned how they have
behaved’ (Amsden, 1989). One stream of research has concentrated on both catch-up
strategies and the development of productive and technological capabilities, based on em
pirical evidence from the late industrializers of the late twentieth century (Amsden, 1989;
Best, 2018; Lee, 2013, 2019a). Lee’s work on East Asian, especially Korean, economic
catch-up shows that it is not automatic and cannot be accomplished merely by imitating
forerunners, because forerunners are also constantly advancing and developing higher
capabilities. Hence, a different catch-up path is required.
(p. 33) Different strategies are used for different stages of catch-up. Focus on imitation,
learning from forerunners and insertion into the global value chain are essential in early-
stage industrialization, but this alone will not enable sustained catch-up. According to Lee
(2019b), latecomers face catch-up paradoxes: each latecomer has to be different, the pace
of catch-up has to be faster and requires detour strategies, while leapfrogging requires
exceptional innovation and development of R&D capabilities. In the early phases, relying
on foreign firms, insertion into global value chains, and excelling in learning by doing and
imitation is essential, together with growing investment in human capital, innovation ca
pabilities, and the development of big businesses. In the second stage, where the focus is
on adaptive imitation, the latecomer needs to work closely with frontrunner technology,
even though this poses a significant threat. Strictly enforced intellectual property and
other protections create new difficulties for latecomers seeking to move into innovation
leadership. According to Lee (2019b), latecomers need to focus on leapfrogging in short-
Page 14 of 48
cycle industries, gradually moving to long-cycle technologies, and building local value
chains into global ones.
Recent research has shown that catch-up from middle-income to high-income levels
rarely occurs, and a new term, ‘middle-income trap’ (MIT) has been coined. However,
while it is true that more and more countries are remaining trapped at middle-income lev
el, the evidence used to explain the notion of the middle-income trap is based on a partial
premise.38 The explanation focuses on conditions, rather than dynamic factors such as
productive and technological capabilities, which are key drivers of productivity growth.39
A more plausible explanation would be low investment in technological and innovation ca
pability, which becomes more difficult as productivity converges to that of the frontrun
ners (see Abramovitz, 1989, 1994; Paus, 2019; Lee, 2019b). Paus (2019: 2) highlights that
‘Productive transformation from commodity production to higher value added, more
knowledge-intensive activities is at the heart of the transition from a middle-income to a
high-income economy…Productivity growth is the distinguishing characteristic between
upper middle-income countries which transitioned to high-income countries and those
that did not.’ Similarly, according to Lin, ‘the middle-income trap is a result of a middle-
income country’s failure to grow labour productivity through technological innovation
and industrial upgrading’ (Lin, 2016). For UNCTAD, catch-up is dependent on the nature
of the industrialization path or trajectory followed, (p. 34) which may lead to successful
catch-up or getting caught in the low- or middle-level development trap. UNCTAD classi
fies these stages as catch-up industrialization, stalled industrialization, and premature
de-industrialization (UNCTAD, 2016).40
2.3.3 Technological Learning and Industrial Policy
2.3.3.1 Technological Learning

From a catch-up perspective, learning is central in late development. It has two dimen
sions: technological learning in firms and policy learning by governments.41 Learning is
the prime source of learning and learning is embedded in learning by doing (Oqubay and
Ohno, 2019). Governments and firms learn from other firms or other governments, but al
so and perhaps more importantly, from themselves. Emulation involves learning from oth
ers in the form of copying or imitation, adaptive innovation, or radical or new innovations.
Learning requires an open system, inter-firm interactions, contact with foreign knowl
edge, the compulsion and readiness to learn, while depth of learning depends on absorp
tive capacity. The experience of the East Asian countries shows that they were not only
pragmatic, but also strategic in their choices of policies and sectors, and in designing
their industrial and international trade policies. Hence, pragmatism is at the heart of in
dustrial policy as policy decisions have to be based on what works or does not, and no
policy action is immune from the test of practice. Pragmatism rejects dogmatism or ro
manticism: it adopts what works and serves the long-term vision and strategy, while con
stantly adapting to international and domestic reality.
Page 15 of 48
The sources and dynamics of learning are diverse and constantly changing. Successful
economies such as the East Asian newly industrialized economies have perfected (p. 35)
their trials and errors, and have strictly adhered to a pragmatic national vision and devel
opment strategy.42 The driver of technological learning and industrial policy is the syner
gy between industrialization or productive capacity, exports or market focus, and innova
tion. Exports provide the broader market opportunity to expand economies of scale and
scope, and exploit productivity gains. In addition to their strategic role as a key driver of
growth, exports remain a prime disciplining device for firms, promoting international
learning, energizing technological learning and innovation, and attesting to their interna
tional competitiveness (Amsden, 1989, 2001; Wade, 2004).43
Advances in the technological and knowledge frontier are intimately tied to accumulation
of capital. Hirschman highlights that ‘the complementary effect of investment is the es
sential mechanism by which new energies are channelled toward the development
process…where one disequilibrium calls forth the development mover [which] in turn
leads to a similar equilibrium and so on ad infinitum’ (1958: 72).
This is in line with the proposition of technological learning as the conduit of structural
change and driver of economic catch-up, the foundations of the production-centric para
digm explored by leading thinkers on capability development (Best, 2001, 2018; Lee,
2019a). Michael Best’s How Growth Really Happens (2018) describes the ‘capability tri
ad’ of production capability, skills, and governance in firms. This perspective integrates
the catch-up drive, the active role of the state as organizer of production, the dynamism
of firms, and shifting production systems.
The development of technological and innovation capabilities underpins the symbiotic and
dynamic relationship between technological learning, industrial policy, and catch-up. A
coherent and strategic approach to R&D, technology commercialization, support mecha
nisms, education, and skills targets key dynamic industries (and firms) and new technolo
gies, rewarding learning and providing consistent and comprehensive support.
2.3.3.2 Innovation Systems and Infrastructure

The structure of innovation systems, and the science and technology infrastructure, are
foundations for developing innovation and technological capability. The notion of innova
tion systems was popularized in the 1980s by Nelson and Winter (1982), Mowery and
Rosenburg (1989), and Kim and Nelson (2000), among others. The innovation system
framework provides a holistic view of the conditions necessary to create an environment
that supports innovative activities by enterprises. In this respect, it describes a nation’s
capabilities, encompassing firms, industries, institutional networks, and the interaction
among the key players.44 National innovation systems (NISs) have become increasingly
open with regard to knowledge flows, including (p. 36) technologies and know-how ac
quired from abroad through capital and intermediate goods, purchases of patents and li
cences, technical alliances between firms in different countries, trade in services such as
technical consultancies, FDI, and so on. This has enhanced the abilities of local firms to
innovate through, for example, learning, linkages, and demonstration effects. With the in
Page 16 of 48
fluence of external actors, the NIS framework is not as closed or ‘rigid’ as it was once
perceived. Cohen and Levinthal (1990: 128) highlight that ‘the ability to exploit external
knowledge is a critical component of innovative capabilities’.45
2.3.3.3 Implications for Technological Learning and Industrial Policy

This premise has important implications for technological learning and industrial policy.
First, while national innovation systems focus on country or economy level, the NIS may
be augmented by sectoral innovation systems as key shifts occur in industries. Malerba
states that a wider view of industrial sectors that includes related support industries is
critical for innovation and technological capabilities, and for ecosystems and cluster dy
namics. Breschi and Malerba (2005: 4) suggest ‘some key drivers of agglomeration are
sector-specific, leading to distinct patterns of concentration’. This view has implications
for the life cycle of industrial hubs, with distinct drivers of processes during the initiation
of industrial hubs and their maintenance.
Second, technological learning is associated with the evolving institutional structures of

research systems. With the increasing role of big businesses in the twentieth and twenty-
first centuries in the United States, Western Europe, Japan, the East Asian economies and
China, national economies have single-mindedly pursued the development of national
champions, for example, in South Korea and China (Lee, 2019a; Nolan, 2014).46 While the
role of big business is critical for the exploitation of the global market, its increasing in
vestment in R&D is also important and should be matched with support of new technolo
gy start-ups and spin-outs to continuously reinvigorate innovation.47 It is through new
start-ups, especially in the technology sphere, that the economy continuously renews and
upgrades.
Third, collaboration among the key economic players (or actors) is critical for
(p. 37)
building and developing a dynamic innovation system. The roles of industrial firms, the
academic community, and government-funded independent research centres (as well as
some private ones), as well as the orientation and focus of higher education and industry‒
university linkages, are critical elements of the science and technology infrastructure.
The development of innovation and technological capabilities should not be separated
from the development of production capacity and market promotion. Although policy in
novations vary across countries and times, they enhance coordination among lead agen
cies and strengthen strategic development.
Fourth, the ‘middle-income trap’ can be associated with inadequate investment in new in
novations and technologies. Resource allocation is a proxy to measure the focus on inno
vation and should be reviewed with related issues, such as the contribution of firms, uni
versities, and research centres.
Finally, the distinction between innovative activities and the nature of innovation has sig
nificant implications for policy approaches.48 At early stages of industrialization, copying,
imitation, FDI, and global value chains may be used as prime sources of learning. Howev
er, economies increasingly need to develop their own capabilities, and add incremental in
Page 17 of 48
novation by moving to ‘adaptive imitation’ (a term coined by Linsu Kim, meaning internal
ization of knowledge and technology). Ultimately economies have to focus on new indus
tries and technologies (‘leapfrogging’, as Keun Lee terms it) primarily by building on
growing investment in innovation and production capability.
The mechanisms and instruments vary depending on the context (industry, technology, ex
ternal environment) and require thoroughly thought-out policy design. For instance, re
verse engineering, licensing, insertion into GVCs, attraction of FDI, and sub-contracting
are common during the early and middle stages of industrialization. Technological insti
tutes, the targeted circulation of talent, and overseas training and education apply in the
middle stages of industrialization. Ultimately, a focus on domestic capability and the pro
motion of big business, leapfrogging to new industries and technologies with a focus on
short-cycle industries, massive investment in R&D, and the acquisition of foreign firms
are critical for building an innovation-driven economy.49 In the electronics and semicon
ductor industries, local firms have been used as original equipment manufacturer (OEM)
during initial industrialization, upgrading to own design and manufacturer (ODM), and
subsequently own brand manufacturer (OBM). This approach will, however, differ across
sectors.50
A commitment to structural transformation and developing technological capability re

quires the support and intervention of the state, a creative state–market interaction, and
has to be embedded in political economy. This is discussed in section 2.4. (p. 38)
2.4 The State and the Politics of Industrial Poli

cy
2.4.1 The Dynamics of the State‒Market Mechanism
The existence of self-regulating markets is a common myth; in reality the internal work
ings and consequences of capitalism cannot function without state intervention. Polanyi
(1944: xxvi) highlights that ‘the idea of a self-adjusting market implied a stark utopia.
Such an institution could not exist for any length of time without annihilating the human
and natural substance of society…Real market societies need the state to play an active
role in managing markets, and that role requires political decision making; it cannot be
reduced to some kind of technical or administrative function.’ This perspective has signifi
cant implication for the nature of industrial policy (Mazzucato, 2011; Wade, 2004; Ams
den, 1989).
State‒market forces are in constant tension in various sectors, economic activities, and at
different times, and a perfect match is always difficult to achieve. There are three vari
ants on the state‒market mechanism relationship. The first is market fundamentalism
which rejects government intervention as harmful. It advocates that governments should
restrain themselves from intervening in the functioning of the market with the rational re
allocation of resources. The neo-liberal Washington Consensus, orchestrated by IFIs and
Page 18 of 48
mainstream scholars, typified this laissez-faire (‘leave it to the market’) view, advocating
the dismantling of state apparatus, privatization, and the opening up of markets. This
view idolizes the free-market economy and rejects the need for industrial policy.
A second variant is the market failure view, which argues that market mechanisms and
the market economy do not function optimally in all times and conditions. This view ac
cepts that the government should intervene selectively to stabilize the macro economy,
for example, where there is market failure, such as infrastructure development, or devel
opment of human capital. The point of market failure policy recommendations is that they
are designed to replicate how a perfectly competitive market would look. The state is con
fined to a ‘facilitative’ role, its interventions ‘qualified’, and is constrained to follow latent
comparative advantage as this does not imply strategic intervention. This view has gained
widespread influence in mainstream economics and has become dominant within neoclas
sical economics.
A third variant is the state activism perspective that believes market mechanisms are in
stitutions that need to be nurtured, and advocates that the state should intervene to en
hance structural transformation and technological catch-up, playing a leading role as or
ganizer of production, with the aim of promoting a dynamic private sector (Best, 2018;
Lee and Malerba, 2018; Wade, 2004; Amsden, 1989). This view recognizes that capitalism
involves economic players, primarily the state, firms, labour, families, and non-govern
ment players, and sees capitalism as a mixed economy where markets and (p. 39) non-
market forces function. The activist state intervenes beyond market failure to shape and
create market forces, with strategic interventions to develop competitive advantage, and
works hand in hand with the private sector. Mazzucato (2011: 5) highlights that ‘When
not taking a leading role, the State becomes a poor imitator of private-sector behaviours,
rather than a real alternative. It is a key partner of the private sector—and often a more
daring one, willing to take risks that business won’t.’ The purpose and modality of inter
ventions are crucial, as Johnson (1982: 17) highlights: ‘the issue is not one of state inter
vention in the economy…all states intervene in their economies for various reasons…the
question is how the government intervenes and for what purpose’.
The state activism view is supported by both economic history and deductive observa
tions. Such governments are known variously as ‘activist’, ‘entrepreneurial’, and ‘devel
opmental’ states. The term ‘developmental’ tends to be used as a neat and rigid charac
terization of a government. The purpose of a developmental state, however, changes dur
ing different stages of industrialization and economic catch-up, and across sectors. Exist
ing literature on the developmental state takes two views: the static and the dynamic.
Some scholarly works have seen the role of industrial policies as dependent on the quality
of bureaucracy to the extent of being dissociated from politics. Another variant views de
velopmental states as ‘pure prototypes’ detached from the international and national con
text where they function. A more dynamic perspective is to link developmental states with
the deliberate pursuit of developmental goals, a grand vision interacting with society, in
which the nature of the state constantly evolves on the predatory‒developmental state
continuum (Chang, 2003; Oqubay, 2015; Thurbon and Weiss, 2016).51 It should be noted
Page 19 of 48
that there is a broader issue here. Not just that the very idea of a developmental state is
usually ahistorical, but that industrial policy does not exist within an institutional and po
litical vacuum. The policies that have ‘worked’ in one place may fail in other places, and
this may be because of the political and institutional context (Haggard, 2018; Thurbon
and Weiss, 2016). Schwartz (2010) argues for the essential role of the state in latecomers
or late-latecomers.
2.4.2 Politics and Political Economy
While it is widely recognized that politics and political economy matters for economic
policies, the various schools have divergent views. The Marxist view revolves around
class struggle and exploitation, while market-friendly views (based on methodological in
dividualism and rational choice theory) focus on profit maximization, ensuring market
property rights and reducing transaction costs, and on abstract notions of good gover
nance and institutions. A plausible analytical framework is offered in the (p. 40) develop
mental state literature (which focuses on ideological commitment and the nature of the
ruling elite) and the political settlement framework, sharing common features such as the
balance of power among elites, the features of institutions, and prevailing ideas.52
In the important, broader political and political economy dimension of industrial policy,
power relations, and political variables shape the nature and outcomes of policies (Hall,
1987). Industrial policy shapes market structure and the nature and direction of private-
sector growth. If industrial policy is focused on the transformation of production and posi
tioning in international trade, it is more likely that private-sector manufacturing, exports,
and technological advancement will grow and have greater political influence, which in
versely affects economic policies. The development of production capabilities may slow
down in situations where industrial policy is oriented towards extractive industries or the
finance sector. The politics and political economy of industrial policy is a key variable
shaping its design, execution, and outcomes at both sectoral and national level. Hence,
understanding the internal and external political constraints is essential for the success
of industrial policies.
All economic policies affect the existing political economy, and industrial policymaking is
in its turn shaped by politics. Hall (1987: i, 4) highlights that ‘economic policymaking
must be understood as an essentially political process…a process deeply conditioned by
broader struggles between competing parties, ideologies, and social classes’. He adds
that ‘economic policy is made by governments, and governments are political creatures’.
Even when a consensus is less likely among key interest groups and political forces, they
at least need to accept decisions even if they can’t shape them directly. The cohesion and
historical intellectual formation and shared experiences of the political elite have a signif
icant bearing on the functioning of government and the policymaking process, which has
to be managed by the government through structures coordinated along horizontal and
vertical divisions of power. An agreement among the ruling elites and key influencers is
necessary to hold and impose power to pursue industrial policy as consensus on values
Page 20 of 48
and grand projects may not necessarily be achieved but it acts as a glue enabling the pur
suit of industrial policies.
2.4.3 Disciplining the Private Sector and Reciprocity
All economic policies, including industrial policy, are the outcome of power relations, and
each policy generates losers and winners (Hirschman, 1963, 1968). The stark contrast be
tween the success of industrial policy in East Asia and weak performance in Latin Ameri
ca and Africa has been associated with the political economy variable (p. 41) and state ca
pacity (see, e.g. Kohli, 2004; Chang, 1999; Khan and Blankenburg, 2009). This is also true
at the international level, where positions on environmental sustainability and climate
change exemplify the political constraint, power relations, and interest groups. The con
straints on policy space and the international governance system reflect political and
power relations globally.
A key aspect of the interaction of politics and political economy with industrial policy is
disciplining the private sector; the provision of productive rent or incentives to shape its
behaviour and promote learning is critical for the success of industrial policies. This is a
critical political economy dimension. Incentives, to be productive, have to be carefully de
signed and adjusted to have a positive impact. However, the ability of the government to
apply specific types of incentives depends on the political constraint of the state. Khan
and Blankenburg (2009: 348) highlight that ‘managing rents for technology acquisition is
not just constrained by state capacities, but also and often primarily by political con
straints that prevent specific strategies of rent management from being implemented…
From a policy perspective, potentially growth-enhancing rents can become growth reduc
ing if the rent-management capacities of the state are missing.’ Political ability has a di
rect and indirect bearing on the allocation of ‘development rents’ and this differs across
sectors, different stages of development, and countries. Institutional strength, i.e. the ca
pacity to design and administer rents, is also another dimension that is political. Hence,
political considerations and understanding the political context of each sector is essential.
The dynamism of economic players is a major variable that has an important impact on
policymaking. The design of incentive structures and support schemes requires an under
standing of reciprocity and the dynamics of sectors. Moreover, linking incentives with
performance helps strengthen developmentalism and productive contribution. Induce
ments such as linkages and latitude for performance will also help to leverage activities
in line with the desired political-economy and development outcomes.
History and path dependency play an important role in shaping political economy and in
dustrial policy. Land reform in East Asia significantly changed the political economy, mak
ing it conducive to export-led industrialization, while in Latin America it constrained the
political economy (Kay, 2002; Ocampo, 2019; see Chapter 28). Internal and external
threats, diverse legacies of colonial rule, and cultural traditions also shape the political
economy (Doner, Ritchie, and Slater, 2005).
Page 21 of 48
2.4.4 Conflict and Political Stability
Politics and political economy are significant because policies (like all economic policies)
have conflicting distributional effects across sectors and regions (Hirschman, 2013). The
rules of capitalism include conflict, and the outcomes of industrial policies produce win
ners and losers. Conflicts occur with considerable frequency, in a variety of forms, (p. 42)
and navigating through them calls for constant compromise and bargaining, a process
which society learns to manage. Hirschman (2013: 355‒6) states:
Conflict is indeed a characteristic of pluralist market society that has come to the
fore with remarkable persistence…many conflicts of market economy are over the
distribution of the social product among different classes, sectors, or regions…all
it can aspire to accomplish is to ‘muddle through’ from one conflict to the next.
Polanyi (1944) and Hirschman argue that economic transformation and technical changes
constantly generate new conflicts and social configurations across sectors or regions be
cause of newly emerging inequalities. Such conflicts may involve sectors and regions as
some decline and some rise.
From this perspective, political stability is a much more complex issue than usually pre
sented in ‘business climate’ literature. It could be argued that while efforts may be made
to improve business climate and with it political stability and security, it is important to
recognize the complexity and limitations of this approach. Even if political stability is se
cured, a focus on productive transformation and developing technological capability is
what ultimately matters. Political stability and security, a critical and related issue, is the
minimum but insufficient condition for economic growth and industrial policies. This is
particularly important for long-term investments, productive FDI, and the attraction of
talent. However, while this may be the case in most instances, many industrial policies
and periods of economic growth and technical change have thrived amidst horrible insta
bility and warfare.
2.4.5 Myths Surrounding Industrial Policy
Myths around industrial policy arguably reflect the underlying political and ideological
views of the role of the state in industrial policy. The most common misconception around
industrial policy equates it with merely ‘picking winners’ rather than considering its
wider perspectives. Another myth is the ideologically driven dichotomy between the role
of the state and market forces. There has never been a market reality that has not been
shaped by state intervention, although the type and nature of interventions differ. Eco
nomic history and empirical evidence point to the state’s role in promoting industrializa
tion and structural transformation. Industrial policy is wrongly equated with state inter
vention, while the dominance of the state in a command economy does not necessarily
guarantee an activist industrial policy aiming to build a dynamic private sector.
Page 22 of 48
There is often a dichotomy between export-led and import-substitution industrialization.

In reality, however, they are complementary and industrial policy is not about ISI per se,
as some scholars suggest. Similarly, FDI firms are seen as an enclave focused only on
their own outcomes, while experience shows that FDI could be directed to support the de
velopment of domestic firms. Another myth confuses instruments with the nature of in
dustrial policy. As this chapter has shown, a vigorous industrial policy is (p. 43) based on
structural transformation and economic catch-up perspectives, the recognition of a dy
namic view that emphasizes long-termism, synergy and connectivity, and constant adapta
tion to the changing external and domestic environment and the shifting dynamics of in
dustrial policy.
Wade (2019) articulates clearly the reasons for the failure of neoclassical and neo-liberal
economists to design and apply industrial policies promoting structural transformation
and catch-up. He writes:
Neoclassical economics is a misleading guide to development policy. It thinks in

terms of curves, not step changes. And it makes no distinction between activities
(or sectors) according to their growth potential—it treats each unit of GDP as hav
ing equal potential for future growth. Yet the governments of virtually all the suc
cessful catch-up countries recognized that they had to nurture ‘superior’ or ‘star’
economic activities by means of trade protection, subsidies, and regulations,
buffering them from—but not insulating them from—competitive pressures.
(Wade, 2019: 33)
Industrial policy is as essential for the pursuit of structural transformation and economic
catch-up in the twenty-first century as it was in earlier centuries. Economic history and
empirical evidence from advanced, emerging, and developing economies alike illustrate
this reality. Industrial policies, like all economic policies, are influenced by experience;
learning and building experiences is critical to their success. Furthermore, industrial poli
cy is inseparable from politics and the political economy and forms an integral part of
broader economic and social transformation. Learning from international and local expe
rience needs to be supported by research that vigorously seeks to build knowledge on the
subject, helping to constantly adapt industrial policy based on both lessons and the
changing external environment and domestic reality. It is hoped that this Handbook will
motivate researchers to undertake new studies to help policymakers and practitioners
produce creative policies.
2.5 The Practice of Industrial Policy: Dynamics,

Adaptability, and Principles
A broad and comprehensive understanding of industrial policy has to be tied to structural
transformation and to notions such as protection of infant industry and the state‒market
mechanism. Industrial policies, far from being empty declarations and plans, underlying
Page 23 of 48
purposes and patterns, consist variously of structural change, employment, nationalism,

or defence. Most industrial policies are designed and executed at a national level, though
there are also regional levels (such as the European Union), and sub-national or provin
cial levels.53 Instruments of industrial policy may (p. 44) vary and can involve mid- to long-
term plans (five years and beyond), target setting and performance reviews, associated
legislation (proclamations, bills, regulations), specific policies, executive decisions and di
rectives, support institutions, and the organization of government‒private sector dia
logue.
This section reviews the principles and features of successful industrial policies, with
structural transformation and technological catch-up as key frameworks. It also high
lights the linkage dynamics, ecosystems, and the politics and principles of industrial poli
cy that are needed to produce stronger dynamism, coherence, synergy, and complemen
tarity.
2.5.1 Linkage Dynamics and Interdependence
The pioneering work on linkage effects as a prime tool of industrialization was by

Hirschman. Hirschman ([1992: 74] 2013) states that linkage is a conceptual tool that as
sists in ‘detecting how one thing [activity] leads or fails to lead to another’, and is ‘a more
or less compelling sequence of investment decisions occurring in the course of industrial
ization and, more generally, of economic development’ (Hirschman, [1992: 56] 2013). The
linkage effect is relevant for the selection and support of priority sectors likely to have
the strongest or most lasting effect in accelerating structural change. According to
Oqubay, ‘understanding the industrial structure of different sectors, and leveraging lati
tudes for performance are valuable in industrial policymaking. Hirschman’s linkage con
cept (the favouring of industries with strong linkages…) is essential in bringing dynamics
and impetus to new activities and increasing returns and in creating economic space in
developing economies’ (Oqubay, 2015: 288; own emphasis).
Hirschman (1958) argued—though this was not widely accepted—that the key constraint
in developing countries is not scarcity of resources as such but the inability to make deci
sions that promote investment and induce productive activities. Thus, for latecomer coun
tries, the correct identification of linkage effects, the direction of the linkages and the
various forms of linkages, and changing dynamics are important for industrial policy. Lati
tude for performance is an associated concept that helps to identify and exploit the inter
nal dynamics (especially the technological and economic characteristics) present in the
industrial structures of sectors, projects, and productive activities.54 While linkage effect
is central to the selection of sectors, it is essential that technical change or the potential
for technological advancement and demand formation and elasticity are also considered.
Ocampo, Rada, and Taylor (2009) underline the importance of focusing on ‘innovative ac
tivities that generate domestic spill over’ in new industries, new products, new markets,
new linkages, and new institutions.55
Page 24 of 48
From a linkage effects perspective, the prime focus is inducing investment for de
(p. 45)
velopment, and promoting interdependencies and complementarities. There are three

principal issues. First, the relationship between export-led and import-substitution indus
trialization leads strategic choices and patterns of industrial policies. The increased den
sity of players in the international market should not lead countries to discount the prima
ry role of exports and abandon export-led industrialization. From a production-centric
perspective, relying on exports to spearhead the industrialization drive generates more
dynamism, with growth in productivity gains, larger market size and demand, and a
source of international learning; by generating foreign exchange it also resolves the con
straints hampering the growth of the economy.56 While competition is naturally strongest
in international markets, this choice is a prime guarantee of sustained growth and inter
national competitiveness.57 Complementarity is heightened by ensuring that the export
sector leads all other sectors, including those in the domestic markets, and that domestic
rivalries are fostered.
A second interdependent and complementary relationship is that between manufacturing

and agriculture, especially for late latecomers in the early stages of development. The al
ternatives of manufacturing and agriculture are usually regarded as mutually exclusive,
but a plausible option is to search for an approach that maximizes their synergy and com
plementarity. All developing countries need to focus on boosting agriculture for their ini
tial economic take-off, laying the foundations for accelerated industrialization. Kaldor
highlights that ‘any relative growth of industry in an economy presupposes a correspond
ing growth in the excess of agricultural production…it is not an accident that all coun
tries which succeeded in developing manufacturing industry on a large scale possess a
highly efficient or largely “commercialized” agriculture, with high yields per acre and
high productivity per man’ (Kaldor, 1980: 241).
Kaldor thus argued that agriculture generates the initial demand for the manufacturing
sector in the early stages of industrialization, before penetration of the international mar
ket. However, this may not be the case in all countries; the focus may be on meeting the
demands of other priorities and on manufactured imports. It is important that the techno
logically intensive, high-productivity agricultural sub-sectors with stronger linkage dy
namics should qualify as targeted sub-sectors. Research and study, supported by experi
ments and learning from experience and innovation elsewhere, should determine comple
mentarities. Technological advances (in bio-technology, digitization, economics of fresh
ness, and automation) are shaping and reshaping agricultural practices as we know
them.58
A third important aspect of industrial policy is the relationship between FDI and
(p. 46)
domestic firms, a challenging balance that is often affected by political tension. Policy
choices depend on the international environment, a dynamic domestic private sector, the
nature of the specific industry, and the stage of industrialization. For instance, in the
highly globalized production network, the role of FDI as a source of productive invest
ment has increased immensely. However, building an innovation-driven economy requires
a shift from imitation to innovation, which is driven by domestic capability. Different
Page 25 of 48
countries, for instance Japan, South Korea, Singapore, and Taiwan, China, have followed
different paths, according to their technological capabilities and industrial structure.
2.5.2 Industrial Ecosystem and Cluster Dynamics
Agglomeration economies and cluster dynamics foster specialization and division of

labour, increasing returns and productivity gains, innovation and learning, and linkages,
acting as critical drivers of positive externalities. Internal economy of scale applies to
firms, while external economy of scale applies to agglomeration within the same industry
and across industries (these are termed respectively localized and urbanization agglomer
ation).59 Positive externalities are also associated with productive cities that are centres
of new ideas and learning (Jacobs, 1969, 1984). While the organic formation and evolu
tion of industrial hubs is universal, the deliberate development of industrial hubs that of
fer a learning ecosystem and facilitate linkages is a post-Second World War
phenomenon.60
Developing dynamic industrial hubs that move beyond static benefits to become dynamic
and organic evolving entities needs to be an integral part of the industrial policy frame
work (Best, 2001). Breschi and Malerba suggest that ‘clusters should not be examined on
ly as a static framework and at a given point in time. Rather, clusters have specific stages
of development and transformation’ (Breschi and Malerba, 2005: 22).61 The dynamics of
these industrial hubs are shaped by a culture of learning and openness, and open systems
involving a network of institutions such as universities, research centres, government
bodies, standards authorities, and intermediary institutions.
Empirical evidence shows that the cluster dynamics of industrial hubs promote learning
and innovation, linkages, productivity gains, increasing returns to scale, and (p. 47) espe
cially, technological and innovation capabilities. If built on an open culture, they will pro
mote knowledge exchanges and combine collective efficiency, cooperation, and competi
tion.62
Effective policies for cluster development include talent attraction and brain gain, a net
work of organizations, the attraction of new start-ups and enhanced spin-outs.63 Different
strategies are required during the initial stage and the growth and maturity stages.64
Best (2001: 87) highlights that:
A static industrial district lacks the internal/external growth dynamic. Such dis
tricts can enjoy Marshall’s locational economies, but they do so within an un
changing ‘production chain’. Dynamic districts, in contrast, are continuously up
grading, redefining, and reconstituting the production chain. Most of the world’s
industry is conducted within static industrial districts that lack entrepreneurial
firms. Capabilities are not advancing and innovation is limited. They do not drive
growth.65
Page 26 of 48
2.5.3 Tensions in Industrial Policy
Different contexts (national or sectoral, or different stages) call for various industrial poli
cy instruments to serve different goals and strategies. Despite their varied content, inher
ent tensions are evident during the design and execution stage which impact on the out
come of industrial policy. These tensions are exhibited in multiple ways.
2.5.3.1 Strategic Compatibility and Coherence

The overarching principle in the design and application of industrial policy is to choose
instruments that serve the country’s grand vision, purpose, goals, and development strat
egy. Key to this decision is the targeting of priority sectors and new productive (p. 48) ac
tivities. Another aspect is engaging firms and dynamic social groups that have a high
stake in the desired outcomes. Focus on the development of productive capacity, techno
logical and domestic capability, and learning is critical to ensure that industrial policy
serves as a conduit of transformation and catch-up. However, ensuring compatibility be
tween the industrial goals, and picking and nurturing winners, or even establishing state-
owned enterprises and transfers to the private sector, are in constant tension.
Dynamic shifts in productivity require both learning by doing and ‘true’ innovations,
which as key drivers of technological learning are highly complementary. Solow under
lines that:
Learning by doing would soon exhaust itself were it not for the intermittent occur
rence of major innovations…The overall productivity trend will represent both
true innovation and learning by doing. Innovations will bulk larger the faster they
come and the larger they are. Learning by doing will bulk larger the more rapid
the pace of investment and the greater the intensity of learning.
(Solow, 1997: 20, 33)
In the long term, what matters is not the design and content of a single instrument. In al
most all situations, multiple instruments are needed for full and sustained impact, so co
herence between them as well as between short-term and long-term directions is critical
to achieving maximum benefits. However, coherence is not automatic and must be
achieved through appropriate planning and design. Aligning plans to policy decisions,
proclamations, regulations, and other executive directives and implementing institutions
is a difficult exercise and is full of tensions. A major constraint on policy outcomes is in
consistency of policy execution, which can limit both effectiveness and learning opportu
nities. Inconsistencies across firms, time, regions, and sectors creates frustration and un
dermines effectiveness. Gaps should therefore be consistently identified to improve policy
implementation. Enhancing coherence and consistency can be targeted during the de
sign, execution, and review stages of the policy. Independent studies, continuous re
search, and curiosity are important for continuous improvement. However, one of the pit
falls is that many policymakers and practitioners perceive instruments as static.
Page 27 of 48
2.5.3.2 Synergy and Connections to Other Policies

As we have seen, industrial policy clearly does not exist in a vacuum; it is related to other
economic policies and the general political economy. Industrial policies will only have a
partial effect and indeed are likely to fail in the long term if complementarity is not maxi
mized. Of particular importance are policies that impact the transformation of productive
capacity, export and domestic capacity, and the development of technological capability,
and those associated with building human capital and the education system, the develop
ment of physical infrastructure, and ensuring macroeconomic stability.
Human capital and education have a direct bearing on the vigour and outcome of
(p. 49)
industrial policy. Empirical evidence and research show that education is a necessary con
dition for structural transformation and industrial catch-up. Its importance increases as
the economy progresses towards the advanced stage when technological capability and
innovation become more critical. However, experience elsewhere shows that education
alone does not necessarily guarantee a breakthrough in the economic catch-up of late in
dustrializers. At an early stage of development, access to general education is important
to promote agriculture and shared growth. However, as industrialization advances, tech
nical schools, technical and vocational education centres, and universities become more
important. Skills formation, and engineering- and technology-driven courses become
higher priorities. In an innovation-driven economy, links between research centres, uni
versities, and industry become important. Education should focus not only on producing
high-quality technicians and technologists, but also on nurturing the talent that will be
needed for innovation. The tension inherent in this connection is exacerbated by the num
ber of players—multiple organizations, political interests, and interest groups.
Another critical policy dimension that has a direct bearing on the outcomes of industrial
policies is the development of physical infrastructure. Infrastructure, including reliable
and competitive energy and electricity supplies, transport and logistics, is the most im
portant factor in the production process and in downstream and upstream activities, and
it is relevant to both quality and cost competitiveness. Infrastructure requirements
change during the different stages of development. At an early stage, road connectivity
(especially rural and regional) and universal access to electricity, water, and telecommu
nications will be mandatory. At a later stage, high-speed road networks and high-produc
tivity infrastructure such as large power supply, railways, and airports become necessary.
Infrastructure development is capital intensive and requires not only long-term planning
but also economic viability and full utilization, building international competitiveness and
capacity in managing and maintaining infrastructure facilities.
Equally important for the attraction of investment, performance, and viability of exports,
labour costs, and productivity, is the presence of a stable and conducive macro economy,
including favourable interest rates and sound monetary and fiscal policies. The connec
tion of industrial policy and macro-policies is essential as macro-policies should be con
ducive to industrial policy and support the production-centric pathway and the produc
tion transformation. This implies arguably not only creating macroeconomic stability but
also developing support for the production transformation in terms of exchange rates,
Page 28 of 48
taxation, fiscal policy, public investment, labour market institutions, and income distribu
tion (Cimoli, Dosi, and Stiglitz, 2009). Imbalances and tensions are the basic feature of
this relationship and the aim should not be to create harmony but to use the imbalances
and tension to promote enduring economic transformation. Macroeconomic policy deci
sions should also aim to maximize synergy with industrial policy, ensuring that exchange
rates do not undermine exports, that domestic savings and investment are conducive to
the expansion of (p. 50) production capability, inflation is controlled so that wages are not
undermined, and that the tax regime is directly linked to supporting production capability
and innovation.66
2.5.3.3 Competitiveness and Discipline

A vital principle of industrial policy is the support offered to industries and firms, and an
incentive structure linked to performance standards and enhancing productivity, exports,
learning, and innovation. One of the most important lessons from the East Asian experi
ences was the linking of incentives and support to performance and the control of hand-
outs and unproductive rent. Amsden highlights that ‘the guiding principle of the best bu
reaucracies—politics permitting—was to give nothing away for free. Reciprocity was ideal
…The reciprocity principle in Korea operated in almost every industry’ Amsden (2007: 94,
96).67 Outcomes are far from automatic if the principle of reciprocity is not strictly ad
hered to; its successful implementation depends above all on the state’s political support,
a social base that ensures enforcement, and whether the government has the political
clout and capability to insulate itself from interest groups. In addition, designing and
managing incentives and support schemes is a complex process that requires careful se
lection, together with an effective bureaucracy, and continuous evaluation of its purpose
and shortcomings during the implementation process.
Instruments such as export discipline (linking incentives to exports to discipline firms and
enhance international learning) and domestic rivalry (to ensure that incentives are tied to
intense competitiveness) may also be beneficial. Devices such as latitude to performance
are important; the system has to be measurable and execution as simple as possible. The
nature of these incentives should be measured against international practices and policy
space, and constantly renewed based on learning from others. These are critical decision
areas that have strategic implications. The incentive structure should differentiate be
tween big businesses, small businesses, and technology start-ups, and should focus on
sectors with the strongest linkage effects. It should also be based on the principle of ‘cre
ative destruction’ to ensure that the death of decaying firms is accelerated and new start-
ups and spin-outs that generate new energy are supported.
The point is that ensuring complementarity, coherence, and compatibility across policies
and within industrial policy is a difficult exercise which is constrained by institutional ca
pability and political constraints. Ensuring greater coherence at planning and during im
plementation stages may not be sufficient as these tensions and imbalances are constant.
Arguably, thriving by creating tensions and imbalances is an alternative but complemen
tary approach to produce better outcomes: ‘the view that (p. 51) economies should devel
op by prioritizing key sectors rather than relying on comprehensive, overarching
Page 29 of 48
plans’ (Hirschman, 2013: 103; see also Oqubay, 2015; Cramer, Sender, and Oqubay,
2020).68
2.5.4 Summary
The practice of industrial policy requires:
a) Effective policy instruments which are relatively easy to design, measure, and
adapt at sectoral level. This helps to link instruments with the sector’s life cycle and
with upgrading and deepening of the industrial structure, and to create an ecosys
tem that offers growth opportunities and continuous learning.
b) Not all instruments require the same level of execution capacity. For instance, de
valuation is important for promoting exports, but it does not require sophisticated
executive capacity at sectoral level. Nonetheless, some export incentives (such as the
‘voucher system’) may require higher professional bureaucracy. Understanding this
variation is important for choosing and designing policy instruments.
c) The purpose of these instruments is to promote productive investment, production
capability, domestic capability, export performance, and linkage effects and innova
tion capability. It needs a balancing act so that not only productive investment, but
more importantly international and local talents are targeted, accelerating creative
destruction in terms of firm dynamism, and balancing the attraction of FDI against
domestic firms, and big business against small firms.
2.6 Concluding Remarks: Observations and

Emerging Issues
Industrial policy has not been static, and successful industrial policies have constantly
adapted to the changing external environment and local conditions. Empirical evidence
shows that industrial policy and outcomes are uneven across sectors and are primarily
shaped by variations in the industrial structure, politics, and linkage effects of sectors
(Oqubay, 2015: 277, 284).69 However, three key factors carrying significant implications
for policymakers emerge as drivers of unevenness and variation in (p. 52) industrial poli
cies: the industrial structure of each sector; politics and the political economy; and link
age effects.
2.6.1 The Middle-income Trap
More recently, new debates have emerged focusing on economic catch-up, and particular
ly on the difficulty and rare instances of climbing the development ladder from middle to
high income. While there is less agreement on the ‘middle-income trap’ concept, increas
ing evidence shows that the vitality of technological capability and innovation are critical
for avoiding it. This may be partly explained by Abramovitz’s hypothesis that catch-up
slows as latecomers converge with the productivity levels of the leaders or forerunners
Page 30 of 48
(Abramovitz, 1994). Accelerated catch-up, dynamic transformation to an innovation-dri

ven economic path, stalled industrialization, where technological catch-up has become
impossible, and premature de-industrialization, where manufacturing has run out of
steam for creating jobs, output, and export earnings (UNCTAD, 2016), can all be imag
ined. As China emerges as the world’s manufacturing and exports powerhouse, its shift to
an innovation-driven economy, combined with its sheer size, continues to impact the in
tensity of global competition, and the struggle to escape the ‘middle-income trap’ (Paus,
2019).
2.6.2 Changing Patterns of Global Production and World Economy
A major shift in the last five decades has been the internationalization of production, the
rise of global value chains and global production networks, and the dominance of the
global business revolution (Gereffi, 2018; Dicken, 1990; Coe and Yeung, 2015; Nolan,
2014). Major shifts in international governance systems that have had an uneven impact
on developing, emerging, and advanced economies and on the policy space for designing
industrial policy include the introduction of binding multilateral trade rules under the
WTO, tighter protection of intellectual property through the TRIPS agreement, a Sino-
centric global economy, and the post-2008 global economic slowdown. Least developed
economies face intensified competition to exploit limited opportunities. Wade character
izes this process as follows: ‘the combination of global value chains, knowledge monopoly,
and financialization makes for slow or no long-run convergence upwards of the majority
of emerging and developing economies’ (2019: 33).70
Since the 2008 economic crisis three tendencies have surfaced which reflect the
(p. 53)
contemporary global political economy. First, emerging economic powers and significant
players in international markets have pursued a pro-WTO trading system advocating free
trade and shrinking protection. This view is in line with multilateralism and the multi-po
lar global power structure. Second, however, there has also been a backlash against this
trend with a preference for bilateral trade negotiations and a withdrawal from regional
economic blocks. Third is increased integration of regional markets, such as ASEAN in
the Asia Pacific, an inclusive initiative related to the Trans Pacific Partnership.71
2.6.3 Accelerated Technological Advancement
The pace of technological change has accelerated the emergence and fusion of new tech
nologies (such as ICT, artificial intelligence, 5G technology, automation, robotization, and
digitization) that are shaping and shaking each industry in unexpected ways. While what
is called the Fourth Industrial Revolution, or the digital economy, offers many opportuni
ties, the unrestrained power these technologies offer to leading technology MNCs under
mines the roles of government and the rights of citizens.72 ‘Technological romanticizing’
or ‘technology hype’ advocates technological determinism, ignoring the fact that tech
nologies should serve the well-being of society and the indispensable role of social trans
formation.
Page 31 of 48
Technological advances do not shape all sectors uniformly, and understanding their scope
at economy and sectoral level is critical. The implications for the development of human
capital and the expansion of smart infrastructure, financial services, and R&D are partic
ularly significant. Industrial policies should take this fundamental shift into account and
provide effective mitigation and adjustment in response to these challenges.
2.6.4 Sustainability and Green Industrial Policy
The early industrial revolution and twentieth-century industrialization were characterized

by neglect of climate change and the critical role of environmental sustainability (Math
ews, 2016a, 2017).73 The position taken by the corporate world has varied according to
economic interests and economic power. Increased awareness on the part of consumers
has still not been sufficient to pressure corporations and industries to refrain from dam
aging the environment. One perspective to emerge is that economic (p. 54) growth and
environmental sustainability are not mutually exclusive. Sustainable development re
quires environmental protection to be at the centre of economic policies. It involves edu
cation, a regulatory regime that does not reward polluters, industrial policy that rewards
sustainability and resource savings, and increased innovation in environmental science
for job creation and economic growth. From a long-term perspective, integrating environ
mental sustainability and putting it at the centre of industrial policy is a necessity.74
Acknowledgements
The author is grateful to Christopher Cramer, John Mathews, José A. Ocampo, and Taferre
Tesfachew for their constructive comments and discussion, and inputs from the reviews
workshop. The author thanks Deborah Kefale, Samuel Arkebe, and Binyam Arkebe for in
puts to improve the draft and for their continued support.
References
Abramovitz, Moses (1989) ‘Catching up, Forging ahead, and Falling behind’, in Moses
Abramovitz, Thinking About Growth: And Other Essays on Economic Growth and Welfare.
Cambridge: Cambridge University Press, pp. 220–44.
Abramovitz, Moses (1994) ‘Catch-up and Convergence in the Growth Boom and After’, in
William Baumol, Richard Nelson, and Edward Wolf (eds), Convergence of Productivity:
Cross-national Studies and Historical Evidence. Oxford: Oxford University Press, pp. 86–
128.
Akyűz, Yilmaz (2017) Playing with Fire: Deepened Financial Integration and Changing
Vulnerabilities of the Global South. Oxford: Oxford University Press.
Amsden, Alice (1989) Asia’s Next Giant: South Korea and Late Industrialization. Oxford:
Oxford University Press.
Page 32 of 48
Amsden, Alice (2001) The ‘Rise of the Rest’: Challenges to the West from Late-Industrial
izing Economies. Oxford: Oxford University Press.
Amsden, Alice (2007) Escape from Empire: The Developing World’s Journey through
Heaven and Hell. Cambridge, MA: MIT Press.
Amsden, Alice (2012) ‘Elites and Property Rights’, in Alice Amsden, Alisa DiCaprio, and
James A. Robinson (eds) The Role of Elites in Economic Development. Oxford: Oxford Uni
versity Press, pp. 19–28.
Amsden, Alice and Wan-wen Chu (2003) Beyond Late Development: Taiwan’s Upgrading
Policies. Cambridge, MA: MIT Press.
Andreoni, Antonio and Ha-Joon Chang (2017) ‘Bringing Production and Employment back
into Development: Alice Amsden’s Legacy for a New Developmentalist Agenda’, Cam
bridge Journal of Regions, Economy and Society 10(1): 173–87. (p. 55)
Andreoni, Antonio and Ha-Joon Chang (2019) ‘The Political Economy of Industrial Policy:
Structural Interdependencies, Policy Alignment and Conflict Management’, Structural
Change and Economic Dynamics 48: 136–50.
Arrow, Kenneth (1962) ‘The Economic Implications of Learning by Doing’, The Review of
Economic Studies 29(3): 155–73.
Atkinson, Robert, and Stephen Ezel (2012) Innovation Economics: The Role for Global Ad
vantage. New Haven, CT: Yale University Press.
Babbage, C. (2005 [1832]) On the Economy of Machinery and Manufactures. Cirencester:

Echo Library.
Best, Michael (2001) The New Comparative Advantage. Oxford: Oxford University Press.
Best, Michael (2018) How Growth Really Happens: The Making of Economic Miracles
Breschi, Stefano and Franco Malerba (eds) (2005) Clusters, Networks, and Innovation.
Oxford: Oxford University Press.
Chandler, Alfred (2004) Scale and Scope: The Dynamics of Industrial Capitalism. Cam
bridge MA: Harvard University Press.
Chang, Ha-Joon (1994) The Political Economy of Industrial Policy. London: Macmillan.
Chang, Ha-Joon (1999) ‘The Economic Theory of the Developmental State’, in Meredith
Woo-Cummings (ed.) The Developmental State. Ithaca, NY: Cornell University Press, pp.
182–99.
Page 33 of 48
Cherif, Reda, and Fuad Hasanov (2019) ‘The Return of the Policy That Shall Not Be
Named: Principles of Industrial Policy’. IMF Working Paper No. WP/19/74. March.
Chung, Moon Y. and Keun Lee (2015) ‘How Absorptive Capacity is Formed in a Latecomer
Economy: Different Roles of Foreign Patent and Know-how Licensing in Korea’, World De
velopment 66: 678–94.
Cimoli, Mario, Giovanni Dosi, and Joseph Stiglitz (2009) The Political Economy of Industri
al Policy and Development. Oxford: Oxford University Press.
Coe, Neil and Henry Yeung (2015) Global Production Networks: Theorizing Economic De
velopment in an International World. 1st edition. Oxford: Oxford University Press.
Cohen, Stephen, and Bradford DeLong (2016) Concrete Economics: The Hamilton Ap
proach to Economic Growth and Policy. Boston, MA: Harvard Business Review Press.
Cohen, Wesley and Daniel Levinthal (1990) ‘Absorptive Capacity: A New Perspective on
Learning and Innovation’, Administrative Science Quarterly 35(1). Special Issue: Technol
ogy, Organizations, and Innovations, pp. 128–52.
Cramer, Christopher and John Sender (2019) ‘Oranges Are Not Only Fruit: The Industrial
ization of Freshness and the Quality of Growth’, in Ravi Kanbur, Akbar Noman, and
Joseph Stiglitz (eds) The Quality of Growth in Africa. New York: Columbia University
Press, pp. 209–33.
icy and the Implications for Industrial Hubs’, in Arkebe Oqubay and Justin Yifu Lin (2020)
The Oxford Handbook of Industrial Hubs and Economic Development. Oxford: Oxford
University Press.
Cramer, Christopher, John Sender, and Arkebe Oqubay (2020) African Economic Develop
ment: Evidence, Theory, Policy. Oxford: Oxford University Press. (p. 56)
Dicken, Peter (1990) Global Shift: Reshaping the Global Economic Map in the 21st Centu
ry. London: Sage.
Doner, Richard, Bryan Ritchie, and Dan Slater (2005) ‘Systematic Vulnerability and Ori
gins of Developmental States: Northeast and Southeast Asia in Comparative Perspective,
International Organization 59(2): 327–61.
Dosi, Giovanni and Richard Nelson (2018) ‘Technical Change and Industrial Dynamics as
Evolutionary Processes’, in Bronwyn Halland and Nathan Rosenberg (eds) Economics of
Innovation, Vol. I. Burlington, VI: Academic Press, pp. 51–128.
Drucker, Peter F. (1987) ‘What We Learn from Germany’, in Peter Drucker, The Frontiers
of Management. Oxford: Butterworth Heinemann, pp. 81–115.
Page 34 of 48
Evans, Peter (1997) Embedded Autonomy: States and Industrial Transformation. Prince
ton, NJ: Princeton University Press.
Gereffi, Gary (2018) Global Value Chains and Development. Cambridge: Cambridge Uni
versity Press.
Gerschenkron, Alexander (1962) Economic Backwardness in Historical Perspective. Cam

bridge, MA: Harvard University Press.
Goodrich, Carter (1965) The Government and the Economy, 1783–1861. New York: Bobbs-
Merrill.
Gray, Hazel (2018) Turbulence and Order in Economic Development: Institutions and Eco
nomic Transformation in Tanzania and Vietnam. Oxford: Oxford University Press.
Haggard, Stephan (2018) Developmental States. Cambridge: Cambridge University Press.
Hall, Peter (1987) Governing the Economy: The Politics of State Intervention in Britain
and France. Cambridge: Polity Press.
Hall, Peter, and David Soskice (2001) Varieties of Capitalism: The Institutional Founda
tions of Comparative Advantage. Oxford: Oxford University Press.
Hamilton, Alexander (1934) Papers on Public Credit, Commerce and Finance, ed. Samuel
McKee. New York: Columbia University Press.
Hirschman, Albert (1958) The Strategy of Economic Development. New Haven, CT: Yale
University Press.
Hirschman, Albert O. (1963) Journeys toward Progress: Studies of Economic Policy-mak

ing in Latin America. New York: Twentieth Century Fund.
Hirschman, Albert O. (1967) Development Projects Observed. Washington, DC: The

Brookings Institute.
Hirschman, Albert O. (1968) ‘The Political Economy of Import-substitution Industrializa

tion in Latin America’, Quarterly Journal of Economics 82(1): 2–32.
Hirschman, Albert ([1992] 2013) ‘Social Conflicts as Pillars of Democratic Market Soci
ety’, in Jeremy Adelman (ed.) Essential Hirschman. Princeton, NJ: Princeton University
Press, pp. 345–62.
Hirschman, Albert (2013) ‘The Gaucher-Dubeil Thesis: Conflict as the Builder of Commu
nity’, in Jeremy Adelman, The Essential Hirschman, Princeton, NJ: Princeton University
Press.
Hobday, Michael (2000) ‘East versus South East Asian Innovation Systems’, in Linsu Kim
and Richard Nelson (eds) Technology, Learning, and Innovation: Experiences of Newly In
dustrializing Economies. Cambridge: Cambridge University Press, pp. 335–58.
Page 35 of 48
Hymer, Stephen H. (1976) The International Operations of National Firms: A Study of Di
rect Foreign Investment. Cambridge, MA: The MIT Press.
Jacobs, Jane (1969) The Economy of Cities. New York: Vintage Books. (p. 57)
Jacobs, Jane (1984) Cities and the Wealth of Nations: Principles of Economic Life. New
York: Vintage Books.
cy, 1925–1975. Stanford, CA: Stanford University Press.
Kaldor, Nicholas (1967) Strategic Factors in Economic Development. Ithaca, NY: New
York State School of Industrial and Labour Relations, Cornell University.
Kaldor, Nicholas (1980) Essays on Economic Stability and Growth. 2nd edition. New York:
Holmes & Meier Publishers.
Kay, Cristobal (2002) ‘Why East Asia Overtook Latin America: Agrarian Reform, Industri
alisation and Development’, Third World Quarterly 23(6): 1073–102.
Khan, Mushtaq H. (2010) ‘Political Settlements and the Governance of Growth-Enhancing

Institutions’. Available at https://eprints.soas.ac.uk/9968/1/
Political_Settlements_internet.pdf.
Khan, Mushtaq H. and Stephanie Blankenburg (2009) ‘The Political Economy of Industrial
Policy in Asia and Latin America’, in Mario Cimoli, Giovanni Dosi, and Joseph E. Stiglitz
(eds) Industrial Policy and Development: The Political Economy of Capabilities Accumula
tion. Oxford: Oxford University Press, pp. 336–77.
Kim, Linsu (2000) ‘Korea’s National Innovation System in Transition’, in Linsu Kim and
Richard Nelson (eds) Technology, Learning, and Innovation: Experiences of Newly Indus
trializing Economies. Cambridge: Cambridge University Press, pp. 335–58.
Kim, Linsu and Richard R. Nelson (eds) (2000) Technology, Learning and Innovation: Ex
periences of Newly Industrializing Economies. Cambridge: Cambridge University Press.
Kohli, Atul (2004) State-Directed Development: Political Power and Industrialization in

the Global Periphery. Cambridge: Cambridge University Press.
Komiya, Ryutaro, Masahiro Okuno, and Kotaro Suzumura (eds) (1988) Industrial Policy of
Japan. London: Academic Press.
Kuznets, Simon (1966) Modern Economic Growth: Rate, Structure and Spread. New
Haven, CT: Yale University Press.
Kuznets, Simon (1973) ‘Modern Economic Growth: Findings and Reflections’, American
Economic Review 63: 247–58.
Page 36 of 48
Lee, Keun (2013) Schumpeterian Analysis of Economic Catch-up: Knowledge, Path-Cre

ation, and the Middle-income Trap. Cambridge: Cambridge University Press.
Lee, Keun (2019a) ‘The Origin of Absorptive Capacity in Korea: How Korean Industry
Learnt’, in Arkebe Oqubay and Kenichi Ohno (ed.) How Nations Learn: Technological
Learning, Industrial Policy, and Catch-up. Oxford: Oxford University Press, pp. 125–48.
Lee, Keun (2019b) The Art of Economic Catch-up: Barriers, Detours and Leapfrogging in
Lee, Keun and Franco Malerba (2018) ‘Economic Catch-up by Latecomers as an Evolu
tionary Process’, in Richard Nelson (ed.) Modern Evolutionary Economics: An Overview.
Cambridge: Cambridge University Press.
Lin, Justin Yifu (2016) ‘The Latecomer Advantages and Disadvantages: A New Structural
Economics Perspective’, in Martin Andersson and Tobias Axelsson (eds) Diverse Develop
ment Paths and Structural Transformation in Escape from Poverty. Oxford: Oxford Uni
tries Conform to Comparative Advantage or Defy It? A Debate between Justin Lin and Ha-
Joon Chang’, Development Policy Review 27(5): 483–502. (p. 58)
Lin, Justin Yifu and Volker Treichel (2012) ‘Learning from China’s Rise to Escape the Mid
dle-income Trap: A New Structural Economics Approach to Latin America’. Policy Re
search Working Paper No. 6165. Washington, DC: World Bank.
List, Friedrich (1856) National System of Political Economy, Vols I–IV. Memphis, TN: Lip
pincott.
Marshall, Alfred (1920) Principles of Economics. London: Palgrave Macmillan.
Marx, Karl ([1867] 1990) Capital: A Critique of Political Economy, Volume I. London: Pen
guin Classics.
Mathews, John (2016a) Greening of Capitalism: How Asia is Driving the Next Great Trans
formation. Stanford, CA: Stanford University Press.
Mathews, John (2016b) ‘Latecomer Industrialization’, in Erik S. Reinert, Jayati Gosh, and
Rainer Kattel (eds) Handbook of Alternative Theories of Economic Development. Chel
tenham: Edward Elgar, pp. 613–36.
Mathews, John (2017) Global Green Shift: When CERES Meet GAIA. London: Anthem
Press.
tor Myths. London: Anthem Press.
Page 37 of 48
Mazzucato, Mariana (2017) The Value of Everything: Making and Taking in the Global
Economy. London: Allen Lane.
Mowery, David C. and Nathan Rosenberg (1989) Technology and the Pursuit of Economic
Growth. Cambridge: Cambridge University Press.
Mytelka, Lynn (2000) ‘Local Systems of Innovation in a Globalized World Economy’, In
dustry and Innovation 7(1): 33–54.
Mytelka, Lynn and Taffere Tesfachew (1998) ‘The Role of Policy in Promoting Enterprise
Learning during Early Industrialization: Lessons for African Countries’. UNCTAD/GDS/
MDPB/Misc.7. Geneva: UNCTAD.
Nayyar, Deepak (2016) Catch Up: Developing Countries in the World Economy. Oxford:
Nayyar, Deepak (2019) ‘Foreword’, in Arkebe Oqubay and Kenichi Ohno (eds) How Na
tions Learn: Technological Learning, Industrial Policy, and Catch-up. Oxford: Oxford Uni
versity Press, pp. vi–xi.
Nelson, Richard and Sidney Winter (1982) An Evolutionary Theory of Economic Change.
Nolan, Peter (2014) Chinese Firms, Global Firms: Industrial Policy in the Age of Globaliza
tion. New York: Routledge.
Ocampo, José Antonio (2017) ‘Dynamic Efficiency: Structural Dynamics and Structural
Change in Developing Countries’, in Akbar Noman and Joseph E. Stiglitz (eds) Efficiency,
Finance, and Varieties of Industrial Policy. New York: Columbia University Press, pp. 65–
102.
Ocampo, José Antonio (2019) ‘The Transformation of Latin American Economies’, in

Célestin Monga and Justin Yifu Lin (eds) The Oxford Handbook of Structural Transforma
Ocampo, José Antonio, Cordina Rada, and Lance Taylor (2009) Growth and Policy in De
veloping Countries: A Structuralist Approach. New York: Columbia University Press.
OECD (2005) Guidelines for Collecting and Interpreting Innovation Data: Oslo Manual.
OECD and Eurostat.
Ohlin, Bertil (1933) Interregional and International Trade. Cambridge: Cambridge Univer
sity Press. (p. 59)
Ohno, Kenichi (2019) ‘Meiji Japan: Progressive Learning of Western Technology’, in

Arkebe Oqubay and Kenichi Ohno (eds) How Nations Learn: Technological Learning, In
dustrial Policy, and Catch-up. Oxford: Oxford University Press, pp. 85–106.
Page 38 of 48
Oqubay, Arkebe (2015) Made in Africa: Industrial Policy in Ethiopia. Oxford: Oxford Uni
versity Press.
Oqubay, Arkebe (2019a) ‘Ethiopia: Lessons from an Experiment’, in Célestin Monga and
Justin Yifu Lin (eds) The Oxford Handbook of Structural Transformation. Oxford: Oxford
University Press, pp. 591–618.
Oqubay, Arkebe (2019b) ‘Industrial Policy and Late Industrialization’, in Fantu Cheru,
Christopher Cramer, and Arkebe Oqubay (eds) The Oxford Handbook of the Ethiopian
Economy. Oxford: Oxford University Press, pp. 505–629.
Oqubay, Arkebe and Justin Yifu Lin (eds) (2020) The Oxford Handbook of Industrial Hubs
and Economic Development. Oxford: Oxford University Press.
Oqubay, Arkebe and Kenichi Ohno (eds) (2019) How Nations Learn: Technological Learn
ing, Industrial Policy, and Catch-up. Oxford: Oxford University Press.
Pasinetti, Luigi (1981) Structural Change and Economic Growth: A Theoretical Essay on
the Dynamics of the Wealth of Nations. Cambridge: Cambridge University Press.
Pasinetti, Luigi (1993) Structural Economic Dynamics: A Theory of the Economic Conse
quences of Human Learning. Cambridge: Cambridge University Press.
Paus, Eva (ed.) (2018) Confrontation Dystopia: The New Technological Revolution and the
Future of Work. Ithaca, NY: Cornell University Press.
Paus, Eva (2019) ‘Innovation Strategies Matter: Latin America’s Middle-income Trap
Meets China and Globalisation’, The Journal of Development Studies 56(4): 657–79.
Peerenboom, Randall (2001) ‘Seek Truth from Facts: An Empirical Study of Enforcement
of Arbitral Awards in the PRC’, The American Journal of Comparative Law 49(2): 249–327.
Polanyi, Karl ([1944] 2001) The Great Transformation: The Political and Economic Origins
of our Time. Boston, MA: Beacon Press.
Porter, Michael (1990) The Competitive Advantage of Nations. New York: Palgrave
Macmillan.
Prebisch, Raúl (1950) The Economic Development of Latin America and its Principal
Problems. New York: UN Department of Economic Affairs.
Reinert, Erik (2008) How Rich Countries Get Rich and Why Poor Countries Stay Poor.
London: Constable.
Reinert, Erik, Jayati Gosh, and Rainer Kattel (eds) (2016) Handbook of Alternative Theo
ries of Economic Development. Cheltenham: Edward Elgar.
Rosenberg, Nathan (1994) Exploring the Black Box: Technology, Economics and History.
Page 39 of 48
Schumpeter, Joseph ([1934] 2017) The Theory of Economic Development. New York:
Routledge.
Schumpeter, Joseph (1950) Capitalism, Socialism, and Democracy. 3rd edition. New York:
Harper and Brothers.
Schwartz, Herman (2010) States versus Markets: The Emergence of a Global Economy.
3rd edition. New York: Palgrave.
Serra, Narcis and Joseph E. Stiglitz (2008) The Washington Consensus: Towards a New
Global Governance. Oxford: Oxford University Press.
Smith, Adam ([1776] 1976) An Inquiry into the Nature and Causes of the Wealth of Na
tions, Vols I and II. London: W. Strahan and T. Cadell. (p. 60)
Solow, Robert (1997) Learning from “Learning by Doing”: Lessons for Economic Growth.
Stanford, CA: Stanford University Press.
Thirlwall, Anthony (2002) The Nature of Economic Growth: An Alternative Framework for
Understanding the Performance of Nations. Cheltenham: Edward Elgar.
Thirlwall, Anthony (2011) ‘Balance of Payments Constrained Growth Models: History and
Overview’, PSL Quarterly Review 64(259): 307–51.
Thirlwall, Anthony (2013) Economic Growth in an Open Developing Economy: The Role of
Structure and Demand. Cheltenham: Edward Elgar.
Thurbon, Elizabeth and Linda Weiss (2016) ‘The Developmental State in the Late Twenti
eth Century’, in Erik Reinert, Jayati Gosh, and Rainer Kattel (eds) Handbook of Alterna
tive Theories of Economic Development. Cheltenham: Edward Elgar, pp. 637–50.
Tregenna, F. (2009) ‘Characterising Deindustrialisation: An Analysis of Changes in Manu

facturing Employment and Output Internationally’, Cambridge Journal of Economics
33(3): 433–66.
Tregenna, Fiona (2013) ‘Deindustrialization and Reindustrialization’, in Adam Szirmai,

Wim Naudé, and Ludvico Alcorta (eds) Pathways to Industrialization in the Twenty-First
Century: New Challenges and Emerging Paradigms. Oxford: Oxford University Press, pp.
76–101.
UNCTAD (2006) Developing Productive Capacities. The Least Developed Countries Re
port 2006. Geneva: United Nations.
UNCTAD (2013) Global Value Chains: Investment and Trade for Development. World In
vestment Report. Geneva: United Nations.
UNCTAD (2016) Structural Transformation for Inclusive and Sustained Growth. Trade
and Development Report. Geneva: United Nations.
Page 40 of 48
UNCTAD (2019) World Investment Report. Geneva: UNCTAD.
UNCTAD-UNIDO (2011) Fostering Industrial Development in Africa in the New Global En
vironment. Economic Development in Africa Report 2011. Geneva: UNIDO and UNCTAD.
UNECA (2016) Greening Africa’s Industrialization. Economic Report on Africa. Addis Aba
ba: United Nations Economic Commission for Africa.
Veblen, Thornstein ([1915] 2016) Imperial Germany and the Industrial Revolution. Hard
Press, Classic Series. New York: The Macmillan Company.
Wade, Robert (2004) Governing the Market: Economic Theory and the Role of the Govern
Wade, Robert (2018) ‘The Developmental State: Dead or Alive’, Development and Change
49(2): 518–46.
Wade, Robert (2019) ‘Catch-up and Constraints in the Twentieth and Twenty-first Cen
turies’, in Arkebe Oqubay and Kenichi Ohno (eds) How Nations Learn: Technological
Learning, Industrial Policy, and Catch-up. Oxford: Oxford University Press, pp. 15–37.
Weiss, John (2013) ‘Industrial Policy in the Twenty-first Century: Challenges for the Fu
ture’, in Adam Szirmai, Wim Naudé, and Ludovico Alcorta (eds) Pathways to Industrializa
tion in the Twenty-first Century: New Challenges and Emerging Paradigms. Oxford: Ox
ford University Press, pp. 393–414.
Woo-Cummings, Meredith (ed.) (1999) The Developmental State. Ithaca, NY: Cornell Uni
versity Press.
World Bank (2012) China 2030: Building a Modern, Harmonious, and Creative High-in
come Society. Washington, DC: World Bank.
Young, Allyn (1928) ‘Increasing Returns and Economic Progress’, Economic Journal
38(152): 527–42.
Notes:
(1) According to Weiss (2013: 396) two competing perspectives emerge, namely the ‘pro
motional’ that focuses on interventionist and leading role of government, and the ‘market-
based’ approach which sees government as a facilitator with the role of addressing mar
ket failures and the malfunctioning of the market.
(2) See Komiya, Okuno, and Suzumura (1988) and Ohno (2019) on Meiji’s restoration and
Japanese industrial policy in the nineteenth and twentieth centuries.
(3) Reinert, Gosh, and Kattel (2016). See also Chapter 19.
(4) See Mazzucato (2017). See also Chapter 3.
Page 41 of 48
(5) See Chapter 4.
(6) According to Ocampo, dynamic efficiency is the ‘capacity to generate new waves of
structural change’ (see Chapter 3). See also Ocampo (2017).
(7) For example, the outlawing of local content requirements, which have been a staple of
industrial policy. LDCs may find smart ways to get around imposed restrictions.
(8) See Chapters 10 and 15. For example, sustainability and the development of eco-indus
trial hubs has become a critical aspect of industrial policies.
(10) Axioms of Deng Xiaoping capture this thinking: ‘It doesn’t matter if a cat is black or
white, so long as it catches mice’; ‘Seek truth from facts’; and ‘Crossing the river by feel
ing the stones’. See also Peerenboom (2001).
(11) According to UNCTAD (2006), productive capacity is ‘the productive resources, entre
preneurial capabilities and production linkages’ which together enable a country to devel
op the technological and production capabilities necessary to produce a wide range of
goods (UNCTAD, 2006).
(13) This view influences economic views promoted by the United Nations Economic Com
mission for Latin America and the Caribbean (ECLAC) and in Latin American policymak
ing.
(14) See Chapter 3.
(15) Pasinetti underlines that industrial wealth is associated with technical knowledge and
capability. In capitalism or industrial society, ‘wealth is not a stock of material goods
(which only represent the external expression of it)—it is a stock of technical knowledge…
today it has become to emulate them and do better’ (Pasinetti, 1981: 276). ‘It is only by
absorbing technical knowledge that the poor countries will be able permanently to in
crease their wealth’ (1981: 275).
(16) See Kaldor’s laws as elaborated by Thirlwall (2013) and Mathews (2016b).
(17) Knowledge and technology intensity are increasing in some high-value activities in
primary goods and services such as health care.
(18) See also Chapters 3 and 4.
(19) The intersectoral linkages are noticed through technical and mental shifts, as Kaldor
highlights that ‘the key to an accelerated growth of the underdeveloped areas of the
world lies in bringing about fundamental changes in both the mental outlook and the
technical knowledge and skill of their peasant population’ (Kaldor, 1980: 242).
Page 42 of 48
(20) For instance, see Serra and Stiglitz (2008). This view is represented in various publi
cations of international financial institutions and proponents of free trade and economic
liberalization.
(21) Pasinetti states: ‘The real difficulty is that technological knowledge is far less mobile,
or rather … far less quickly mobile, than goods, so that—when all possible efforts have
been made to improve technical knowledge, and only when all such efforts have been
made—a country can obtain further gains by expansion of international trade’ (Pasinetti,
1981: 272).
(22) Thirlwall (2011: 321) highlights that the constraint on balance of payments is the
prime constraint for sustained growth in an open economy. The balance of payments con
strained growth model is ‘the proposition that no country can grow faster than that rate
consistent with balance of payments equilibrium on current account, unless it can finance
ever-growing deficits, which in general it cannot. There is a limit to the deficit/GDP ratio,
and international debt/GDP ratio, beyond which financial markets get nervous.’
(23) Reinert states that import substitution has preceded exports in England (see Chap
ters 19 and 24).
(24) See UNCTAD (2016: 97‒138).
(25) Hymer (1976), in his pioneering work on FDI and the importance of national origin,
states that ‘direct investments are the capital movements associated with international
operations and firms’ and ‘their nationality is of the utmost importance, for it affects the
way they behave, and it affects the treatment they receive’ because of legal aspects of na
tionality, control of firm operations and taxation, the conversion of profit to the home cur
rency, and a preference for building R&D at home (1976: 29, 30).
(26) See also Goodrich (1965) and Cohen and DeLong (2016).
(27) Oqubay (2015: 26) highlights that ‘the theory of infant industry is based on the as
sumption that the manufacturing sector should play the key role in the economy, and that
its promotion requires jumping ahead of comparative advantage, thus necessitating pro
tection of infant industries, the use of industrial policies, and an indispensable role for the
state’.
(28) This is exemplified in Lin and Chang’s debate on the strategy of conformity (compara
tive advantage defying following, CAF) or defying comparative advantage defying (CAD)
(Lin and Chang, 2009).
(29) As coined by Schwartz (2010).
(30) Marx ([1867] 1990: 399) highlights that ‘modern industry never looks upon and treats
the existing form of a process as final’.
Page 43 of 48
(31) Veblen (1915: 60) argues that ‘the prime mover among these factors [population
growth, military power, industrial efficiency] of the nation’s unfolding power has been its
increased industrial efficiency rather than either the other two factors’. Kim and Nelson
(2000: 1) state that ‘technology advance is the key driving force of economic growth’ and
‘technological advance accounted for the lion’s share of growth in worker productivity’.
(32) Solow highlights the distinction between incremental and true innovations: ‘Innova
tion is the discretion or outcome of researches that bring those development changes to
the nature of the product or the nature of the production process in existing industries, or
may even lead to the creation of recognizably new industries. The less obvious process is
usually described as “continuous improvement” of products and processes. It consists of
an ongoing series of minor improvements in the design and manufacture of standard
products. It leads to advances in customers’ satisfaction, in quality, durability, and relia
bility, and to continuing reductions in the cost of production.’
(33) Paus (2019: 5) states that ‘innovation is primarily the result of the incorporation of
knowledge developed elsewhere: through the use of licenses, the incorporation of new
capital goods, and spillovers from foreign investment in the developing country, if there is
domestic absorptive capacity. The next step is an increase in the development of domestic
innovation, where firms generate new processes and products that make them interna
tionally competitive.’ The Oslo Manual (OECD, 2005) states that innovation ‘goes far be
yond the confines of research labs to users, suppliers and consumers everywhere—in gov
ernment, business and non-profit organisations, across borders, across sectors, and
across institutions’ and proposes four types of innovation: product innovation, process in
novation, marketing innovation, and organizational innovation.
(34) In contrast to articulated or explicit knowledge, tacit knowledge is neither codified

nor explained.
(35) Market fundamentalism advocates a laissez-faire approach and intellectual property

protection (reflecting property rights and rewarding innovators), rejecting the role of the
state.
(36) See Mowery and Rosenberg (1989), Lee (2019a), Dosi and Nelson (2018).
(37) It is useful to note that the notion of catch-up and late development is a notion which
is old and neoclassical economists may well agree with some of this because of the as
sumption of diminishing returns and market-facilitated convergence.
(38) According to the World Bank (2012), only thirteen of the 101 middle-income
economies in 1960 have escaped the middle-income trap. The methodology and data are
flawed as the classification is based on GNI per capita; a static indicator based on a World
Bank classification that was introduced in the late 1970s for purposes other than
analysing economic catch-up. It can be argued that this threshold for high-income econo
my is lower than present times warrant (i.e. about US$12,000). See also Lin and Treichel
(2012) and Lin (2016).
Page 44 of 48
(39) The thirteen economies are Equatorial Guinea, Greece, Hong Kong SAR (China), Ire
land, Israel, Japan, Mauritius, Portugal, Puerto Rico, the Republic of Korea, Singapore,
Spain, and Taiwan, China. Some of these countries have low technological capability and
their economy and exports are not based on diversification and technological intensity
(for example, exports of crude oil account for 98 per cent of Equatorial Guinea’s econo
my).
(40) See Tregenna (2009, 2013) on manufacturing properties and premature de-industrial
ization. See also UNCTAD (2016). According to UNCTAD (2016: 76):
Catch-up industrialization, with robust growth of production, investment, income,

and technological and trade linkages built around a large and increasingly diversi
fied manufacturing sector gives rise to a strong catch-up growth dynamic result
ing in narrowing the productivity gap with lead economies. Stalled industrializa
tion is characterized by stagnant shares of industrial output and employment, and
sporadic growth episodes that generate linkages that are not large or strong
enough for industrial growth to withstand shock and setbacks resulting in contin
ued vulnerability. In general, such a trajectory results in a widening productivity
gap with lead economies. Finally, there is premature deindustrialization in which
the shares of industrial output and employment fall prematurely, at levels of per
capita income much lower than those at which developed economies started to
deindustrialize. This is accompanied by delinking along several dimensions and a
sharp drop in relative productivity levels.
(41) Arrow (1962: 155) states that learning is ‘the acquisition of knowledge…learning is
the product of experience. Learning can only take place through the attempt to solve a
problem and therefore only takes place during activity.’ Moreover, Arrow emphasizes that
a steadily evolving situation acts as stimulus to generate growing productivity.
(42) See also Mathews (2016b).
(43) Hobday (2000: 156) confirms that in East Asia, ‘exports, whether within TNCs or
from local firms, acted as a focusing device for technological learning and investment’.
(44) See also Hall and Soskice (2001) on various national innovation systems that reflect
the varieties of capitalism.
(45) Cohen and Levinthal associate absorptive capacity as like ‘creative capacity’ in the
field of psychology. They emphasize the critical role of learning intensity, diverse back
ground, knowledge diversity, and openness to external environment are critical for facili
tating innovation capability. ‘Learning is cumulative, and learning performance is great
est when the object of learning is related to what is already known…Learning capabilities
involve the development of the capacity to assimilate existing knowledge, while problem-
solving skills represent a capacity to create new knowledge’ (1990: 130).
(46) See Chandler (2004). See also Chapters 23 and 25.
Page 45 of 48
(47) A policy choice of industrial policy is related to the focus on large firms in contrast to
small and medium firms. Large firms are important for developing technological capabili
ty and engaging in international markets, while small and medium firms play an indis
pensable role in employment creation and in production networks with large firms. Ger
many and Japan are examples where such production networks are widely used and have
contributed to international competitiveness.
(48) See Atkinson and Ezel (2012).
(49) For a thorough review on economic catch-up and leapfrogging strategies, see The Art
of Economic Catch-up (Lee, 2019b).
(50) See Amsden and Chu (2003) on Taiwan’s industrial upgrading.
(51) See also Evans (1997), Woo-Cummings (1999), and Wade (2018).
(52) See Khan (2010), Amsden (2012), and Gray (2018).
(54) See Hirschman (1967) on latitude for performance.
(55) Oqubay underlines how ‘understanding the industrial structure of different sectors
and leveraging latitudes for performance are valuable in industrial policymaking.
Hirschman’s linkage concept (the favouring of industries with strong linkages…) is essen
tial in bringing dynamics and impetus to new activities and increasing returns and in cre
ating economic space in developing economies’ (Oqubay, 2015: 288, own emphasis).
(56) See Drucker (1987) on Germany’s industrial competitiveness in machinery and indus
trial goods exports.
(57) In a highly open economy, it is possible that the open domestic market can be as com
petitive as international markets. Thus, competition is naturally strangest in international
markets if the domestic market is totally or partially protected.
(58) See Cramer and Sender (2019).
(59) See Smith (1776), Babbage (2005), Marshall (1920), and Ohlin (1933), among others.
For a comprehensive understanding of theories, context, and empirical perspectives, see
The Oxford Handbook of Industrial Hubs and Economic Development (Oqubay and Lin,
2020).
(60) See UNCTAD (2019), which states there are more than 6,000 industrial hubs (special
economic zones, export-processing zones, industrial parks, and technology parks) global
ly, the lion’s share of these are in Asia.
(61) See Breschi and Malerba (2005) and Best (2001).
Page 46 of 48
(62) See Oqubay and Lin, The Oxford Handbook of Industrial Hubs and Economic Develop
ment (2020). See also Marshall (1920); Jacobs (1969, 1984); Porter (1990) among others.
(63) Babbage (2005: 90): ‘It is found in every country, that the situation of large manufac
turing establishments is confined to particular districts.’ Marshall (1920: 156) states: ‘The
mysteries of the trade become no mystery…good work is rightly appreciated, inventions
and improvements in machinery, in processes and the general organization of business
have their merits promptly discussed: If one man starts a new idea, it is taken up by oth
ers and combined with suggestions of their own; and thus it becomes a source of further
new ideas.’
(64) Breschi and Malerba (2005: 9–10) highlight that ‘starting a cluster and sustaining
clusters are different in processes and economies: a) availability of technical and manage
rial skills, exploiting technological and market opportunities, availability of government
agencies and research institutions, importing intellectual, organizational, technical inputs
from outside the local area’. They emphasize the importance of external linkages: ‘In par
ticular, external linkages are vital in order to establish and maintain a dense local net
work of relationships, for both emerging and established clusters’ (Breschi and Malerba,
2005: 13).
(65) See also UNCTAD (2013: 168): ‘Agglomeration and clustering facilitates economic
benefit from GVC participation’, generating collective efficiency from geographical prox
imity, facilitating learning and business interaction.
(67) See also Oqubay (2015: 288) who highlights that ‘The principle of reciprocity is im
portant in almost all conditions, despite the challenges of implementation and its depen
dence on the state’s political clout’.
(68) See Hirschman (2013: 351), who ‘emphasized the positive role of imbalance in eco
nomic development and of crisis in the achievement of social and economic reform in
Latin America’. See also Hirschman (1958, 1963).
(69) See also Oqubay (2019a, 2019b).
(70) Wade further states that: ‘The key point is that the new phase of globalization charac
terized by GVCs (since the late 1980s) tips the balance of power in the world economy
firmly in favour of MNCs, because if one host government does not agree to their condi
tions, or if labour costs in one country rise too high, the firm can readily shift production
elsewhere’ (2019: 24).
(71) The United States and the United Kingdom are typical examples. See also Chapters 8
and 9.
(72) See Paus (2018).

Page 47 of 48
(74) See also Mathews (2016a) Green Capitalism.
Arkebe Oqubay
Arkebe Oqubay (PhD) is a senior minister and special adviser to the prime minister of
Ethiopia and has been at the centre of policymaking for over twenty-five years. He is
the former mayor of Addis Ababa, winner of the Best African Mayor of 2005 and fi
nalist in the World Mayor Award 2005, for his work transforming the city. He is a re
cipient of the Order of the Rising Sun, Gold and Silver Star, presented by the Emper
or of Japan. He currently serves as board chair of several leading public organiza
tions and international advisory boards. He is an ODI Distinguished Fellow and a re
search associate at the Centre of African Studies in the University of London, and
holds a PhD in development studies from SOAS, University of London. His recent
works include the path-breaking Made in Africa: Industrial Policy in Ethiopia (OUP,
2015); The Oxford Handbook of the Ethiopian Economy (OUP, 2019); How Nations
Learn: Technological Learning, Industrial Policy, and Catch-up (OUP, 2019); China‒
Africa and an Economic Transformation (OUP, 2019); African Economic Develop
ment: Evidence, Theory, and Policy (OUP, 2020); and The Oxford Handbook of Indus
trial Hubs and Economic Development (OUP, 2020). He was recognized as one of the
100 Most Influential Africans of 2016, and a ‘leading thinker on Africa’s strategic de
velopment’ by the New African, for his work, both theoretical and practical, on indus
trial policies.
Page 48 of 48
Industrial Policy, Macroeconomics, and Structural Change
Industrial Policy, Macroeconomics, and Structural

Change
The main argument of this chapter is that the key to rapid economic growth is the dynam
ic efficiency of economic structures, defined as their capacity to generate new waves of
innovative activities. After reviewing the empirical literature that confirms this view, the
chapter argues that the dynamics of production structures may be visualized as the inter
action between two basic forces: innovations (broadly defined) and the learning process
es associated with them; and the complementarities, linkages, or networks among firms
and production sectors. The chapter then analyses the specific structural transformation
challenges that natural resource-dependent economies face. It finally argues that indus
trial policies should be accompanied by appropriate macroeconomic and financial policies
that should guarantee in particular a competitive and stable real exchange rate and long-
term financing for innovative activities, with a crucial role for national development
banks.
Keywords: innovations, complementarities, structural change, dynamic efficiency, commodities real exchange
rate, national development banks
3.1 Introduction
THIS chapter argues that structural change is at the heart of a dynamic process of eco
nomic development, and that active industrial (production-sector development) policies
must be at the heart of an appropriate development strategy. The major policy focus of
that strategy should, therefore, be on the dynamic efficiency of economic structures, de
fined as their capacity to generate new waves of structural change.1 This concept is in
sharp contrast with static efficiency, the central focus of traditional microeconomic and
international trade theories. Dynamic efficiency requires degrees of state intervention
that traditional defendants of static efficiency would also consider unacceptable.
Page 1 of 33
What this means is that economic growth in emerging and developing countries—the fo
cus of this chapter—is intrinsically tied to the dynamics of production structures, the
learning processes associated with technological catch-up and the capacity to gradually
join the world of innovators, and the specific policies and institutions created to support
these processes. The promotion of dynamic efficiency in these countries also includes the
creation of linkages among domestic firms and sectors, and the adequate management of
natural resources in countries that have a strong static comparative advantage in com
modity production. It equally involves the reduction of the heterogeneity of their produc
tion structures, due to the coexistence of low-productivity (informal) activities alongside
high-productivity (modern) firms—a phenomenon (p. 64) that has been alternatively called
both ‘dualism’ and ‘structural heterogeneity’. Avoiding macroeconomic instability is also
essential, particularly guaranteeing competitive and stable real exchange rates, which
are critical for adequate structural change, in the face of terms-of-trade fluctuations and
capital account volatility.
These are basic ideas that have been advanced by all brands of—broadly defined—‘struc
turalism’ in economic thinking.2 The work of Schumpeter, the neo-Schumpeterian, and
evolutionary schools are included within this concept. This encompasses the view, which
originates in Schumpeter’s (1939) analysis of business cycles, that technological revolu
tions come in waves of innovation that gradually spread through the economic system
(Freeman and Soete, 1997; Pérez, 2002: part I), replacing previous technologies and the
firms and sectors that used them, and generating a process of ‘creative
destruction’ (Schumpeter, 1962: ch. VIII). In relation to developing economies, some of
the ideas come from different brands of Latin American structuralism that followed the
work of Raúl Prebisch and the United Nations Economic Commission for Latin America
and the Caribbean (ECLAC), including its most recent brand, ‘neo-structuralism’. We
should also embrace within this broad concept of structuralism the emphasis of classical
development economics on industrialization and external economies as core elements of
economic development, including the notions of backward and forward linkages associat
ed with the work of Hirschman (1958). We can add the growth-productivity connections
associated with Kaldor’s (1978: chs 1 and 2) analysis of economic growth, as well as the
role of increasing returns in contemporary neoclassical models of economic growth.3
Contextual conditions for a dynamic development process have also been emphasized in
the literature. However, they generally play the role of background conditions rather than
that of direct determinants of changes in the growth momentum.4 They include an ade
quate education system and a proper physical infrastructure. They also include an institu
tional context that guarantees a measure of stability in the basic social contract, a non-
discretionary legal system, an impartial (and, ideally, efficient) state bureaucracy, and
smooth business‒labour‒government relations. There are, however, significant differ
ences in concepts about what constitutes a proper institutional context, and certain insti
tutional features are fairly constant over decades in specific countries, whereas growth is
not.5 This chapter therefore leaves aside the analysis of these contextual conditions, re
ferring only to institutions that directly relate to structural change.
Page 2 of 33
It should be added that the distributive effects of structural transformation strate

(p. 65)
gies depend on how far such strategies help reduce the heterogeneity of production
structures through training, technological diffusion, appropriate financing channels, the
promotion of different forms of associations among small entrepreneurs, and commercial
links between large and small firms. However, this is a subject that is only marginally cov
ered here.
The chapter is divided into six sections, the first of which is this introduction. The second
takes a look at growth patterns identified in the old and more recent literature. The third
focuses on the dynamics of production structures, and particularly on its two fundamental
elements: innovations and complementarities. The fourth considers the specific issues
raised by natural resource dependence. The fifth looks at the interrelated issues of fi
nancing structural change and managing financial fluctuations. The last section draws
major policy implications.
3.2 Growth Patterns

The large empirical literature has identified that economic growth involves the simultane
ous movement of a series of economic variables: improved technology, human capital ac
cumulation, investment, savings, and systematic changes in production structures.6
Disentangling cause and effect is not always an easy task. Thus, higher investment and
savings ratios are generally seen as essential for growth accelerations, but they may be
the result rather than the cause of faster growth: the effects of the accelerator mecha
nism on investment, and higher savings associated with income growth and redistribution
effects that go along with it, such as raising firms’ retained profits. In turn, the accumula
tion of skills, an essential element of human capital, is mainly the result of learning asso
ciated with production experience, and the expansion of education systems is facilitated
by the increased social spending enabled by economic growth. Productivity improve
ments may also be the result of growth: learning processes, as well as technical improve
ments embedded in new equipment—the causal link emphasized by Kaldor (1978), which
is the opposite of that assumed by neoclassical growth theory since Solow (1956, 2000).
Thus, many of the regularities mentioned in the growth literature may be subject to
sharply differing interpretations, depending on views on the causal links involved.
A few ‘stylized facts’ may serve, however, as a point of departure for this chapter. The
first is the persistence and even enhancement of the vast inequalities in the world econo
my that arose quite early in the history of modern economic growth. As Rodrik (2014) has
emphasized, convergence in per capita incomes has been the exception (p. 66) rather than
the rule. Indeed, using Maddison’s (2001) data, we can estimate that slightly over 60 per
cent of the variance of per capita income levels in the world at the end of the twentieth
century can be explained by income differences that already existed in 1914. This indi
cates that, although there have been changes in the world income hierarchy, these have
been exceptions. Even in the case of developed countries, strong convergence took place
Page 3 of 33
during the post-Second World War ‘golden age’ of 1950‒73, but not before the Second
World War (Maddison, 1991).7
The most important feature, however, is the divergence of incomes between developed
and developing countries in the nineteenth and twentieth centuries, which Pritchett
(1997) aptly characterized as ‘divergence, big time’. Among the exceptions to this rule,
we can also include the rise of Latin America to middle-income levels since the late nine
teenth and early twentieth centuries and through the inter-war period8 and, of course, the
success of Asian newly industrializing economies since the 1960s and that of China since
the 1980s. The first decade of the twenty-first century, and at a slower rate until the end
of the ‘super-cycle’ of commodity prices in 2014, is perhaps the only case of fairly broad
convergence of per capita income between developed and developing countries in history.
The reasons for divergence are well known. They include the ‘poverty trap’ analysed by
classical development economists, as well as the ‘middle-income trap’ identified in the re
cent literature (see, for example, Gill and Kharas, 2007, 2015; Eichengreen et al., 2012,
2013). They also reflect basic international asymmetries: (i) prohibitive entry costs into
mature sectors and technologically dynamic activities; (ii) differences in domestic finan
cial development and in the stability or volatility of external financing; and (iii) macroeco
nomic asymmetries that generate quite different degrees of freedom to adopt counter
cyclical macroeconomic policies and even a tendency for developing countries to adopt
procyclical policies, due to their dependence on unstable external financing (Ocampo,
2001, 2016).
The main implication of this fact is that economic opportunities are largely determined by
the position that a particular country occupies within the world hierarchy. For this rea
son, economic development is not about following ‘stages’ of growth, but carrying out the
associated structural transformations, and employing the appropriate macroeconomic
and financial strategies within the restrictions that each country’s position within the
world hierarchy creates. This was the essential insight of the Latin American structuralist
school and the literature on late industrialization since Gerschenkron (see Gerschenkron,
1962; Amsden, 2001).
As underscored in the classic work by Chenery and collaborators, growth is accompanied

by regular changes in the sectoral composition of output and the patterns of international
specialization (see Chenery et al., 1986). Episodes of convergence have (p. 67) been gen
erally associated with processes of industrialization and the reallocation of labour from
low- to high-productivity sectors subject to economies of scale and scope (specialization),
but many have also ended in truncated convergences or growth collapses (Ros, 2000,
2013; Easterly, 2001).9 Furthermore, a significant feature of developing countries in re
cent decades has been the premature de-industrialization of Latin America and Africa,
that is, the reduction in the share of manufacturing in employment and GDP at much low
er levels of income per capita than has been typical in similar processes in developed
countries—a concept and a trend that was first highlighted by Palma (2005) and Dasgup
ta and Singh (2006), and more recently by Rodrik (2016). De-industrialization may have
Page 4 of 33
been generated by the way economic liberalization was undertaken, and by the ‘Dutch
Disease’ effects of the super-cycle of commodity prices of the early twenty-first century.
Such a trend is in sharp contrast with the persistent industrialization of East Asia and its
spread to a group of middle- and lower-income countries in South East and South Asia.
Asian countries are also the only ones where recent growth accelerations have mixed
structural change with improvements in labour productivity at the sectoral level, whereas
Latin America has lacked the first element and Africa the second (Diao et al., 2017).
Several authors have also pointed out that episodes of structural change come in spurts
rather than as steady flows, an idea that may be seen as related to the concept of waves
of innovation. The capacity to generate a wave of innovations or absorb one that has al
ready been developed in advanced economies depends, in any case, on production experi
ence, and follows, therefore, a process of path dependence (Arthur, 1994). The comple
mentarities (externalities) among sectors are crucial for a strong growth process to take
place (Rosenstein-Rodan, 1943; Taylor, 1991; Ros, 2000, 2013) and, if they cannot be de
veloped simultaneously, they may generate successive phases of disequilibrium
(Hirschman, 1958). These views imply, in short, that the dynamics of production struc
tures are an active determinant of economic growth and, therefore, that this process can
not be reduced to its aggregate dimensions.
Elastic factor supplies play an essential role in facilitating a smooth expansion of dynamic
activities. Financing facilities for innovative sectors are essential in this regard, as em
phasized in the literature on late industrialization in the now developed countries; as we
will see, public-sector development banks can play a crucial role in this regard. In turn,
the reallocation of labour from traditional to modern sectors also plays an essential role,
as underscored by Lewis (1954, 1969) and many later authors.10 At the same time, howev
er, low economic growth may generate the opposite pattern, in which traditional or infor
mal activities—or the public sector—absorb the labour that modern (p. 68) sectors do not
demand (Ocampo et al., 2009). The interplay between labour mobility and economies of
scale has also been the essential insight of regional economics since its origins, generat
ing urban and regional growth poles (for a modern version, see Fujita et al., 1999). The
‘vent for surplus’ models of international trade, which go back to Adam Smith, also pro
vide an alternative source of elastic factor supplies: un- or underexploited natural re
sources (Myint, 1971: ch. 5).
The role of economic policy in these processes has been the subject of heated controver
sies. In recent decades, the orthodox emphasis has been on the positive role that trade
openness plays in facilitating economic growth, but the simplistic relation between trade
liberalization and growth has been shown to be incorrect, as underscored by several au
thors after the seminal paper by Rodríguez and Rodrik (2001). Indeed, to the extent that
scale economies and learning play an important role in international specialization,11
comparative advantages can be or even are generally created. More broadly, successful
development experiences have been associated with variable policy packages involving
different mixes of orthodox incentives with unorthodox institutional features (‘local here
sies’) (see the comparative analyses of development experiences in Helleiner, 1994, and
Page 5 of 33
Rodrik, 2007, 2014). Thus, protection has been a source of growth in some periods in spe
cific countries, but has blocked it in others; the same thing can be said of freer trade—the
degree of openness in the world economy being critical in this regard. Export growth has
been, of course, a crucial element of East Asian success in recent decades, but has in
volved significant elements of state intervention. Mixed strategies have worked well un
der many circumstances. Indeed, an interesting historical observation is the evidence
that successful experiences of manufacturing export growth in the developing world were
generally preceded by periods of import-substitution industrialization (Chenery et al.,
1986). Bairoch (1993: part I) came to a similar view regarding the role of protection in
the growth of the ‘late industrializers’ during the pre-First World War period, concluding
that the fastest periods of growth in world trade before the First World War were not
those characterized by the most liberal trade regimes.
In macroeconomic terms, there is evidence that long-term growth in developing countries

is positively associated with the capacity to guarantee a competitive real exchange rate,
and thus with the idea that an active exchange rate policy can help foster structural
change.12 In this sense, a competitive exchange rate can be viewed as a type of industrial
policy that can partially substitute for traditional industrial policies, particularly in the
face of restrictions on subsidies to production and exports under World Trade Organiza
tion (WTO) rules. However, it should also be complemented by other industrial policies
(e.g. on access to technology and credit) that increase the elasticity of the aggregate sup
ply to the real exchange rate. (p. 69)
3.3 The Dynamics of Production Structures

As stated in the introduction, the capacity to permanently generate new dynamic activi
ties is the essence of successful economic development. In this sense, structural change
is essentially a meso-economic process that includes the variations in the composition of
production, intra- and inter-sectoral linkages, market structures, the functioning of factor
markets, and the institutions that support all of them. Dynamic microeconomic changes
are the building blocks, but it is system-wide structural change that matters. Further
more, this process has a strong impact on macroeconomic dynamics through its effects on
investment, employment, and trade, and macroeconomic policy can also affect this
process.
The dynamics of production structures may be visualized as the interaction between two
basic forces: (1) innovations, broadly defined as new technologies, new activities and new
ways of doing previous activities, and the learning processes that characterize their full
realization and their diffusion through the economic system; and (2) the complementari
ties, linkages, or networks among firms and production activities. The institutions re
quired to enhance these structural processes are crucial and also subject to learning.
Elastic factor supplies are essential to guarantee that these dynamic processes can de
ploy their full potentialities. It is the combination of these factors that determines the dy
namic efficiency of a given production system (Ocampo, 2017b).
Page 6 of 33
3.3.1 Innovations
The definition of innovations used here follows the broad concept of ‘new combinations’
suggested by Schumpeter (1961: ch. II): new qualities of goods and services; new produc
tion methods or marketing strategies; opening up of new markets; new sources of raw
materials; and new industrial structures. Today we would also add new ways of managing
the environment, including mitigating the effects of climate change. The definition in
cludes technological innovations—the more common use of the concept of innovations in
the economic literature—but also a broader set of micro- and meso-economic processes.
As we saw in the Introduction to this chapter, innovations include not only the creation of
firms, production activities, and sectors, but also the destruction of others—or, using
Easterly’s (2001: ch. 9) terminology, complementary and substitution effects.
Schumpeter’s ‘creative destruction’ is, of course, essential if innovations are to lead to
growth. However, there may be other outcomes: limited destruction but also large-scale
destruction or a mixed negative case, ‘destructive creation’, when the destruction pre
vails over the creative parts of the transformation. Also, some locations may (p. 70) con
centrate the creative and others the destructive effects, for example, when a synthetic
substitute is discovered in an industrial centre that puts producers of the natural raw ma
terial located elsewhere out of business.
In industrial countries, the incentive to innovate is provided by the extraordinary profits

that can be earned by the pioneering firms that introduce technical, commercial, or orga
nizational changes, or which open new markets or find new sources of raw materials. This
incentive is necessary to offset the uncertainties and risks involved in the innovators’ de
cisions, the incomplete nature of the knowledge they initially have, and the fact that, due
to the externalities that the innovation generates, they may not be able to fully appropri
ate its benefits.
In contrast, in developing countries, innovations are largely associated with the transfer
of sectors, new products, technologies, and organizational or commercial strategies previ
ously developed in the industrial centres. The industrial countries’ innovations thus repre
sent the ‘moving targets’ which generate the windows of opportunity for developing coun
tries (Pérez, 2001). The extraordinary profits that innovators enjoy in developed countries
may be absent, as they may involve entry into mature activities with thinner profit mar
gins. Thus, in the absence of policy incentives, there may be a suboptimal search for new
economic activities (Hausmann and Rodrik, 2003).
No innovative process is passive, as it requires investment and learning. It requires in

vestments in physical capital as well as in intangibles, including technological learning.
Technical know-how must indeed go through a maturing process that is closely linked to
the production experience. Climbing up the ladder in the world hierarchy entails shorten
ing transfer periods, taking ‘detours’ to manage existing intellectual property rights in
place and, most importantly, gradually becoming a more active participant in technology
generation (Lee, 2019). It requires national innovation systems to be built up, which
should include an institutional framework to coordinate the various actors engaged in in
Page 7 of 33
novation and learning—research and development centres, universities and technology

schools, extension services, and the innovating firms themselves—and to redirect invest
ments over the long term towards new capabilities and, of course, an ambitious educa
tional strategy that supports these processes.
Essential insights into learning dynamics have been provided by ‘evolutionary’ theories of
technical change.13 These theories emphasize the fact that technology is, to a great ex
tent, tacit in nature, that is, that detailed ‘blueprints’ cannot be plotted. This has three
major implications. The first is that technology is incompletely available and imperfectly
tradable. This is associated with the fact that technology is, to a large extent, composed
of intangible human and organizational capital, which implies that even (p. 71) firms that
purchase or imitate it must invest in mastering the acquired or imitated technology, a
process that involves adaptation and even redesigns and other secondary innovations.
Since this process will be specific to each firm, heterogeneous producers will coexist in
any sector of production. The second implication is that technology proficiency cannot be
detached from production experience: it has a strong learning-by-doing component. This
will also apply, at least in part, to technology creation, which implies that the probability
of major innovations would depend on the accumulated technological knowledge and pro
duction experience of firms, which in new technological fields would include new firms.
The third feature of technical change, unrelated to tacitness, is that diffusion of innova
tions implies that innovative firms only imperfectly appropriate their benefits. Intellectual
property rights provide a mechanism for appropriating those benefits more fully in the
case of technological innovations, but they are not present in other forms of innovation
(such as the development of new activities or a new marketing strategy). Innovations,
therefore, have a mixture of private and public good attributes.
It must be emphasized that these three features of technical change—imperfect tradabili

ty, close association with production experience, and private/public attributes—are equal
ly characteristic of other forms of knowledge, particularly organizational and commercial
know-how, and institutional development. Imperfect tradability, due to its social-capital
attributes, is paramount in the case of organizational knowledge. In turn, commercial
know-how and the development of commercial reputation (goodwill) plays a pivotal role
in international trade (Keesing and Lall, 1992). Moreover, familiarity with the market en
ables producers to modify their products and their marketing channels and helps buyers
learn about suppliers, generating client relations that are important to guarantee the sta
ble growth of firms.
3.3.2 Complementarities
Complementarities are associated with the development of networks of suppliers of goods

and specialized services, marketing channels, and organizations and institutions that dis
seminate information and coordinate the relevant agents. This concept summarizes the
role that backward and forward linkages play in economic growth (Hirschman, 1958) but
also that of (private, public, or mixed) institutions that are created to reduce information
costs (e.g. on technology and markets) and to mitigate the coordination failures that char
Page 8 of 33
acterize interdependent investment decisions (Chang, 1994). Complementarities gener

ate positive externalities among agents, which help reduce their costs. They are the basis
of the dynamic meso-economic economies of scale that determine the competitiveness of
production sectors in a given region or country—or the lack of it. Under these conditions,
competitiveness involves more than microeconomic efficiency: it is essentially a meso-
economic or even a system-wide feature (Fajnzylber, 1990).
The cost and quality of the non-tradable inputs are particularly important in this
(p. 72)
regard. They contain specialized services, including knowledge, and logistic and market
ing services, for which closeness to producers who use the inputs or services may be a
critical factor. They may also include specialized financial services, where closeness can
also be important due to asymmetric information. Also, and although importing tradable
inputs from the best supplier worldwide can encourage export competitiveness, the ca
pacity to generate value chains in which exports have a large domestic value-added con
tent based on national clusters determines how much a given country benefits from trade.
Such contents differ considerably among countries. According to OECD-WTO data, and
focusing only on developing countries that are important manufacturing exporters, the
share of value added in gross exports in 2016 was 83.6 per cent for China, where it has
been rising over the past decade, 69.6 per cent for the Republic of Korea, also rising, but
63.6 per cent and falling for Mexico, and only 56.4 per cent, also falling, for Vietnam.14
As we have seen, the ability of innovative activities to attract capital and labour, and to
gain access to the natural resources they need, will be a critical factor in facilitating the
growth of these activities. One factor is the role of national development banks in facili
tating long-term finance for innovative activities. International capital mobility—particu
larly foreign direct investment—can also play an important role. International labour mi
gration may be critical for skilled labour. Unemployed or, more typically, underemployed
natural resources can facilitate the expansion of innovative sectors that require them—for
example, innovative agricultural activities. And, of course, in the developing world, low-
productivity activities, characterized by a considerable element of underemployment (or
informality), act as a residual supplier of the labour required by a surge of economic
growth. The distinction that dualistic models make between ‘traditional’ and ‘modern’
sectors is inappropriate for describing this feature of the developing world, as high- and
low-productivity sectors are heterogeneous in their structure. The term ‘structural het
erogeneity’, coined by Latin American structuralists (Pinto, 1970) to describe this phe
nomenon, is more appropriate and will thus be used in this chapter.
Structural heterogeneity implies that the dynamism generated by innovative activities

and the strength of the linkages they generate determine the efficiency with which the
aggregate labour force is used (i.e. the extent of labour underemployment), as well as the
underemployment of other factors of production, particularly land. A similar process can
be generated by the better use of existing infrastructure. At the aggregate level, these
processes give rise to Kaldorian growth–productivity links of similar characteristics, but
in addition to the micro- and meso-economic dynamic economies of scales associated with
learning and the development of complementarities. This means, of course, that aggre
Page 9 of 33
gate productivity growth is both a cause and an effect of dynamic economic growth (see
section 3.3.4 below). (p. 73)
3.3.3 The Interplay of Innovations and Complementarities
The interplay between these factors will determine the dynamic efficiency of a given
process of structural transformation. Innovations accompanied by strong complementari
ties will be reflected in the absorption of an increasing number of workers into dynamic
activities. The result will be a virtuous circle of high investment, accelerated technologi
cal learning, and institutional development. On the other hand, destructive forces may
prevail, giving rise to a vicious circle of a slowdown in productivity and economic growth,
decline in investment, increased structural heterogeneity as surplus manpower is ab
sorbed into low-productivity activities, and loss of production experience that widens the
technology gap vis-à-vis industrialized countries.
A simple typology of different processes of structural transformation can be suggested

(Ocampo, 2017b). ‘Deep transformation’ is characterized by strong learning (including in
duced technological innovations) and complementarities (economies of agglomeration
and specialization, and knowledge spillovers) and hence also by strong micro- and meso-
economic dynamic economies of scale, and by the additional productivity effects generat
ed by the reduction in underemployment. This tends to be the pattern in periods of rapid
growth in the developing world and has characterized the East Asian success stories of
recent decades. The opposite, which can be called ‘shallow structural transformation’, is
characterized by the weakness of both learning and complementarities. A classic case is
natural resource enclaves, but as we shall see, this may also be true today of certain
forms of assembly manufacturing.
A first mixed case combines strong learning with weak linkages, due to high import re
quirements. Some import-substitution activities of the past were of this type; in this case,
the initial innovative effect may soon be exhausted due to its limited sectoral or systemic
effects. A second mixed case is the combination of strong linkages with weak learning
processes. In this case, productivity growth at the firm level may be low, but there may be
significant aggregate productivity effects associated with reductions in underemployment
and the development of complementarities. The expansion of labour-intensive export
crops (e.g. coffee) is a case in point.
The association with large-scale international networks of suppliers—global value

chains15—that comes from very active foreign direct investment by multinational firms,
and more codified technology that can be borrowed by paying the associated intellectual
property rights, reduces the costs for developing countries to enter new activities. How
ever, the centralization of research and development efforts and the marketing of the as
sociated products—which can be seen as the two poles of the value chain—also risk ‘shal
low’ structural transformation for the developing countries where assembly manufactur
ing takes place; the lack of domestic complementarities also makes these export activities
footloose. Governments would have to challenge this outcome by promoting domestic
Page 10 of 33
complementarities and transition to new sectors that can benefit from existing specializa
tion patterns.
Finally, the typology is useful for understanding some of the social effects of struc
(p. 74)
tural transformations. In this regard, deep transformations have better effects on formal
employment and standards of living than shallow processes. However, if there is technical
bias in the demand for labour in the first case, wage differentials may increase if the edu
cation policy does not rapidly increase the supply of skilled labour. Strong learning with
weak linkages may lead to increased structural heterogeneity, whereas the opposite com
bination may generate strong demand for low-skilled labour.
3.3.4 The Links between Productivity and Growth
The relationships between structural dynamics and long-term growth can be formalized
as a dual link between economic growth and productivity (Ocampo and Taylor, 1998;
Ocampo et al., 2009: ch. 8). On the one hand, economic growth has positive effects on
productivity through four channels: (i) dynamic economies of scale of a microeconomic
character, associated with learning and induced innovations; (ii) if technology is embod
ied in new equipment, a higher rate of investment induced by faster growth will also in
crease productivity; (iii) the productivity effects of the development of complementarities,
associated with the exploitation of intra- and inter-sectoral external economies
(economies of agglomeration and specialization, and knowledge spillovers); and (iv) the
transfer of underemployed workers to higher-productivity activities. Kaldor (1978: chs 1
and 2) called this link between productivity and production growth (shown as TT in Fig
ure 3.1) the ‘technical progress function’. Following the literature on the topic, it could al
so be referred to as the Kaldor-Verdoorn function.
It must be emphasized that the technical progress function is not an aggregate

(p. 75)
production function. Rather, its positive slope implies that there is some underutilization
of resources at any point in time and, therefore, that growth induces a better allocation of
resources—and the lack of growth reduces aggregate productivity, mainly through the un
deremployment of labour.
FIGURE 3.1 Productivity and GDP dynamics
Page 11 of 33
The second relationship, shown as GG in Figure 3.1, focuses on the reverse causality link:
productivity growth increases economic growth. It can have diverse determinants, which
alternately capture either aggregate supply or aggregate demand effects. First, technical
change directly increases aggregate supply; this is the channel most emphasized in the
growth literature. It also generates new investment that increases aggregate demand,
generating a GG curve of a Keynesian nature. If the economy is foreign exchange con
strained—a situation not uncommon in developing countries—the GG function would be
effectively an aggregate supply function determined by the balance-of-payments restric
tions (Thirlwall, 2011a, 2011b).16 Technical change also improves international competi
tiveness and thus the trade balance and aggregate demand; if the economy has a scarcity
of foreign exchange, it weakens this constraint and has aggregate supply effects.
As both curves have positive slopes, the effects that they capture reinforce each other,
generating alternating positive feedbacks but also possible negative feedbacks. A stable
equilibrium A exists when TT is flatter than GG, as shown in Figure 3.1.17 Given the deter
minants of the technical progress function, TT will be flatter if micro- and meso-economic
dynamic economies of scale are not too strong, or labour underemployment is moderate.
In a Keynesian model, the slope of GG will depend on the elasticity of investment to pro
ductivity growth, whereas in foreign-exchange-constrained models, it will depend on the
elasticities of exports and imports to productivity. In both cases, the higher the elastici
ties the flatter GG will be.
It is important to emphasize that the relationships shown here are assumed to be medium
or long term in character (some short-term macroeconomic effects associated with the
balance of payments will be analysed below). If there is a new wave of innovations, the TT
function will shift upward, to T’T´, accelerating both productivity and income growth at a
new equilibrium point B. As this particular wave of innovations becomes fully exploited,
the function may shift down. In turn, a favourable macroeconomic shock—improved in
vestment financing in a Keynesian model, or improved export prospects or access to ex
ternal financing in a foreign exchange-constrained economy—will shift the GG function
rightwards to G’G´, generating a new equilibrium at C; a negative macroeconomic shock
will, of course, have the (p. 76) opposite effect. With positive productivity and macro ef
fects, the two curves could shift, generating a new equilibrium at D.
In Ocampo (2017b), this simple framework is used to analyse the effects of trade liberal
ization on growth. The net effects are uncertain since they depend on many factors that
affect both functions. In the orthodox view, that opening the economy to competition (in
cluding external competition) unleashes more innovations, then the TT function will shift
up. However, if the response of firms to liberalization is a rationalization of their produc
tion (i.e. a defensive attitude) rather than a new wave of innovation and investment, the
TT curve may not be affected; it may even be adversely affected if the static comparative
advantages are in sectors with limited innovations and complementarities (see section
3.4). On the other hand, through either Keynesian mechanisms or the supply effects char
acteristic of a foreign-exchange-constrained economy, the increase in the propensity to
import generated by a trade reform will lead to a leftward shift in the GG function, with
Page 12 of 33
adverse effects on equilibrium growth. Overall there is, therefore, no general presump
tion that trade liberalization will accelerate growth, as the positive microeconomic links
emphasized by defenders of liberalization may be swamped by adverse structural and
macroeconomic effects.
3.4 Commodity Dependence

Commodities have been and will continue to be at the heart of development in several
parts of the developing world, notably Africa, several parts of Latin America (particularly
South America), the Middle East, and some other Asian countries. Furthermore, the re
cent commodity boom, which started in 2004 and lasted for a decade, generated a ‘re-
primarization’ (or ‘re-commoditization’) in several Latin American countries, understood
as a growing share of natural resource goods in the export basket.18
The links between commodity dependence and development have been the subject of
heated debate in the development literature as to whether commodity dependence pro
motes or obstructs structural change, and particularly whether it benefits or harms the
development of manufacturing and modern services. There is also discussion on how the
macroeconomic challenges associated with commodity price trends and fluctuations
should be managed.
The historical debate on these issues started with the Prebisch–Singer hypothesis, which
claimed that commodity prices tended to deteriorate in the long term relative to those of
manufactures (Prebisch, 1973; Singer, 1950). The original hypothesis involved two com
plementary ideas (Ocampo, 1986). The first was that commodities are characterized by
low income and price elasticities of demand. The second and more interesting suggestion
was that there is an asymmetry between the labour markets of (p. 77) advanced and de
veloping countries, which implies that technological progress in manufactures tends to in
crease real wages in developed countries whereas, given the pool of unskilled labour, it
tends to depress the prices of commodities in the developing world. This coincides with
Lewis’s (1969) analysis of the terms of trade of developing countries. It implies that man
ufactures exported from developing countries may face similar pressures.
Focusing first on long-term commodity price trends, the Prebisch–Singer hypothesis was
largely discarded on empirical and analytical grounds in the three decades after its for
mulation. Interestingly, it was revived by the work of Grilli and Yang (1988) at the World
Bank, who showed that real commodity prices had declined through the twentieth centu
ry. These findings triggered a significant flow of empirical contributions.19 A major con
clusion of this literature is that the factors underlying such long-term trends and cycles
vary through time, including, in recent decades, for example, the rising Chinese demand
for commodities, notably for metals, and that this gives rise to specific features in differ
ent periods.
Page 13 of 33
Overall, Grilli and Yang’s view that the terms of trade have deteriorated through the
twentieth century has been confirmed in the empirical literature. According to Erten and
Ocampo (2013), the adverse trend was stronger in terms of length and intensity for tropi
cal agricultural goods than for temperate-zone agriculture and metals, and short and
weak for oil prices. However, there was no adverse long-term trend in the nineteenth cen
tury, and there has not been one in the early twenty-first century, where there has been
rather an upward trend for metal and oil prices, which will probably come to an end due
to the COVID-19 crisis. In turn, non-oil commodities have experienced four long-term thir
ty-to-forty-year cycles since the late nineteenth century (the last still ongoing), with sub
stantial overlap among different commodity groups, as they are largely determined by
trends in world GDP.
The broader macroeconomic effects of commodity dependence should be analysed from

both a short- and a long-term perspective. The short-term dimensions are closely associ
ated with the cyclical patterns of commodity prices, which generate fluctuations in in
come levels, aggregate domestic demand, and the balance of payments. The procyclical
patterns of investment tend to be particularly strong, and are enhanced by those of both
external and domestic finance (see section 3.5). In commodity-exporting countries, financ
ing cycles tend to follow those of commodity prices, and the recent ‘financialization’ of
commodity markets has reinforced this pattern.
Among the main implications of these cyclical fluctuations are their effects on the real ex
change rate. The cyclical fluctuations of this variable tend to reinforce the variations of
aggregate demand in economies with net liabilities in foreign currency: real exchange-
rate appreciation during booms generates windfall wealth gains that enhance spending,
whereas depreciation during crises generates wealth losses, which accentuates the con
traction of spending. The distributive effects go in the same direction: if the appreciation
benefits workers and the depreciation hurts them, there (p. 78) will also be procyclical ef
fects, given the higher propensity to spend out of wages. As the more traditional macro
economic literature has argued, the effects of real exchange-rate fluctuations on the cur
rent account of the balance of payments will tend to be countercyclical (non-primary ex
ports decreasing and imports rising during commodity booms, and the opposite evolution
occurring during crises). However, if there is an initial surplus during the boom (e.g. due
to macroeconomic adjustments adopted to manage the previous commodity crisis), or an
initial deficit during the crisis (as a result of the strong expansion of aggregate demand
during the boom), the initial effect would also be procyclical and the countercyclical ef
fects will come with a lag.
In terms of cyclical behaviour, the critical choice for governments is whether to adopt a
countercyclical stance, as macroeconomic theory recommends, particularly in its Keyne
sian variants, or follow a procyclical pattern, associated with either economic or political-
economy pressures, or both. In commodity-dependent economies, an important counter
cyclical instrument is a commodity stabilization fund through which the government
saves, during the boom, some of the increased revenues from taxes on commodity sectors
and the profits from state-owned enterprises active in those sectors (particularly impor
Page 14 of 33
tant in oil and minerals sectors).20 This also helps to mitigate the procyclical effects of
commodity prices on real exchange rates if those revenues are kept abroad or saved as
foreign exchange reserves by the central bank. There may, however, be strong political
pressures to spend those revenues, in which case the procyclical effects of commodity
prices will be transmitted in a stronger way to the domestic economy. As we will see in
sections 3.5 and 3.6, the government’s ability to counteract the procyclical effects of a
mix of a commodity boom and procyclical financial flows with countercyclical monetary
policy would be limited if there is free movement of capital; the use of some other instru
ments would, therefore, be necessary.
The long-term structural effects of commodity dependence are associated, in turn, with
whether the commodity sectors generate strong or weak linkages with other economic
activities, and whether commodity dependence is associated with strong or weak produc
tivity growth and learning. In classical analyses of commodity dependence, including
those associated with Prebisch and Singer, the basic arguments were that manufacturing
generates stronger linkages and is a better mechanism to transmit technical progress. As
we have seen, the more recent literature has tended to confirm that rapid economic
growth in emerging and developing countries continues to be associated with industrial
ization drives and, in contrast, that the de-industrialization that Latin America and Africa
have experienced in recent decades is an adverse trend.
It can be argued in favour of commodity dependence that the opportunities for technical
progress and linkages with both the manufacturing and service sectors have been behind
the capacity of commodity-dependent developed countries to prosper.21 Pérez (2010)
mounts a strong defence of the development opportunities provided to Latin America by
its natural resources, arguing that there are ample technological opportunities—biotech
nology, nanotechnology, environmentally friendly products—associated with natural re
sources and the opportunities to exploit the whole value chains of natural resource-inten
sive sectors. In contrast, states Pérez, Latin America is too far behind in other technology
sectors and is no longer a low-wage region. There are also strong complementarities with
Asia, an argument that applies even more strongly to Africa, where China is both a major
market and investor. The major challenge in both cases is how to diversify into non-com
modity sectors, using rising commodity revenues (including those from commodity-pro
ducing state-owned enterprises).
The long-term effects are not independent of the cyclical effects of commodity depen
dence. This implies that the structural vulnerabilities associated with commodity depen
dence are combined with short-term macroeconomic vulnerabilities. Those associated
with real exchange-rate fluctuations are particularly important. Real exchange-rate ap
preciation during commodity booms has a negative effect on non-commodity tradable sec
tors (both exporting and import competing) during booms—an effect that has been
strongly emphasized by the ‘Dutch disease’ literature.22 Firms in non-resource tradable
sectors may go bankrupt during commodity booms, generating permanent effects on eco
nomic structures and productivity if the latter is associated with production experience
(Krugman, 1990: ch. 7). The unstable incentives associated with real exchange-rate fluc
Page 15 of 33
tuations through the business cycle also make the profitability of those sectors highly
volatile, reducing investment in structural diversification.
Beyond the structural and macroeconomic vulnerabilities mentioned lie other vulnerabili
ties of a more political-economy or institutional character. In this regard, the literature on
the ‘Dutch disease’ has emphasized the institutional effects of the rentierism associated
with natural resources. There may also be significant distributive effects associated with
land concentration in agriculture and, in the cases of hydrocarbons and mining, high in
dustrial concentration.
3.5 Macroeconomics, Finance, and Structural

Change
There are two crucial links between macroeconomics, finance, and structural change. The
first is the positive contribution made by financial institutions that support innovative sec
tors with long-term risk capital and lend ‘patient capital’ (Mazzucato, 2013). Funding can,
of course, be external or domestic, with national development banks (NDBs) having an
important role to play in domestic financing. The second link, in (p. 80) contrast, concerns
whether finance risks undermining growth, because it generates boom‒bust macroeco
nomic cycles—again of external or domestic origin—that may end up in costly crises. Im
portant links between capital flows and structural change are (i) the possible misalign
ment of exchange rates during booms and (ii) exchange-rate volatility through the busi
ness cycle, with a negative effect on innovative tradable sectors.
Focusing first on international capital flows, the key question is which of two effects on
the domestic economy will prevail. The first is the direct contribution to growth if it leads
to higher domestic investment, and particularly to key areas of innovations or domestic
competitiveness. The contrasting effect is the risk that growth is undermined because ex
ternal financing can be consumed, and in that case substitutes domestic savings, but par
ticularly because it is potentially reversible and can lead to ‘sudden stops’ that generate
costly crises (Calvo, 1998).
If investment is increased, whether it is channelled to sectors with higher productivity or

faster productivity growth will generally depend on domestic economic structures rather
than on capital flows as such. From a policy perspective, it may be possible, however, to
encourage foreign direct investment which brings technological innovations, increases
exports, and/or produces for the domestic and foreign markets with higher domestic val
ue-added contents.
Multilateral development banks (MDBs) can also play a role in supporting innovations, in
cluding the development of relevant domestic institutions. They can also support activi
ties with significant externalities, notably infrastructure and investments that contribute
to combatting climate change. Recent analyses have underscored the role of MDBs and
sovereign wealth funds in providing financing to reduce the large infrastructure gaps that
Page 16 of 33
characterize the developing world, as well as the full development of infrastructure as an

asset class (see, for example, Bhattarcharya et al., 2015).
A major countercyclical role, recognized by all MDBs since the 2007‒09 North Atlantic fi
nancial crisis, is that of increasing their lending or investments by their financial corpora
tions in developing countries when international private capital flows experience a major
downward swing or a sudden stop. MDBs significantly increased their financing during
that crisis and its aftermath (Ocampo, 2017a: table 5.3). All MDBs were also capitalized
during those years; two more recent institutions of this type are the Asia Infrastructure
Investment Bank (AIIB) and the BRICS’ New Development Bank, in both of which China is
playing a leading role.
An extensive literature discusses the macroeconomic risks associated with cross-border

capital flows and the policies required to manage them. The major risks are the volatility
of international private capital flows, the procyclical pattern of country risk premiums,
and the contagion that characterize both booms and sudden stops of external financing.
The literature has identified a sort of hierarchy of volatility, with FDI being the more sta
ble, and short-term bank lending and portfolio flows more unstable (Rodrik and Velasco,
2000).23 These problems are particularly important in (p. 81) relation to emerging
economies, but also affected peripheral Europe during the North Atlantic crisis, and are
also increasingly important to low-income countries—the ‘frontier markets’ in current ter
minology.
From a theoretical perspective, the major problem of volatile flows is that they have nega
tive externalities, as individual investors and borrowers do not take into account the ef
fects of their financial decisions on other investors and, overall, on the level of financial
stability in a particular country (Jeanne and Korinek, 2010; Korinek, 2011). From an em
pirical perspective, the intellectual battle over the effects of capital market liberalization
was settled by a major International Monetary Fund (IMF) study (Prasad et al., 2003),
which showed that it generates stronger business cycles in developing countries, and to a
lesser extent in developed countries. This was also a major conclusion of the Commission
on Financial Stability convened by the Bank of International Settlements after the out
break of the North Atlantic financial crisis (BIS, 2009).
Strong evidence also comes from later studies. Gourinchas and Jeanne (2007), among
others, have shown that countries that have grown more are the ones that have relied
less, not more, on capital flows for growth, and have therefore run stronger current ac
count balances; this result is, of course, related to the links between competitive ex
change rates and growth. The ‘meta-regression’ analysis by Jeanne et al. (2012: ch. 3) al
so found very limited evidence of a link between financial globalization and growth in the
period 1970‒2007.
In terms of macroeconomic and financial policy, the major implications of these effects
are that capital account liberalization generates major risks in emerging and developing
countries, and that a proper macroeconomic policy in these countries should include the
use of capital account regulations (CARs) to manage the risks of cross-border flows, as
Page 17 of 33
part of the broader family of ‘macroprudential’ regulations (Ocampo, 2017a: ch. 4).24
Section 3.6, the final section of this chapter, provides an additional discussion of this is
sue.
At the domestic level, the central problem in many (or even most) emerging and develop
ing countries is that their financial markets are thin, that is, they are characterized by the
strong prevalence of short-term financial assets and liabilities. This means that long-term
financing is limited, forcing firms to rely on short-term loans for their investments, or lim
iting them to what they can finance with retained profits. From a stability perspective,
the major issue is the variable mixes of maturity and currency mismatches in portfolios.
This means that, during crises, creditors may not roll over short-term loans, thus generat
ing a liquidity crunch, or may subject domestic borrowers to interest rate increases at a
time when their revenues are falling. Domestic bond markets—if they have developed—
will also shrink and be subject to shorter maturities and/or higher interest rates. For larg
er firms that have borrowed abroad, (p. 82) debt ratios will rise if exchange rates depreci
ate. The limited development of future markets implies that the capacity of agents to cov
er these mismatches would be very limited.
Given the limitations and stability issues that domestic financial sectors face, NDBs play
an essential role from both growth and stability perspectives. As argued in Griffith-Jones
and Ocampo (2018), NDBs should have five main functions, which help cover associated
market failures: (i) providing countercyclical financing; (ii) promoting innovation and
structural change; (iii) financing infrastructure investment; (iv) enhancing financial inclu
sion; and (v) supporting the provision of public goods, particularly combatting climate
change.
Function (ii) is particularly important for the topics analysed in this chapter, but other di
mensions also potentially are. Function (i) makes development banks an additional instru
ment of countercyclical macroeconomic policy, and (iii) makes them an instrument of in
frastructure financing, two functions we have discussed in relation to MDBs. Promoting
small start-ups or SMEs that link to them, as part of the broader objective of financial in
clusion (iv), may also be essential for structural change. And many of the activities associ
ated with mitigating and adapting to climate change, included under (v), are innovative
activities in themselves. But I will underscore the function of development banks as
providers of ‘patient capital’ to support innovative sectors and firms.
The failure of private financial markets to deliver adequate long-term funding is behind
the history of NDBs in many developing but also developed countries. They are a crucial
feature of financial sectors in successful emerging economies like China, India, and the
Republic of Korea, but also in prosperous developed countries, notably Germany. After a
long period of neglect in the academic and policy literature, they have been the subject of
renewed interest by MDBs25 and by policymakers in several developed and developing
countries, some of which have created NDBs over the past decade.
Page 18 of 33
The evidence from World Bank data indicates that NDBs played a countercyclical role in
the wake of the North Atlantic financial crisis (Luna-Martinez and Vicente, 2012). In turn,
Mazzucato and Penna (2018) have argued that there is mounting evidence that NDBs
have fostered patient, long-term committed finance for mission-oriented investment in in
novative activities. The case studies analysed in Griffith-Jones and Ocampo (2018) corrob
orate this: the role of KfW, the German development bank, in the development of renew
able energy; of the CDB, the Chinese NDB, in nurturing high-tech ventures since the
2000s; of Brazil’s BNDES in financing programmes targeted at high-tech firms and pro
moting a successful venture capital fund; and of the successful start-up programme of
CORFO, the Chilean development agency, among others.
One of the key features of successful NDBs is, of course, providing leverage to attract pri
vate investors and deepen domestic financial markets. The development of new (p. 83) in
struments, such as guarantees, equity—including venture capital—and debt funds have
been major innovations in this regard. Loan instruments continue to be important, though
greater emphasis than in the past is placed on second-tier loans. A key is, of course, the
long-term character of NDBs’ loans (over 50 per cent of their lending is for ten years’ ma
turity or more). In the area of financial inclusion, correspondent stores have also been an
important new instrument that can be widely used by commercial banks (Colombia being
a success story in this regard). In Mexico, NAFINSA also operates an online reverse fac
toring system called Productive Chains, which allows SMEs to sell their accounts receiv
ables from large companies to private banks, providing themselves with working capital.
Several NDBs have helped deepen financial markets by introducing local currency and
green bonds into their local capital market.
Expanding the role of NDBs in countries that have them, or creating them in those that
do not, would therefore help create a financial system that better serves development
needs. These activities should be linked to strong development policies with structural
change at their heart, they should have adequate support from macroeconomic policies,
and be buttressed by good governance structures that guarantee, in particular, that they
are not used for rent-seeking. This does not necessarily imply large government re
sources, as the only public contributions would be an increase in their paid-in capital and
special programmes that governments want to promote. NDBs would then fund their op
erations on the private domestic market, as well as international capital markets—includ
ing through the support of MDBs and their financial corporations.
3.6 Policy Implications

The main conclusion of this chapter is that the key to rapid growth in the developing
world is the dynamic efficiency of economic structures, defined as their capacity to gener
ate new waves of innovative activities. The strategies aimed at promoting structural
change should be mixed with appropriate macroeconomic and financial policies. They al
so require appropriate institutional frameworks, the formation of human capital, and the
development of infrastructure, but these additional conditions (not analysed here) only
Page 19 of 33
serve as the context for the structural transformation and are not in themselves sufficient
to guarantee dynamic economic growth.
The focus on structural dynamics helps to identify the policy areas and specific institu
tions that authorities should target to accelerate economic development. The first is en
couraging innovations—in the broad sense of the term—and the associated learning
processes in the areas of technological development, productive organization, and mar
keting strategies. In emerging and developing countries, the diversification of production
structures that this policy requires may be largely associated with the transfer of sectors
of production from the industrialized world. The second policy area is the development of
complementarities—backward and forward linkages in (p. 84) Hirschman’s terminology—
aimed at guaranteeing that innovations spread through the economy, at the same time en
suring system-wide competitiveness. Non-tradable inputs and specialized services (knowl
edge, logistics, and marketing services) are particularly important, but tradable inputs al
so are, as part of a policy aimed at increasing the domestic value added of a given eco
nomic activity.
An adequate structural transformation strategy should therefore develop a fair balance

between individual entrepreneurial initiatives—microeconomic dynamics—and equally ap
propriate meso-economic processes that diffuse innovations through the economy and
help create appropriate complementarities. This includes the establishment of institu
tions aimed at increasing information and coordination among agents; different mixes of
public and private institutions can play that role, according to the tradition of each coun
try. Moreover, different mixes of international, national, and local institutions can play a
positive role in this regard.
The appropriate strategies should mix horizontal and selective policies. Although a funda
mental advantage of the former is their neutrality vis-à-vis individual agents, selective
policies must be part of an effective structural diversification strategy that reinforces suc
cessful specialization patterns, helps nurture ‘infant sectors’, and creates comparative ad
vantages. These policies must include support for research and development in the rele
vant sectors, the institutions that link firms in those sectors, their export strategies, and
special long-term credit lines from NDBs. Furthermore, when there are limited resources,
any ‘horizontal’ policy must be detailed and, hence, necessarily becomes selective. Clear
examples are the allocation of resources from funds for technological development and
export promotion. Recognizing that there is an implicit selectivity in horizontal policies
will lead to a better allocation of resources than a supposedly neutral stance.
Under current global conditions, emphasis should be placed on integrating into dynamic
global markets and thus on developing competitive export sectors, as well as mixing in
dustrial and competition policies. Incentives should be granted based on performance,
generating ‘reciprocal control mechanisms’, to borrow Amsden’s (2001) term. In this re
gard, the capacity to export is indeed the best control mechanism, as underlined by
Oqubay in his contribution to this volume (see Chapter 2) and by Cherif and Hasanov
(2019)—and, of course, as the East Asian success stories indicate. The institutional struc
Page 20 of 33
ture that guarantees this should be subject to periodic evaluations, within its learning
path.26
A complex issue is the framework of international rules, especially those of the WTO and
the wave of bilateral and plurilateral free-trade agreements. In this regard, although pri
ority should be given to taking advantage of the manoeuvring room provided under exist
ing agreements, there is a strong sense that a larger policy space (to borrow the term ex
tensively used in UN debates) should be made available to the governments of developing
countries, as policy autonomy has been severely restricted in (p. 85) trade negotiations. In
particular, according to the analysis presented in this chapter, they should be allowed to
apply selective policies and performance criteria to encourage innovation and create the
complementarities that are essential for development. To the extent that the current on
going trade wars undermine the world trading system, regional integration processes
among emerging and developing countries may be particularly attractive.
According to our analysis, structural transformation is not a ‘once and for all’ process,
but rather a persistent task, as the structural transformation process is continuous and
may face obstacles at any stage. To the extent that in developing countries innovative ac
tivities are largely the result of the spread of new sectors and technologies previously
created in the industrial centres, these activities may be regarded as the new set of ‘in
fant sectors’ to be promoted—particularly infant export activities. Furthermore, accord
ing to the analysis presented here, the process of transformation is by no means smooth:
destruction is a constant companion of creation, and structural heterogeneity is a persis
tent feature that may increase at different stages. Distributive tensions are presumably
associated with both factors. In this context, supporting the restructuring of firms in old
sectors and regions that concentrate them, avoiding transformation processes that in
crease structural heterogeneity, and working to upgrade low-productivity activities and
generate positive links with high-productivity sectors are critical for achieving a more eq
uitable development process.
As part of their broader set of functions, NDBs can play a crucial role in guaranteeing the
availability of long-term financing for innovative sectors, and should interact closely with
private financial agents. Private investment banking and venture capital can also play a
role, but past and recent experience indicates that they do not automatically expand opti
mally in developing countries. Hence the importance of private financial agents working
together with NDBs, which in their turn should help build deeper domestic financial sec
tors. Access to international financial services of this sort may also be important to guar
antee funding of innovative activities, but this may generate a strong bias in favour of
multinational and large domestic firms and against small and medium-sized enterprises.
Macroeconomic policies should aim, in turn, at smoothing business and investment cy
cles, and guaranteeing a competitive and relatively stable real exchange rate. Smoothing
cyclical commodity price fluctuations and external financing boom‒bust cycles is essen
tial for the relative stability of the exchange rate. Absorbing part of the commodity booms
with stabilization funds or taxes is critical for managing the first of these problems, while
Page 21 of 33
CARs are essential to regulate the second. The latter should be complemented at the do
mestic level with regulatory policies aimed at avoiding unsustainable credit booms, and
managing the maturity and currency mismatches in portfolios, and the incompleteness of
futures markets. NDBs should also be active in the provision of countercyclical financing
at the national level, complementing the role played by MDBs at the international level.
A competitive exchange rate can be seen as a type of industrial policy, indeed perhaps as
the best ‘neutral’ industrial policy, particularly in the face of restrictions (p. 86) on subsi
dies to production and exports under WTO rules. However, exchange-rate policy alone
may fail to encourage diversification: it should be complemented by other industrial poli
cies that increase the elasticity of the aggregate supply of tradables to the real exchange
rate. A competitive exchange rate may also benefit sectors, particularly natural resource
sectors that should not be subject to specific incentives. This implies that an active ex
change-rate policy must be combined with taxes on sectors with no externalities, smaller
learning spillovers, and weak domestic complementarities (Guzman et al., 2018). The sta
bility of the real exchange rate is also essential to provide stable profit incentives, which
would help reduce the uncertainties that characterize investment in innovative sectors.
The policy interventions necessary to guarantee a competitive and stable real exchange
rate should include CARs and exchange-rate management; both aim at facilitating a more
positive relationship between international capital flows and macroeconomic manage
ment (Ghosh et al., 2017; Ocampo, 2016, 2017a: ch. 4). CARs play the dual roles of both
macroeconomic and financial stability tools. As a macroeconomic instrument, they pro
vide greater room for countercyclical monetary policies. During booms, they increase the
space needed for contractionary monetary policies while mitigating the exchange-rate ap
preciation pressures that such monetary policies may generate. During crises, they can
create space for expansionary monetary policies while constraining capital flight as well
as excessive exchange-rate depreciation that would otherwise partly translate into do
mestic inflation and rising debt/GDP ratios. In turn, when viewed as a financial stability
tool, CARs recognize the fact that there is a ‘hierarchy’ of volatility, as reversibility is par
ticularly important for portfolio flows and short-term bank lending.
There is a broad consensus in the literature that CARs help improve the composition of
capital flows towards less reversible flows, and provide room for countercyclical mone
tary policies. As Erten and Ocampo (2017) have shown, they also reduce the ‘foreign ex
change pressure’ generated by capital flows in emerging and developing countries. All
these advantages mean that there is now a broad consensus in the international policy
debate that the full liberalization of the capital account is not desirable and that CARs
can play a positive macroeconomic role—views that can particularly be seen in the IMF’s
‘institutional view’ on capital account management (IMF, 2012).
CARs can and should be combined with active intervention in foreign exchange markets
and effective management of foreign exchange reserves in a countercyclical way: accu
mulation during booms and use of reserves as a stabilization tool during crises. Counter
cyclical foreign exchange reserve management has indeed been a widespread practice in
Page 22 of 33
emerging and developing countries since the East Asian crisis (Ocampo, 2017a: ch. 2).
However, the ‘self-insurance’ that they provide is costly, as it involves accumulating an as
set that has low yields (foreign exchange reserves) to compensate for the entry of private
capital inflows, which have higher costs; if reserve accumulation is sterilized, central
banks will also incur losses. CARs are, therefore, a less costly policy instrument, but
countries may be reluctant to use them because they (p. 87) are seen as a distortion in fi
nancial markets, and there may be restrictions on their use associated with investment
treaties.
In summary, the combination of CARs with exchange-rate and foreign exchange manage
ment, the countercyclical monetary policy that they facilitate, and countercyclical fiscal
policies, forms the appropriate macroeconomic policy package. Aside from its contribu
tions to countercyclical management, this policy package has long-term development im
plications, in that it contributes to maintaining a competitive and relatively stable real ex
change rate. It creates a positive relation between international capital flows, macroeco
nomic stability, structural transformation, and economic growth.
Acknowledgements
This chapter borrows from the author’s previous work on the subject, and from joint work
with Bilge Erten, Stephany Griffith-Jones, Martin Guzman, Codrina Rada, Joseph E.
Stiglitz, and Lance Taylor, whose contributions are gratefully acknowledged. In particular,
the section on the dynamics of production structures borrows from Ocampo (2017b) and
my analysis of financing issues from my work with Stephany Griffith-Jones (Griffith-Jones
and Ocampo 2018). The author thanks Reda Cherif, Fuad Hasanov, Arkebe Oqubay,
Gabriel Porcile, and Rajah Rasiah for comments on a previous draft of this chapter, also
Verónica Pérez for her support in its drafting.
References
Aghion, Philippe and Peter Howitt (1998) Endogenous Growth Theory. Cambridge, MA:
The MIT Press.
Amsden, Alice (2001) The Rise of the Rest: Non-Western Economies’ Ascent in World Mar
kets. Oxford: Oxford University Press.
Arthur, W. Brian (1994) Increasing Returns and Path Dependence in the Economy. Ann Ar
bor, MI: University of Michigan Press.
Bairoch, Paul (1993) Economics and World History: Myths and Paradoxes. Chicago, IL:
University of Chicago Press.
Barro, Robert J. and Xavier Sala-i-Martin (2003) Economic Growth. New York: McGraw-
Hill.
Page 23 of 33
Bértola, Luis and José Antonio Ocampo (2012) The Economic Development of Latin Ameri
ca since Independence. New York: Oxford University Press.
Bhattarcharya, Amar, Jeremy Oppenheim, and Nicholas Stern (2015) Building Sustainable
Development through Better Infrastructure: Key Elements of a Transformation Program,
Brookings Institution, The New Climate Economy and Grantham Research Institute on
Climate Change and the Environment, July.
BIS (Bank for International Settlements) (2009) ‘Capital Flows and Emerging Markets’.
Report of the Committee on the Global Financial System, CGFS Papers No. 33, January.
Blomström, Maguns and Patricio Meller (eds) (1991) Diverging Paths: Comparing a Cen
tury of Scandinavian and Latin American Economic Development. Washington, DC: Inter-
American Development Bank, and Johns Hopkins University Press. (p. 88)
Calvo, Guillermo A. (1998) ‘Capital Flows and Capital Markets Crises: The Simple Eco
nomics of Sudden Stops’, Journal of Applied Economics 1(1): 35–54.
Chang, Ha-Joon (1994) The Political Economy of Industrial Policy. London: Macmillan and
St. Martin’s Press.
Chenery, Hollis, Sherman Robinson, and Moshe Syrquin (1986) Industrialization and
Growth: A Comparative Study. New York: The World Bank, Oxford University Press.
Cherif, Reda and Fuad Hasanov (2013) ‘Oil Exporters’s Dilemma: How Much to Save and
How Much to Invest’, World Development 52(December): 120–31.
Named: Principles of Industrial Policy’. IMF Working Paper WP/19/74, March. Washing
ton, DC: IMF.
Corden, W. Max and J. Peter Neary (1982) ‘Booming Sector and De-industrialization in a
Small Open Economy’, Economic Journal 92(December): 825–48.
Cripps, T. Francis and Roger J. Tarling (1973) ‘Growth in Advanced Capitalist Economies
1950–1970’. Occasional Paper No. 40, University of Cambridge, Department of Applied
Economics.
Dasgupta, Sukti and Ajit Singh (2006) ‘Manufacturing, Services and Premature Deindus
trialization in Developing Countries: A Kaldorian Analysis’. Research Paper No. 2006/49,
World Institute for Development Economic Research (UNU-WIDER).
Diao, Xinshen, Margaret McMillan, and Dani Rodrik (2017) ‘The Recent Growth Boom in
Developing Economies: A Structural Change Perspective’. NBER Working Paper No.
23132. Cambridge, MA: NBER.
Dosi, Giovanni, Christopher Freeman, Richard Nelson, Gerald Silverberg, and Luc Soete
(eds) (1988) Technical Change and Economic Theory, Maastricht Economic Research In
Page 24 of 33
stitute on Innovation and Technology (MERIT)/The International Federation of Institutes

for Advanced Studies (IFIAS). London and New York: Pinter Publishers.
Easterly, William (2001) The Elusive Quest for Growth: Economists’ Adventures and Mis
adventures in the Tropics. Cambridge, MA: MIT Press.
Easterly, William, Michael Kremer, Lant Pritchett, and Lawrence Summers (1993) ‘Good
Policy or Good Luck? Country Growth Performance and Temporary Shocks’, Journal of
Monetary Economics 32(December).
Eichengreen, Barry, Donghyun Park, and Kwanho Shin (2012) ‘When Fast Growing
Economies Slow Down: International Evidence and Implications for China’, Asian Eco
nomic Papers 11: 42–87.
Eichengreen, Barry, Donghyun Park, and Kwanho Shin (2013) ‘Growth Slowdowns Redux:
New Evidence on the Middle-income Trap’. NBER Working Paper No. 18673. Cambridge,
MA: NBER.
Erten, Bilge and José Antonio Ocampo (2013) ‘Super Cycles of Commodity Prices since
the Mid-nineteenth Century’, World Development 44: 14–30.
Erten, Bilge and José Antonio Ocampo (2017) ‘Macroeconomic Effects of Capital Account
Regulations’, IMF Economic Review 65(2): 193–240.
Fajnzylber, Fernando (1990) ‘Industrialization in Latin America: From the “Black Box” to
the “Empty Box”’, Cuadernos de la CEPAL, No. 60, Santiago.
Freeman, Chris and Luc Soete (1997) The Economics of Industrial Innovation. 3rd edi
tion. Cambridge, MA: MIT Press.
Frenkel, Roberto and Martin Rapetti (2014) ‘The Real Exchange Rate as a Target of
Macroeconomic Policy’. Paper No. 74. Buenos Aires: Iniciativa para la transparencia fi
nanciera. Available at www.itf.org.ar. (p. 89)
Fujita, Masahisa, Paul Krugman, and Anthony J. Venables (1999) The Spatial Economy:
Cities, Regions, and International Trade. Cambridge, MA: MIT Press.
Furtado, Celso (1961) Desarrollo y subdesarrollo, Colección Cuadernos 196. Buenos

Aires: Editorial Universitaria.
Gerschenkron, Alexander (1962) Economic Backwardness in Historical Perspective.

Cambridge, MA: Harvard University Press.
Ghosh, Atish R., Jonathan D. Ostry, and Mahvash Saeed Qureshi (2017) Taming the Tide of
Capital Flows: A Policy Guide. Cambridge, MA: MIT Press.
Gill, Indermit and Homi Kharas (2007) An East Asian Renaissance: Ideas for Economic
Growth. Washington, DC: World Bank.
Page 25 of 33
Gill, Indermit and Homi Kharas (2015) ‘The Middle-income Trap Turns Ten’. World Bank
Policy Research Working Paper No. 7403. Washington, DC: World Bank.
Gourinchas, Pierre-Olivier and Olivier Jeanne (2007) ‘Capital Flows to Developing Coun
tries: The Allocation Puzzle’. NBER Working Paper No. 13602. Cambridge, MA: NBER.
Griffith-Jones, Stephany and José Antonio Ocampo (2018) The Future of National Develop
ment Banks. Oxford: Oxford University Press.
Grilli, Enzo R. and Maw Cheng Yang (1988) ‘Primary Commodity Prices, Manufactured
Goods Prices, and the Terms of Trade of Developing Countries: What Long Run Shows’,
The World Bank Economic Review 2(1): 1–47.
Grossman, Gene M. and Elhanan Helpman (1991) Innovation and Growth in the Global
Economy. Cambridge, MA: The MIT Press.
Guzman, Martin, José Antonio Ocampo, and Joseph E. Stiglitz (2018) ‘Real Exchange
Rates for Economic Development’, World Development 110: 51–62.
Hausmann, Ricardo and Dani Rodrik (2003) ‘Economic Development as Self-discovery’,

Journal of Development Economics 72: 603–33.
Helleiner, Gerald K. (ed.) (1994) Trade Policy and Industrialization in Turbulent Times.
New York: Routledge and UNU/WIDER.
IMF (2011) World Economic Outlook (WEO). Tensions from the Two-Speed Recovery: Un
employment, Commodities, and Capital Flows. Washington, DC: International Monetary
Fund.
IMF (2012) The Liberalization and Management of Capital Flows: An Institutional View.
Washington, DC: International Monetary Fund.
Jeanne, Olivier and Anton Korinek (2010) ‘Managing Credit Booms and Busts: A Pigou
vian Taxation Approach’. NBER Working Paper No. 16377. Cambridge, MA: NBER.
Jeanne, Olivier, Arvind Subramanian, and John Williamson (2012) Who Needs to Open the
Capital Account? Washington, DC: Peterson Institute for International Economics.
Kaldor, Nicholas (1978) Further Essays on Economic Theory. London: Duckworth.
Katz, Jorge (1987) ‘Domestic Technology Generation in LDCs: A Review of Research Find
ings’, in Jorge Katz (ed.) Technology Generation in Latin American Manufacturing Indus
tries. London: Macmillan.
Page 26 of 33
Keesing, Donald B. and Sanjaya Lall (1992) ‘Marketing Manufactured Exports from De
veloping Countries: Learning Sequences and Public Support’, in Gerald K. Helleiner (ed.)
Trade Policy, Industrialization, and Development: New Perspectives. New York: Oxford
University Press and WIDER.
Korinek, Anton (2011) ‘The New Economics of Prudential Capital Controls: A Research
Agenda’, IMF Economic Review 59: 523–61. (p. 90)
Krugman, Paul (1990) Rethinking International Trade. Cambridge, MA: MIT Press.
Lall, Sanjaya (1990) Building Industrial Competitiveness in Developing Countries. Paris:

OECD Development Center.
Lall, Sanjaya (2003) ‘Technology and Industrial Development in an Era of Globalization’,

in Ha-Joon Chang (ed.) Rethinking Development Economics. London: Anthem Press, pp.
277–98.
Lee, Keun (2019) The Art of Economic Catch-Up: Barriers, Detours and Leapfrogging in
Lewis, W. Arthur (1954) ‘Economic Development with Unlimited Supplies of Labor’, Man
chester School of Economic and Social Studies 22: 139–91.
Lewis, W. Arthur (1969) Aspects of Tropical Trade, 1883–1965. Stockholm: Almqvist &
Wicksell, Wicksell Lectures.
Lin, Justin Yifu (2012) The Quest for Prosperity: How Developing Countries Can Take Off.
Princeton, NJ: Princeton University Press.
Lucas, Robert E., Jr. (1988) ‘On the Mechanics of Economic Development’, Journal of
Monetary Economics 22(1): 3–42.
Luna-Martinez, José de and Carlos Leonardo Vicente (2012) ‘Global Survey of Develop
ment Banks’. World Bank Policy Research Working Paper No. WPS 5969. Washington, DC:
World Bank.
Maddison, Angus (1991) Dynamic Forces in Capitalist Development: A Long-Run Compar

ative View. Oxford: Oxford University Press.
Maddison, Angus (2001) The World Economy—A Millennial Perspective. Paris: Develop
ment Centre Studies, OECD.
Mazzucato, Mariana (2013) The Entrepreneurial State: Debunking the Public vs. Private
Myth in Risk and Innovation. London: Anthem Press.
Mazzucato, Mariana and Caetano C. R. Penna (2018) ‘National Development Banks and
Mission-Oriented Finance for Innovation’, in Stephany Griffith-Jones and José Antonio
Page 27 of 33
Ocampo (eds) The Future of National Development Banks. Oxford: Oxford University
Press.
Missio, Fabricio José, Frederico G. Jayme Jr, Gustavo Britto, and José Luis Oreiro (2015)
‘Real Exchange Rate and Economic Growth: New Empirical Evidence’, Metroeconomica
66(4): 686–714.
Myint, Hla (1971) Economic Theory and the Underdeveloped Countries. New York: Ox
ford University Press.
Nelson, Richard R. (1996) The Sources of Economic Growth. Cambridge, MA: Harvard
University Press.
Nelson, Richard R. (2019). ‘Development as an Evolutionary Economic Process’, in

Machiko Nissanke and José Antonio Ocampo (eds) The Palgrave Handbook of Develop
ment Economics: Critical Reflections on Globalisation and Development. London: Pal
grave Macmillan, pp. 229–45.
Nelson, Richard R. and Sidney G. Winter (1982) An Evolutionary Theory of Economic

Change. Cambridge, MA: The Belknap Press of Harvard University Press.
Ocampo, José Antonio (1986) ‘New Developments in Trade Theory and LDCs’, Journal of
Development Economics 22(1): 129–70.
Ocampo, José Antonio (2001) ‘Raul Prebisch and the Development Agenda at the Dawn of
the Twenty-first Century’, CEPAL Review 75: 23–37.
Ocampo, José Antonio (2016), ‘Balance-of-payments Dominance: Implications for Macro

economic Policy’, in Mario Damill, Martin Rapetti, and Guillermo Rozenwurcel (eds)
Macroeconomics and Development: Roberto Frenkel and the Economies of Latin
(p. 91)
America. New York: Columbia University Press, pp. 211–28.
Ocampo, José Antonio (2017a) Resetting the International Monetary (Non) System.
Oxford and Helsinki: Oxford University Press and UNU-WIDER.
Ocampo, José Antonio (2017b) ‘Dynamic Efficiency: Structural Dynamics and Structural
Change in Developing Countries’, in Akbar Noman and Joseph E. Stiglitz (eds) Efficiency,
Finance, and Varieties of Industrial Policy. New York: Columbia University Press, pp. 65–
102.
Ocampo, José Antonio, Codrina Rada, and Lance Taylor (2009) Growth and Policy in De
Ocampo, José Antonio and Lance Taylor (1998) ‘Trade Liberalisation in Developing
Economies: Modest Benefits but Problems with Productivity Growth, Macro Prices, and
Income Distribution’, Economic Journal 108(450): 1523–46.
Page 28 of 33
Palma, José Gabriel (2005) ‘Four Sources of “De-Industrialization” and a New Concept of
the “Dutch Disease”’, in José Antonio Ocampo (ed.) Beyond Reforms: Structural Dynamics
and Macroeconomic Vulnerability. Palo Alto, CA: Stanford University Press, ECLAC, and
World Bank, pp. 71–116.
Pérez, Carlota (2001) ‘Technological Change and Opportunities for Development as a

Moving Target’, CEPAL Review 75, Santiago.
Pérez, Carlota (2002) Technological Revolutions and Financial Capital: The Dynamics of
Bubbles and Golden Ages. Cheltenham: Edward Elgar.
Pérez, Carlota (2010) ‘Dinamismo tecnológico e inclusión social en América Latina: Una
estrategia de desarrollo productivo basada en los recursos naturales’, Revista de la
CEPAL 100: 123–45.
Pinto, Aníbal (1970) ‘Naturaleza e implicaciones de la “heterogeneidad estructural” de la

América Latina’, El Trimestre Económico 37(1), México, DF: Fondo de Cultura Económi
ca, January–March; reprinted in Cincuenta años del pensamiento en la CEPAL, vol. II,
Santiago: CEPAL/Fondo de Cultura Económica, 1998.
Prasad, Eswar S., Kenneth Rogoff, S. J. Wei, and Ayhan Kose, M. (2003) ‘Effects of Finan
cial Globalization on Developing Countries: Some Empirical Evidence’. Occasional Paper
No. 220. Washington, DC: International Monetary Fund.
Prebisch, Raúl (1951) ‘Theoretical and Practical Problems of Economic Growth’ (E/CN.
12/221). Mexico City: Economic Commission for Latin America (ECLA).
Prebisch, Raúl (1973) Interpretación del proceso de desarrollo latinoamericano en 1949.

2nd edition. Santiago: ECLAC, Serie conmemorativa del XXV aniversario de la CEPAL.
Pritchett, Lant (1997) ‘Divergence, Big Time’, Journal of Economic Perspectives 11(3): 3–
17.
Pritchett, Lant (2000) ‘Understanding Patterns of Economic Growth: Searching for Hills
among Plateaus, Mountains and Plains’, World Bank Economic Review 14(2): 221–50.
Rapetti, Martin (2013) ‘Macroeconomic Policy Coordination in a Competitive Real Ex

change Rate Strategy for Development’, Journal of Globalization and Development 3(2):
1–31.
Rapetti, Martin, Peter Skott, and Arslan Razmi (2012) ‘The Real Exchange Rate and Eco
nomic Growth: Are Developing Countries Different?’, International Review of Applied Eco
nomics 26(6): 735–53.
Razmi, Arslan, Martin Rapetti, and Peter Skott (2012) ‘The Real Exchange Rate and Eco
nomic Development’, Structural Change and Economic Dynamics 23(2): 151–69. (p. 92)
Page 29 of 33
Rodríguez, Francisco and Dani Rodrik (2001) ‘Trade Policy and Economic Growth: A
Skeptic’s Guide to the Cross-national Evidence’, in Ben S. Bernanke and Kenneth Rogoff
(eds) NBER Macroeconomics Annual 2000, Vol. 15. Cambridge: MIT Press, pp. 261–325.
Rodrik, Dani (1999) The New Global Economy and Developing Countries: Making Open
ness Work. Washington, DC: Overseas Development Council.
Rodrik, Dani (2007) One Economics, Many Recipes: Globalization, Institutions and Eco
nomic Growth. Princeton, NJ: Princeton University Press.
Rodrik, Dani (2008) ‘The Real Exchange Rate and Economic Growth’, Brookings Papers
on Economic Activity Fall: 365–412.
Rodrik, Dani (2014) ‘The Past, Present and Future of Economic Growth’, in Franklin Allen
et al. (eds) Toward a Better Global Economy. Oxford: Oxford University Press, pp. 70–139.
Rodrik, Dani (2016) ‘Premature Deindustrialization’, Journal of Economic Growth 21: 1–

33.
Rodrik, Dani and Andrés Velasco (2000) ‘Short-Term Capital Flows’, in Boris Pleskovic
and Joseph E. Stiglitz (eds) Annual World Bank Conference on Development Economics
1999. Washington, DC: World Bank.
Romer, Paul Michael (1986) ‘Increasing Returns and Long-run Growth’, Journal of Politi
cal Economy 94(5): 1002–37.
Ros, Jaime (2000) Development Theory and the Economics of Growth. Ann Arbor, MI: Uni
versity of Michigan Press.
Ros, Jaime (2013) Rethinking Economic Development, Growth, and Institutions. Oxford:
Rosenstein-Rodan, Paul N. (1943) ‘Problems of Industrialization of Eastern and South-

Eastern Europe’, The Economic Journal 53(210/211): 202–11.
Schumpeter, Joseph (1939) Business Cycles. New York: McGraw-Hill.
Schumpeter, Joseph (1961) The Theory of Economic Development. Oxford: Oxford Univer
sity Press.
Schumpeter, Joseph (1962) Capitalism, Socialism and Democracy. 3rd edition. New York:
Harper Torchbooks.
Singer, Hans W. (1950) ‘The Distribution of Gains between Investing and Borrowing Coun
tries’, American Economic Review, Papers and Proceedings 40(2): 473–85.
Solow, Robert M. (1956) ‘A Contribution to the Theory of Economic Growth’, Quarterly

Journal of Economics 70(5): 65–94.
Page 30 of 33
Solow, Robert M. (2000) Growth Theory: An Exposition. 2nd edition. New York: Oxford
University Press.
Stiglitz, Joseph E. and Bruce Greenwald (2014) Creating a Learning Society: A New Ap
proach to Growth, Development, and Social Progress. New York: Columbia University
Press.
Taylor, Lance (1991) Income Distribution, Inflation, and Growth. Lectures on Structuralist
Macroeconomic Theory. Cambridge, MA: MIT Press.
Taylor, Lance (1994) ‘Gap Models’, Journal of Development Economics 45: 17–34.
Thirlwall, Anthony Philip (2011a) The Economics of Development: Theory and Evidence.
9th edition. London: Macmillan.
Thirlwall, Anthony Philip (2011b) ‘Balance of Payments Constrained Growth Models: His
tory and Overview’, PSL Quarterly Review 64(259): 307–51.
van Wijnbergen, Sweder (1984) ‘The Dutch Disease: A Disease after All?’, Economic Jour
nal 94(373): 41–55.
Notes:
(1) This concept is borrowed from Ocampo (2017b). Note that it is entirely different from
that of ‘dynamic efficiency’ used in neoclassical optimal growth models.
(2) See, among an extensive literature, Prebisch (1951, 1973), Furtado (1961), Nelson and
Winter (1982), Dosi et al. (1988), Wade (1990), Taylor (1991), Chang (1994), Nelson
(1996, 2019), Aghion and Howitt (1998), Rodrik (1999, 2007), Ros (2000, 2013), Amsden
(2001), Ocampo et al. (2009), Lin (2012), Stiglitz and Greenwald (2014), and Cherif and
Hasanov (2019).
(3) See the classical contributions by Romer (1986), Lucas (1988), and Barro and Sala-i-
Martin (2003).
(4) See in this regard the differentiation between ‘proximate’ and ‘ultimate’ causality of
growth processes by the economic historian Maddison (1991: ch. 1).
(5) See, for example, Easterly et al. (1993) and Pritchett (2000).
(6) Nonetheless, it has also been argued that there is much less association between some
of these variables and economic growth than was traditionally assumed. This has been
claimed particularly in relation to physical and human capital. See Easterly (2001: part
II).
Page 31 of 33
(7) The development that took place in Japan after the Meiji Restoration, and the escala
tion of this country to the top group of developed countries in the post-Second World War
years should be included as successful convergence processes.
(8) In relation to Latin America, see Bértola and Ocampo (2012: ch. 1).
(9) The lasting effects of the debt crises of the 1980s in Africa and Latin America are the
most telling example in this regard, but that of several peripheral European countries
(notably Greece) after the 2007‒9 North Atlantic financial crisis has similar features (I
use this term rather than the more commonly used ‘global financial crisis’, because al
though the crisis had global effects, it centred in the United States and Western Europe).
(10) Interestingly, Kaldor (1978: ch. 4) and Cripps and Tarling (1973) show the importance
that this process also had during the post-Second World War golden years in industrial
countries.
(11) See the seminal analysis of this issue in Krugman (1990), Grossman and Helpman
(1991), and, in relation to developing countries, Ocampo (1986).
(12) See Rodrik (2008), Rapetti et al. (2012), Razmi et al. (2012), Rapetti (2013), and for a
review of the literature, Frenkel and Rapetti (2014) and Missio et al. (2015).
(13) See Nelson and Winter (1982), Nelson (1996), and Dosi et al. (1988) and, with re
spect to developing countries, Katz (1987), Lall (1990, 2003), and Lee (2019). Similar con
cepts have been developed in some versions of the new neoclassical growth theory, in
which ‘knowledge capital’ is a form of ‘human capital’ with three specific attributes: it is
‘embodied’ in particular persons, it is capable of generating significant externalities, and
it is costly to acquire (Lucas, 1988).
(14) See OECD, Trade in value added database, https://www.oecd.org/sti/ind/measuring-

trade-in-value-added.htm#access.
(15) Now, of course, being challenged by US protectionist policies and the coronavirus
pandemic.
(16) See also the October 2019 issue of the Review of Keynesian Economics in honour of
Thirlwall. There may also be savings constraints. For a full analysis of the gaps in macro
economic adjustment that may be reflected in the GG curve, see Taylor (1994).
(17) Under significant initial labour underemployment or underutilization of other re

sources, the slope of TT could be steeper than that of GG, generating an unstable equilib
rium at A. In this case, any displacement from saddle point A will lead the economy into
explosive virtuous or vicious growth processes.
(18) See Chapter 28 in this volume.
(19) See a review of the literature in Erten and Ocampo (2013), the conclusions of which
are summarized in the next paragraph.
Page 32 of 33
(20) This choice is part of a broader dilemma of how much to save or invest out of com
modity booms. For an analysis of this choice in oil economies, see Cherif and Hasanov
(2013).
(21) One interesting analysis is the comparative history of Scandinavian vs. Latin Ameri
can historical development, in the essays collected in Blomström and Meller (1991).
(22) On the ‘Dutch disease’, see, among many others, Corden and Neary (1982), van Wijn
bergen (1984), and Krugman (1990: ch. 7).
(23) As the IMF (2011: ch. 4) has shown, there is evidence that FDI has also become more
volatile, largely because it has become partly financialized. This relates to greater use by
multinationals of intra-corporate and other international loans to fund subsidiaries, as
well as derivatives, both to hedge their exposure, but also to speculate on currencies.
(24) I prefer the term ‘capital account regulations’ to ‘controls’, because most are not di
rect regulation and are rather similar to other prudential regulations.
(25) See, for example, the work of the World Bank economists Luna-Martinez and Vicente
(2012).
(26) See also on these issues Hausmann and Rodrik (2003).
José Antonio Ocampo is professor at Columbia University’s School of International

and Public Affairs, and chair of the Committee for Development Policy of the United
Nations Economic and Social Council (ECOSOC). He has occupied numerous posi
tions at the United Nations and his native Colombia, including UN under-secretary-
general for economic and social affairs, executive secretary of the UN Economic
Commission for Latin America and the Caribbean (ECLAC), and member of the board
of directors of Banco de la República (Colombia’s central bank), minister of finance,
minister of agriculture and director of the National Planning Office of Colombia.
Page 33 of 33
Industrial Policies, Patterns of Learning, and Development: An Evolutionary
Perspective
Industrial Policies, Patterns of Learning, and Develop

ment: An Evolutionary Perspective
Mario Cimoli, Giovanni Dosi, and Xiaodan Yu
This work discusses the role of industrial policies within an evolutionary view of innova
tion and learning as drivers of economic development. Building on the notions of techno
logical paradigms and trajectories, it links the processes of catching up with the dynam
ics of capability accumulation within and across firms. Technological paradigms provide a
structure to the knowledge embodied in individual and, even more importantly, in organi
zational problem-solving procedures and routines, which collectively define the techno
logical and organizational capabilities of firms and whole countries. The improvement of
such capabilities is at the core of the process of development and catching up, well be
yond sheer capital accumulation. In turn, such processes are embedded in broader na
tional systems of innovation in which industrial policies play a pivotal role.
Keywords: technological paradigms, catching up, theory of production, absolute and comparative advantages, na
tional systems of innovation, industrial policies, economic evolution and development
4.1 Introduction
IT is now generally acknowledged by both economists and economic historians that there
is a strong relationship between technical change and economic development. However,
the precise causal links are still a matter of debate. Although it is quite intuitive that im
provements in the efficiency of production techniques and in product performances may
be a determinant or at least a binding precondition of growth in per capita incomes and
consumption, ‘what ultimately determines what’ remains debatable. Is it resource accu
mulation that primarily fosters the exploration of novel innovative opportunities, or, con
versely, does innovation drive capital accumulation? Do new technological opportunities
emerge mainly from an extra-economic domain (‘pure science’) or are they primarily dri
ven by economic incentives? Should one assume that the institutions supporting technical
change are sufficiently adaptive to adjust to the dynamics of market interactions, or, con
Page 1 of 64
Perspective
versely, are they inertial enough to shape the rates and directions of innovation and diffu
sion?
These are obviously quite complex questions. However, the last four decades have seen a
profusion of studies on the sources, mechanisms, and patterns of technological innova
tion, diffusion, and imitation. And the opening of the technological ‘black box’ has often
gone hand in hand with important insights into innovation-driven market competition.
Business historians have finally achieved some cross-fertilization with (some breeds of)
economic theorizing. And institutional understanding of the socio- (p. 94) economic fabric
of contemporary societies is starting to show fruitful complementarities with economic
analyses. A number of these contributions have been made by evolutionist or institution
alist scholars. Indeed, the common threads linking these diverse streams of research
highlight the co-evolution of technologies, corporate organizations, and institutions (more
in Freeman and Louca, 2001; Freeman, 2019; and Reinert, 2007; but a few of the intu
itions are already in the classics of development theory: cf. for example, Gerschenkron,
1962; Hirschman, 1958). These threads—linking evolutionary analyses of the microeco
nomics of innovation and learning all the way to generalizations on some invariant fea
tures of the process of development—are the subject of this contribution. It does not
claim to be a comprehensive survey, but rather a sort of ‘roadmap’.
We start by discussing the theoretical implications of what we know about the dynamics
of innovative activities at micro and sectoral levels. Technical change is structured by
technological paradigms and follows relatively ordered trajectories. In such a view,
knowledge accumulation plays a central role. This view has major implications also for
the theory of production. There are firms and countries that are simply ‘better’, that is,
more efficient and more innovative, than others, irrespective of relative prices. This im
plies that technological asymmetries or gaps are permanent features across firms and,
even more so, countries.
Technical change can be viewed as an evolutionary process with invariances and speci
ficities in patterns of change at sectoral and national level, which in turn can be interpret
ed in terms of some underlying features of the processes of collective learning, market se
lection, and institutional governance.
4.2 The Fundamental Properties of Technology

4.2.1 Technological Paradigms and Trajectories
A variety of concepts have been put forward in recent decades to define the nature of in
novative activities: technological regimes, paradigms, trajectories, salients, guideposts,
dominant designs, and so on. The names are not so important (although some standard
ization could make the diffusion of ideas easier). More crucially, these are overlapping
concepts seeking to capture a few common features of the procedures and direction of
Page 2 of 64
Perspective
technical change (for discussions and references, see Dosi, 1988; Dosi and Nelson, 2010).
Let us consider some of them.
The notion of technological paradigm is based on a view of technology grounded on three

fundamental ideas. First, any satisfactory description of what technology is and how it
changes must also embody the representation of the specific forms of knowledge (p. 95)
on which a particular activity is based. In other words, technology cannot be reduced to
the standard view of a set of well-defined blueprints; rather, it concerns problem-solving
activities and also involves tacit forms of knowledge embodied in individuals and organi
zational procedures.
Second, paradigms entail specific heuristics and visions of ‘how to do things’ and how to
improve them, often shared by the community of practitioners in each particular activity
(engineers, firms, technical societies, etc.) that share common cognitive frames (Con
stant, 1985).
Third, paradigms generally also define basic models of artefacts and systems, which over
time are progressively modified and improved. These basic artefacts can also be de
scribed in terms of some fundamental technological and economic characteristics For ex
ample, the basic attributes of an airplane are described in terms not only of inputs and
production costs, but also on the basis of some salient technological features such as
wing-load, take-off weight, speed, distance it can cover, etc. What is interesting is that
technical progress seems to display patterns and invariances in terms of these product
characteristics. Similar examples of technological invariances can be found, for example,
in semiconductors, agricultural equipment, automobiles, and a few other micro-techno
logical studies.
Technological trajectories relate to the progressive realization of the innovative opportu

nities associated with each paradigm which can in principle be measured in terms of the
changes in the fundamental techno-economic characteristics of artefacts and the produc
tion process. The core ideas involved in this notion of trajectories are the following.
First, each particular body of knowledge (i.e. each paradigm) shapes and constrains the
rates and direction of technological change irrespective of market inducements. Second,
as a consequence, regularities and invariances can be seen in the patterns of technical
change that hold under different market conditions and whose disruption is correlated
with radical changes in knowledge bases (in paradigms). Third, technical change is partly
driven by repeated attempts to cope with the technological imbalances which it itself cre
ates.
It is now widely acknowledged in the innovation literature that learning is local and cu
mulative. Local means that the exploration and development of new techniques is likely to
occur in the neighbourhood of the techniques already in use. Cumulative means that cur
rent technological development—at least at the level of individual business units—often
builds upon past experiences of production and innovation, and it proceeds via specific
problem-solving sequences (Vincenti, 1992). Clearly, this is consistent with the ideas of
Page 3 of 64
Perspective
paradigmatic knowledge and the ensuing trajectories. A crucial implication, however, is
that at any point in time the agents involved in a particular production activity will have
little scope for substitution among techniques, if by that we mean the easy availability of
blueprints different from those actually in use, which could be put efficiently into opera
tion according to relative input prices.
(p. 96) 4.2.2 Paradigms, Routines, Organizations
A locus classicus in the analysis of the profound intertwining between technological

learning and organizational change is certainly Alfred Chandler’s reconstruction of the
origins of the modern multi-divisional (the M-form) corporation and its ensuing effects on
the American competitive leadership over several decades (Chandler, 1990, 1992a, 1993).
And, as Chandler himself has argued, there are strict links between story and evolution
ary theories (Chandler, 1992b). While it is not possible to enter into the richness of the
Chandlerian analysis here, let us just recall one of the main messages:
It was the institutionalizing of the learning involved in product and process devel
opment that gave established managerial firms advantages over start-ups in the
commercialization of technological innovations. Development remained a simple
process involving a wide variety of usually highly product-specific skills, experi
ence and information. It required a close interaction between functional special
ists, such as designers, engineers, production managers, marketers and man
agers…Such individuals had to coordinate their activities, particularly during the
scale-up processes and the initial introduction of the new products on the mar
ket…Existing firms with established core lines had retained earnings as a source
of inexpensive capital and often had specialized organizational and technical com
petence not available to new entrepreneurial firms.
(Chandler, 1993: 37)
These organizational dynamics can be interpreted as an evolutionary story of competence

accumulation and development of specific organizational routines (Chandler, 1992b). Did
seemingly superior organizational forms spread evenly throughout the world? The Chand
lerian enterprise diffused, albeit slowly, through other OECD countries (Chandler, 1990;
Kogut, 1992). However, organizational forms, strategies, and control methods have devel
oped differently from nation to nation, because of the difference in national environments
(Chandler, 1992a: 283). Moreover, the diffusion of the archetypical M-form corporation
has been limited to around half a dozen already developed countries (and even in coun
tries like Italy, it involved very few companies). Similar differences can be found in the
processes of international diffusion of American principles of work organization—for ex
ample, Taylorism and Fordism—(for an analysis of the Japanese case, see Coriat, 1990). A
growing literature looks at the early corporate histories of the German, Japanese, or Ital
ian systems of production which continue to influence contemporary forms of organiza
tion and learning (see Chandler, 1990; Coriat, 1990; Kogut, 1993; Dornseifer and Kocka,
1993; Dosi, Giannetti, and Toninelli, 1993).
Page 4 of 64
Perspective
Different initial organizational conditions, as well as different patterns of learning, can be
seen across developing countries. The last three decades have witnessed increased tech
nological dynamics in some of them (in primis China), with a subsequent development of
the ‘modern’ industrial structures and an impressive technological catching up.
The evolutionary path of technological learning involves both the capacity to acquire
technologies (capital goods, know-how, etc.) and the capability to absorb these technolo
gies and adopt them to the local conditions. Microeconomic/micro-technological (p. 97)
evidence highlighting the mechanisms which stimulate and limit endogenous learning in
the NIEs suggest the existence of some characteristics in the paths of technological
learning at the firm level (see also Cimoli, 1990; Cimoli and Dosi, 1988). In particular, one
might be able to identify some relatively invariant sequences in the learning processes,
conditional on the initial organizational characteristics of the firms and the sectors of
principal activity.
A first set of regularities regards the varying combinations between acquisition of outside
technologies and endogenous learning. As is well known, the transfer of technology to de
veloping economies is a common source for the subsequent development of learning ca
pabilities at the firm and sectoral levels. So, Amsden and Hikino identify the ability to ac
quire foreign technology as a central characteristic ‘of late industrialization at the core of
which is borrowing technology that has already been developed by firms in more ad
vanced countries. Whereas a driving force behind the First and Second Industrial Revolu
tions was the innovation of radically new products and processes, no major technological
breakthrough has been associated with late-industrializing economies. The imperative to
learn from others, and then realize lower costs, higher productivity and better quality in
mid-tech industries by means of incremental improvements, has given otherwise diverse
twentieth-century industrializers a common set of properties’ (Amsden and Hikino 1993:
37).
4.3 Technological Dominance, Micro-Hetero

geneity, and Non-substitution
The notion of paradigm contains elements of both a theory of production and a theory of
innovation. Loosely speaking, we should consider such a theory at the same level of ab
straction as, say, a Cobb-Douglas production function or a production possibility set. That
is, all of them are theories of what are deemed to be some stylized but fundamental fea
tures of technology and, relatedly, of production processes.
In fact, one finds a few remarkable assumptions underlying conventional production theo
ries. As already mentioned, technologies—at least in a first approximation—are conven
tionally seen as a set of blueprints describing alternative input combinations. Moreover,
at any one time there must be many of them, in order to be able to interpret empirical ob
servations as the outcome of a microeconomic process of optimal adjustment to relative
prices. Information about these blueprints is generally assumed to be freely available (un
Page 5 of 64
Perspective
less appropriated through the patent system). Finally, it is assumed that activities leading
to the efficient exploitation of existing blueprints can be separated from those leading to
the development of new ones (exogeneity of technical progress is its extreme version). Of
course, this is only a trivialized account of a family of models that can be made much
more sophisticated, by, for example, adding details (p. 98) on how blueprints are ordered
with respect to each other (more technically, issues like continuity and convexity come
under this heading). However, it still seems fair to say that the basic vision of production
—also carried over in aggregate growth and development models—focuses on questions
of choice among well-defined techniques, generally available to all producers, who also
know perfectly well what to do with all the recipes when they see them.
The theory of production based on paradigms, however, is based on an opposing set of

theoretical building blocks, many of which yield empirically testable hypotheses. Al
though it undertakes the same interpretations at the same level of generality, it is more in
tune with the microeconomic evidence and is directly linked to theories of innovation.
The evolutionary theory makes the following predictions:
a) At any point in time there is one or very few best-practice techniques which domi
nate the others irrespective of relative prices.
b) Different agents are characterized by persistently diverse (better and worse) tech
niques.
c) Over time the observed aggregate dynamics of technical coefficients in each par
ticular activity is the joint outcome of the process of imitation/diffusion of existing
best-practice techniques, the search for new ones and of market selection among
heterogeneous agents.
d) Changes over time of the best-practice techniques themselves highlight regular
paths (i.e. trajectories) both in the space of input coefficients and also in the space of
the core technical characteristics of outputs.
Catching-up in the productivity distribution is the first fundamental mark of successful

catching-up processes more generally (more in Malerba and Nelson, 2011; Lee and
Malerba, 2017; Landini and Malerba, 2017). The striking success of China is a case in
point.
Table 4.1 shows the dramatic labour productivity growth in incumbent manufacturing
firms in China. The overall productivity of incumbents grew at 9.98 per cent per annum
between 1998 and 2007. All sectors display positive productivity growth rates, (except
petroleum refining, which had negative growth during the 1998–2002 period).
Further, note the remarkable differences in productivity growth across sectors, as such
circumstantial evidence of significant inter-sectoral differences in absorptive capacities
(Cohen and Levinthal, 1989) of ‘frontier’, generally foreign, technologies, and of corre
sponding differences in the average catching-up rates. Figure 4.1 offers three snapshots
of the non-parametric kernel density distribution of labour productivity in China, com
pared with Italy and France, illustrating the overall technology gap with two higher-in
Page 6 of 64
Perspective
come countries.1 At a first glance readers might find such a comparison as somewhat far-
fetched if one has in mind a ‘world production function’, (p. 99) possibly multiplied by
some country-specific scalar. After all, Chinese wages have been/are at least an order of
magnitude lower than Italian and French ones. As a consequence, one would expect to
see the three countries on very different positions on such production functions. But is it
really the case? If it were so, one would also expect, first, major differences between Chi
na, on the one hand, and Italy and France, on the other, in capital/output ratios—the ap
propriate proxy for ‘capital intensities’ when (p. 100) ‘ (p. 101) production functions’ differ.
And of course one should expect strong correlations between labour productivities and
capital/labour ratios within each country and within each sector.
Premise to the following discussion: the proxies for capital are very noisy on Italian and
French data and just more so on the Chinese data.2
Remarkably, what the evidence suggests is rather at odds with the conventional wisdom.
First, capital/output ratios also at sectoral levels do not differ very much between China
and the two European countries considered (Table 4.2). Indeed they tend to be higher in
China. Second, the within-country, within-sector micro correlations between labour pro
ductivities (VA/L) and capital/output ratios (K/VA), for whatever proxy for K is used, is ro
bustly negative in China and is mildly negative in Italy and France (statistics available up
on request). In other words, labour and capital productivity are strongly positively corre
lated. Indeed, conventional theories suggest that, given uniform relative prices, one
should not expect distribution of productivities at all. However, they are persistently
there even in developed countries (more in Syverson, 2011; Dosi and Grazzi, 2006) and
much more so in developing ones. Third, even within China, labour productivities and
capital/labour ratios—as a proxy of degrees of production mechanization/automation—are
basically orthogonal (see Figure 4.2 for a sector illustration).
Overall, the evidence suggests that very little action comes from ‘moving along isoquants’
in response to relative prices. Rather, ‘best practice’ techniques involve a more efficient
use of both labour and capital, and relatedly, catching-up fundamentally involves improve
ments on both dimensions. It is a world of complementarities rather than substitution, in
which technology-gaps and learning efforts are both reflected by labour productivity dif
ferences, quite independently from relative prices, while TFP proxies might well yield a
quite distorted picture of the development process. Indeed, given the ubiquitous comple
mentarities between labour and capital, labour productivities alone turn out to be a ro
bust proxy for the lower bound of ‘true’ efficiency distributions within countries, but also
across countries, with the added advantage of avoiding any explicit or implicit hypotheses
on interfactor substitutability and capital measurements.
Page 7 of 64
Industrial Policies, Patterns of Learning, and Development: An Evolutionary Perspective
Table 4.1 Annual growth rate of labour productivity amongst firms over 1998–2007, and subperiods 1998–2002 and
2002–2007 amongst ‘continuing’ firms (i.e. firms remaining in the same two-digit sector over the relevant period)
Page 8 of 64
PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individ
ual user may print out a PDF of a single chapter of a title in Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).
CIC Sector 1998–2007 1998–2002 2002–07
13 Food processing of 11.25 8.55 13.46

agricultural prod
ucts
14 Other foodstuffs 8.87 5.73 11.22
15 Beverages 10.63 6.20 12.80
16 Tobacco 15.29 11.08 10.65
17 Textiles 9.79 10.14 10.54
18 Garments, footwear, 7.84 4.84 10.57

etc.
19 Leather, fur, feath 7.55 6.52 10.29

er, etc.
20 Processing of tim 11.87 7.61 14.43

ber, manuf. of wood,
bamboo, etc.
21 Furniture 6.19 4.82 10.40
Page 9 of 64
22 Paper and paper 10.33 9.48 11.75

products
23 Printing, reproduc 8.92 7.54 8.17

tion of recording
media
24 Articles for culture, 8.39 7.06 10.03

education, and
sports
25 Processing of petro 3.77 -1.61 6.36

leum, cokeries, nu
clear fuel
26 Raw chemical mate 11.40 10.44 11.85

rials and chemical
products
27 Pharmaceuticals 8.94 10.64 7.53
28 Chemical fibres 10.31 12.05 8.85
29 Rubber 8.80 7.01 9.39
30 Plastics 5.83 6.43 6.24
Page 10 of 64
31 Non-metallic miner 12.76 9.86 14.71

al products
32 Smelting and pro 13.86 12.68 14.14

cessing of ferrous
metals
33 Smelting and pro 12.45 13.73 12.64

cessing of non-fer
rous metals
34 Metal products 5.44 7.84 4.32
35 General purpose 15.40 13.76 15.72

machinery
36 Special purpose ma 16.23 13.09 15.13

chinery
37 Transport equip 12.67 11.64 13.05

ment
39 Electrical machin 9.51 8.84 9.32

ery and equipment
Page 11 of 64
40 Communication 5.64 8.37 3.30

equipment, comput
ers, etc.
41 Measuring instru 9.62 9.46 9.57

ments and machin
ery
42 Artwork and other 10.01 9.32 12.70
Average 9.98 8.73 10.66
Source: Yu et al. (2015).
Page 12 of 64
Perspective
FIGURE 4.1 Empirical density (Pr, vertical axis) of

labour productivities, whole manufacturing of China,
France, and Italy, 1998, 2002, and 2006
Note: The first row: constant 2000 prices and ex

change rates (IMF source); the second row: PPP ad
justed price (World Bank source).
Page 13 of 64
Table 4.2 Median capital intensity (capital/output ratios) by sector, China, Italy, and France, 1998, 2002, and 2006
NACE Sector China Italy Franc China Italy Franc China Italy Franc
e e ea
1998 2002 2006
173 Finish 2.772 1.971 1.863 0.732 0.755 0.694 1.228 1.512 1.546
ing of
textiles
175 Car 1.672 1.327 0.789 0.752 0.775 0.688 0.891 0.987 1.055
pets,
rugs,
and
other
textiles
182 Appar 1.052 0.785 0.620 0.268 0.276 0.226 0.318 0.318 0.336
el
193 Footwe 1.062 0.885 0.529 0.29 0.331 0.288 0.488 0.631 0.645
ar
Page 14 of 64
203 Wood 1.477 0.954 0.629 0.728 0.734 0.763 0.773 0.744 0.736
prod
ucts
for
con
struc
tion
212 Paper 1.475 1.367 1.123 0.824 0.901 0.988 1.025 1.206 1.217
and
paper
board
221 Pub 3.873 5.250 5.716 0.259 0.19 0.117 0.229 0.204 0.192
lishing
222 Print 2.559 2.456 2.084 0.508 0.566 0.562 0.700 0.797 0.792
ing
241 Pro 2.547 1.784 1.049 0.977 1.045 1.153 2.081 2.443 2.811
duc
tion of
basic
chemi
cals
Page 15 of 64
243 Paints, 1.312 1.086 0.852 0.584 0.544 0.57 0.946 0.936 1.052
var
nishes,
inks,
mas
tics
244 Phar 1.707 1.514 1.508 0.57 0.623 0.656 0.666 0.83 0.837
ma.,
med.
chemi
cals,
botani
cal
prod
ucts
246 Other 1.436 1.167 0.707 0.588 0.628 0.636 0.973 1.004 1.072
chemi
cal
prod
ucts
251 Rub 1.587 1.479 0.974 0.514 0.588 0.495 0.951 1.088 1.03
ber
prod
ucts
Page 16 of 64
252 Plastic 1.614 1.394 1.055 0.714 0.795 0.818 0.969 0.991 1.035
prod
ucts
261 Glass 1.696 1.442 1.079 0.579 0.594 0.742 0.996 1.169 1.198
and
glass
prod
ucts
266 Con 2.084 1.643 1.676 0.93 0.847 0.965 1.365 1.399 1.253
crete,
plas
ter,
and ce
ment
275 Cast 1.113 0.937 0.698 0.669 0.815 0.734 0.886 1.127 1.128
ing of
metals
281 Struc 1.290 1.176 0.870 0.433 0.481 0.455 0.547 0.505 0.569
tural
metal
prod
ucts
Page 17 of 64
284 Forg 1.981 1.289 0.820 0.574 0.695 0.618 0.77 0.913 0.967
ing,
press
ing,
stamp
ing of
metal
285 Treat 1.113 0.980 0.923 0.452 0.515 0.467 0.673 0.762 0.803
ment
and
coat
ing of
metals
286 Cut 1.554 1.068 0.940 0.471 0.584 0.559 0.734 0.861 0.892
lery,
tools,
and
gener
al
hard
ware
Page 18 of 64
287 Other 1.337 1.018 0.788 0.586 0.626 0.566 0.818 0.921 0.871
fabri
cated
metal
prod
ucts
291 Ma 2.041 1.524 1.012 0.48 0.491 0.408 0.674 0.76 0.714
chin
ery for
prod.
use of
mech.
power
292 Other 1.756 1.321 0.905 0.323 0.315 0.272 0.372 0.364 0.361
gener
al pur
pose
ma
chin
ery
294 Ma 2.530 1.669 0.961 0.343 0.391 0.289 0.425 0.465 0.466
chine
tools
Page 19 of 64
295 Other 2.177 1.486 0.977 0.358 0.337 0.332 0.520 0.585 0.614
special
pur
pose
ma
chin
ery
311 Elec 1.570 1.200 0.767 0.369 0.452 0.397 0.510 0.526 0.501
tric
mo
tors,
gener
ators,
and
trans
form
ers
312 Manuf. 1.409 1.127 0.781 0.335 0.453 0.352 0.640 0.648 0.553
of elec
tricity
distrib
ution
and
control
equip
Page 20 of 64
316 Electri 1.163 0.863 0.671 0.299 0.288 0.275 0.498 0.516 0.497
cal
equip
ment
not e/
where
classi
fied
343 Parts 1.781 1.336 1.094 0.526 0.63 0.534 1.088 1.311 1.185
for mo
tor ve
hicles
and
their
en
gines
361 Furni 1.293 1.092 0.798 0.593 0.61 0.564 0.633 0.674 0.722
ture
366 Manu 0.793 0.830 0.808 0.467 0.485 0.405 0.576 0.669 0.781
factur
ing
n.e.c.
Mean 1.713 1.419 1.127 0.534 0.574 0.550 0.780 0.871 0.888
Page 21 of 64
Medi 1.578 1.305 0.914 0.520 0.586 0.561 0.717 0.814 0.820
an
Note: aData for France in the last column are for 2004.
Sources: Yu et al. (2015), CMM, INSEE (on France), and ISTAT-Micro 3 (on Italy).
Page 22 of 64
Perspective
FIGURE 4.2 Scatterplot of log (VA/L) versus log (K/

L) for corn milling sector (CIC 131), 1998
Note: OLS regression: coefficient = 0.038 (standard

error 0.029), R2 = 0.0006, number of observations =
2,838.
In fact, the empirical elasticities of substitutions implied by the negative micro relation
between labour productivities and capital/output ratios (i.e. positive correlations between
labour and capital productivities) are positive in sign: the isoquants do not look like stan
dard isoquants but are more similar to rays out of the origin.
Granted all that, let us now focus on the micro picture offered by the data and its dynam
ics. First, note the different upper bounds of the three country distributions, as (p. 102)
(p. 103) (p. 104) such an impressionistic proxy of different inter-country lags and leads (to
gether of course with different sectoral compositions of output).
Second, the width of the support of the distribution of China is much larger, revealing
much greater technological asymmetries across Chinese firms. The dynamics of catching-
up in China’s manufacturing productivity are associated with (i) a rightward movement of
the mean of the distributions; (ii) a corresponding rightward movement of the support;
and (iii) as we shall analyse in more detail below, a shrinking of the support itself. Labour
productivity distribution is asymmetric and left-skewed. The evolving pattern of the left-
tail and that of the right-tail are also different, with a significant left-tail shift towards
higher levels of productivity, compared with a relatively mild movement of the right tail.
These dynamics match what, in the old development literature, was called a ‘reduction of
the dualistic structure economy’ consisting of a shrinking traditional/relatively backward
part of manufacturing and an expanding ‘modern’ one which, however, is only just begin
ning to push ‘frontier technologies’ further.
An important piece of evidence on intra-sectoral asymmetries in efficiency and their

changes over time is the top-to-bottom ratio of labour productivities. Table 4.3 displays
the ratio of the 9th decile over the 2nd decile for each sector from 1998 to 2007. The ra
Page 23 of 64
Perspective
tios decrease in most sectors, indicating a reduction of productivity dispersion, plausibly
due both to learning by laggard firms and selection (exit) of the worst performers. The ra
tios are generally lower in ‘traditional’ ones (CIC 17–24 including textile, garments,
leather, furniture, paper manufacturing, etc.) and higher in relatively technology-inten
sive sectors (e.g. transport equipment, electrical machinery, and (p. 105) communication
equipment). The ratios drop more rapidly in the first part of the period under considera
tion, which is also a period of retreat by SOEs from the so-called ‘competitive sectors’. At
the same time, the ratios in several ‘heavy industries’ such as petroleum refining and
non-ferrous metals sectors grows, hinting at some sort of persistent ‘dualism’ within
these industries (note that growing intra-sectoral asymmetries can and often do go hand
in hand with high average growth rates). How much of the dynamics in overall productivi
ty distribution is due to inter-sectoral relocation of production?
Page 24 of 64
Table 4.3 Ratio of the average labour productivity of the second highest decile over the second lowest decile, China,
1998, 2002, and 2007
Page 25 of 64
CIC Sector 1998 2002 2007

agricultural prod
ucts
15 Beverages 14.89 11.82 9.06
16 Tobacco 17.05 22.95 26.44
17 Textiles 8.61 7.01 6.07
18 Garments, footwear 6.51 5.45 5.42

etc.
19 Leather, fur, feather 7.80 7.17 6.80

etc.

bamboo etc.
21 Furniture 9.29 7.15 6.93
Page 26 of 64
22 Paper and paper 7.44 6.16 6.27

products

tion of recording
media

education, and
sports
25 Processing of petro 8.82 12.26 11.23

leum, cokeries, nu
clear fuel

rials and chemical
products
29 Rubber 6.56 7.49 7.42
30 Plastics 8.65 7.18 7.02
Page 27 of 64

al products
32 Smelting and press 9.57 8.58 8.40

ing of ferrous met
als

ing of non-ferrous
metals
35 General purpose 8.77 6.68 6.56

machinery

chinery
37 Transport equip 11.69 8.19 7.09

ment

ery and equipment
Page 28 of 64
40 Communication 13.52 11.08 8.36

equipment, comput
ers etc.

ments and machin
ery
Page 29 of 64
Perspective
Table 4.4 displays the time series of value-added shares of each two-digit sector in overall
manufacturing. It is remarkable that relatively little structural change has (p. 106) oc
curred over the period under investigation, even if indeed in the ‘right direction’. So, for
example, the shares of transport equipment, electrical machinery and equipment, and
communication equipment, computers etc. are among the highest from the start of the
period under consideration, and their total share just increases from 20.7 per cent in
1998 to 22.5 per cent in 2007. A synthetic view of the relative importance of the within-
vs. between-sectors contributions to productivity growth is presented in Table 4.5 (for de
tails on the shift-and-share decomposition method of productivity growth, see Appendix).
Of course, the precise relative measures of sector-specific learning vs. structural change
(what nowadays is often referred to as ‘re-allocation’) depend a great deal on (p. 107)
(p. 108) (p. 109) the techniques of measurement (e.g. whether the sectoral weights are in
terms of employment or value added). So, for example, Paus (2019) finds a contribution of
the latter of around 19 per cent. However, no matter the measure, the ‘within component’
dominates—a sign indeed that China achieves quite early a ‘modern’ industrial structure.
However, as we shall discuss later, this is an exception in the overall picture of catching-
up experiences.
Interestingly, this evidence seems to contradict Kuznets’ view of increasing productivity

due to a large extent to structural change, that is, movements from low-productivity sec
tors to high-productivity ones also within manufacturing. On the contrary, our evidence
suggests that, unlike what happened in the 1980s (cf. Wang and Szirmai, 2008; see also
Akkemik, 2005), the movement of the overall manufacturing means is mainly due to sec
tor-specific dynamics. Incidentally, note that ‘virtuous’ structural change is by no means
automatic or inevitable. Indeed, the apparent failure to undertake that path appears to be
at the heart of the middle-income trap which, for example, Latin American countries have
experienced: more in Paus (2019).
Of course, the relative stability of sectoral shares at the two-digit sectoral level does not
rule out much more turbulence at finer levels of disaggregation within each two-digit sec
tor: indeed, there is very intensive ‘micro structural change’. However, the evidence
marks a difference from other episodes of industrialization and catching-up, in that in the
period of our observations, China appears to be already quite mature in terms of its broad
manufacturing structure. For example, when South Korea had the same real per capita in
come that China had in 1998, which was in 1973 (Maddison’s historical statistics,
www.ggdc.net/maddison/oriindex.htm), the share of around 22 per cent of textile and
clothing in total manufacturing was around 22 per cent (World Development Indicators
database), compared to a 1998 Chinese share of 12 per cent. In the literature a quite
common claim is that export and productivity growth go together (possibly with causality
running in both directions).
China displays a dramatic rise in the share of exports in total manufacturing output, cou
pled with a dramatic growth in productivity. However, the case of China lends little sup
port to the notion of ‘learning by exporting’.
Page 30 of 64
Perspective
Figure 4.3 shows the labour productivity distribution of exporters and non-exporters for
the years 1998 and 2007 in selected sectors (chemical, electrical machinery, and commu
nication equipment), which illustrates a more general pattern. Note that in 1998 ex
porters have a higher level of productivity and their support of distribution is narrower
than that of non-exporters. However, a significant catch-up by non-exporters takes place,
so that in 2007, exporters and non-exporters have similar productivity distributions and
similar widths of support.3
Page 31 of 64
Table 4.4 Contribution of each two-digit sector to total manufacturing value added, China, 1998, 2002, and 2007
(percentages)
Page 32 of 64
CIC Sector 1998 2002 2007

agricultural prod
ucts
15 Beverages 3.51 2.69 2.01
16 Tobacco 5.70 5.13 3.11
17 Textiles 6.39 5.81 5.23
18 Garments, footwear 3.02 2.76 2.41

etc.
19 Leather, fur, feather 1.78 1.76 1.58

etc.

bamboo etc.
21 Furniture 0.50 0.53 0.69
Page 33 of 64
22 Paper and paper 2.11 2.17 1.86

products

tion of recording
media

education, and
sports
25 Processing of petro 3.56 3.79 3.52

leum, cokeries, nu
clear fuel

rials and chemical
products
29 Rubber 1.33 1.12 1.02
30 Plastics 2.35 2.47 2.28
Page 34 of 64

al products

ing of ferrous met
als

ing of non-ferrous
metals
35 General purpose 4.75 4.51 5.46

machinery

chinery
37 Transport equip 7.41 8.54 7.54

ment

ery and equipment
Page 35 of 64
40 Communication 7.39 9.64 8.44

equipment, comput
ers etc.

ments and machin
ery
Total 100 100 100
Page 36 of 64
Perspective
FIGURE 4.3 Empirical density of (log) labour pro

ductivity of exporters and non-exporters of transport
equipment (CIC 37) and electrical machinery and
equipment (CIC 39) sectors in selected years (1998,
2003, and 2007)
Page 37 of 64
Table 4.5 Within-sector learning vs. structural change in productivity growth
CIC Sec P0 PT An In Shif To P0 PT An In Shif To
tor nu tra t tal nu tra t tal
al (%) (%) (%) al (%) (%)
gro gro
wth wth
(%) (%)
1998 2002
– –7
2002
13 Food 40.6 74.3 16.2 4.20 0.02 4.22 74.3 165. 17.3 5.49 0.00 5.49
pro 8 0 5 0 12 2
cess
ing
of
agri
cul
tural
prod
ucts
Page 38 of 64
14 Oth 35.2 62.0 15.1 1.70 0.01 1.71 62.0 128. 15.7 2.16 0.00 2.16
er 6 7 8 7 80 2
food
stuff
s
15 Bev 50.3 86.3 14.4 2.57 0.00 2.57 86.3 174. 15.0 2.32 0.00 2.32
er 0 3 6 3 04 5
ages
16 To 311. 634. 19.4 6.05 0.00 6.05 634. 1448 17.9 4.79 0.00 4.79
bac 94 26 1 26 .41 6
co
17 Tex 18.9 35.6 17.1 5.78 0.00 5.78 35.6 76.1 16.4 6.41 0.00 6.41
tiles 1 3 6 3 9 2
18 Gar 23.6 30.3 6.41 1.05 – 0.14 30.3 55.7 12.9 2.40 – 1.85
men 6 4 0.91 4 3 4 0.56
ts,
foot
wear
, etc.
Page 39 of 64
19 Leat 25.9 34.5 7.39 0.73 – 0.27 34.5 55.6 10.0 1.22 – 0.36
her, 7 4 0.46 4 2 0 0.87
fur,
feat
her,
etc.
20 Pro 25.9 44.5 14.4 0.68 – 0.63 44.5 94.4 16.2 1.14 – 0.72
cess 4 1 5 0.06 1 7 4 0.42
ing
of
tim
ber,
man
uf.
of
woo
d,
bam
boo,
etc.
21 Fur 32.3 43.4 7.65 0.22 – 0.15 43.4 67.5 9.22 0.42 – –
ni 4 5 0.07 5 2 0.67 0.25
ture
Page 40 of 64
22 Pa 27.2 55.1 19.2 2.29 0.00 2.29 55.1 132. 19.2 2.77 0.00 2.77
per 7 1 3 1 98 6
and
pa
per
prod
ucts
23 Print 30.4 59.6 18.2 1.22 0.00 1.22 59.6 111. 13.3 0.95 0.00 0.95
ing, 5 3 9 3 41 2
re
pro
duc
tion
of
reco
rd
ing
me
dia
Page 41 of 64
24 Arti 24.8 31.2 5.87 0.30 – 0.03 31.2 53.9 11.5 0.65 -0.1 0.49
cles 5 3 0.27 3 1 4 6
for
cul
ture,
edu
catio
n,
and
spor
t ac
tivi
ty
25 Oil 67.0 109. 12.9 2.16 0.00 2.16 109. 126. 2.97 0.37 0.00 0.37
re 7 05 2 05 24
fin
ing,
cok
ing,
nu
clear
fuel
Page 42 of 64
26 Raw 30.8 67.5 21.6 8.75 0.00 8.75 67.5 175. 21.0 10.1 0.00 10.1
che 8 2 0 2 35 3 8 8
mi
cal
ma
teri
als
and
che
mi
cal
prod
ucts
27 Phar 44.9 87.2 18.0 3.21 0.17 3.38 87.2 150. 11.5 2.21 0.00 2.21
mac 4 9 5 9 83 6
euti
cals
28 Che 45.2 76.3 13.9 0.87 0.00 0.87 76.3 158. 15.7 0.92 0.00 0.92
mi 5 5 7 5 49 3
cal
fi
bres
29 Rub 27.8 53.9 17.9 1.24 0.00 1.24 53.9 98.4 12.7 0.92 0.00 0.92
ber 5 4 6 4 0 8
Page 43 of 64
30 Plas 34.8 56.0 12.6 1.73 – 1.70 56.0 85.3 8.76 1.43 – 1.02
tics 6 5 1 0.03 5 1 0.41
31 Non- 21.6 38.5 15.5 4.96 0.00 4.96 38.5 108. 23.0 8.28 0.00 8.28
meta 1 0 3 0 60 5
llic
min
eral
prod
ucts
32 Smel 34.6 80.2 23.4 8.70 0.00 8.70 80.2 208. 21.0 10.1 0.00 10.1
ting 2 7 0 7 33 1 0 0
and
pro
cess
ing
of
fer
rous
met
als
1998 2002
– –7
2002
Page 44 of 64
33 Smel 31.8 66.4 20.1 2.47 0.00 2.47 66.4 208. 25.6 4.70 0.01 4.70
ting 6 4 7 4 07 5
and
pro
cess
ing
of
non-
fer
rous
met
als
34 Met 31.2 52.6 13.9 2.27 – 2.22 52.6 78.3 8.26 1.52 – 0.92
al 2 9 8 0.04 9 7 0.60
prod
ucts
35 Gen 22.6 51.2 22.7 6.01 0.00 6.01 51.2 145. 23.2 9.00 – 8.90
eral 0 6 2 6 54 1 0.10
pur
pose
ma
chin
ery
Page 45 of 64
36 Spe 18.9 47.3 25.7 4.73 0.00 4.73 47.3 144. 25.0 6.01 0.00 6.01
cial 4 5 4 5 79 5
pur
pose
ma
chin
ery
37 Tran 35.1 85.6 24.8 11.3 0.00 11.3 85.6 202. 18.8 11.7 0.00 11.7
spor 9 2 9 5 5 2 69 1 3 3
t
equi
pme
nt
39 Elec 40.5 71.5 15.2 5.13 0.14 5.26 71.5 116. 10.2 4.30 – 3.61
tri 1 8 9 8 70 7 0.69
cal
ma
chin
ery
and
equi
pme
nt
Page 46 of 64
40 Com 68.2 123. 15.9 7.53 1.98 9.51 123. 116. – – – –

mu 8 44 6 44 55 1.14 0.75 0.05 0.81
nica
tion
equi
pme
nt,
com
put
ers,
etc.
41 Mea 31.2 58.4 16.9 1.21 – 1.20 58.4 129. 17.2 1.70 – 1.63
sur 9 5 1 0.01 5 34 2 0.06
ing
in
stru
men
ts
and
ma
chin
ery
Page 47 of 64
42 Art 22.4 34.2 11.1 0.68 – 0.43 34.2 74.2 16.7 1.25 0.00 1.25
work 6 9 5 0.25 9 4 1
and
oth
er
Who 32.7 62.9 17.7 99.8 0.21 100. 62.9 126. 14.9 104. – 100.
le 3 0 4 9 00 0 15 3 58 4.58 00
man
ufac
tur
ing
Note: P0 is the aggregate productivity in the first year of the period. PT is the aggregate productivity in the last year
of the period. Unit: 1000 yuan at 1998 constant prices. ‘Annual growth’ is the compound annual growth rate of ag
gregate labour productivity. ‘Intra’ is the percentage contribution of within-sector productivity growth to overall ag
gregate productivity growth. ‘Shift’ is the percentage contribution of between-sector employment reallocation to
overall aggregate productivity growth. Total is the overall contribution (i.e. the sum of ‘Intra’ and ‘Shift’ effects) of
each two-digit sector to aggregate productivity growth. The row ‘Whole manufacturing’ shows the contribution of
‘Intra’ and ‘Shift’ effects for the aggregated manufacturing sector. Sectors with zero shift effects are the shrinking
ones. (For details on the shift-and-share decomposition method of productivity growth, see Appendix).
Source: Authors’ elaboration on CMM data.
Page 48 of 64
Perspective
4.4 Some General Interpretative Implications

From a theoretical point of view, the above argument implies a radical de-linking of in
come distribution, production theory, and development. Conventionally, there is an obvi
ous link between technological conditions, input availabilities, and remuneration.
It is well known that if production functions are well behaved (homogeneous, degree-one,
hence no increasing returns, etc.), relative scarcities determine relative input intensities.
And if the estimates fall short of fully ‘explaining’ output, then that goes under the head
ing of the famous ‘Solow residual’, also renamed as Total Factor Productivity.
The consequences also for trade theories are straightforward: The Heckscher-Ohlin-Sa
muelson theorems easily follow.
And what about the interpretation of why per capita incomes differ so much across coun
tries? Over the last few decades a disproportionate amount of effort has gone into the
search for arguments to add to the Kamasutra of variables entering the ‘production
function’ (nowadays not only questionable proxies for ‘culture’ and ‘institutions’ but also
sinister notions like ‘genetic endowments’). Here we have taken the opposite route and
explored the implications for development of ‘opening up the black box of technology’, to
use the felicitous definition of Nate Rosenberg.
Within the black box, there are no production functions, and even less so, Cobb-Douglas
ones, but rather painstaking efforts aimed at knowledge accumulation, nested in more or
less supportive organizations and institutions.
4.5 Structural Change as a Fundamental Fea

ture of Catching Up
4.5.1 The Evolution of Technological Capabilities and Production Spe
cializations
With the mentioned partial exception of China—which in a sense entered the catching-up
phase already ‘mature’ in terms of sectoral composition of output—most countries under
go major transformation in the sectors in which they operate and in the products they
manufacture (and China is no exception). However, not every country is successful, with
many remaining in the ‘middle-income trap’.
In many respects, catching-up entails ‘climbing up the ladder’ not only of production effi
ciency—well captured by the dynamics in the productivity distributions discussed in sec
tion 4.4—but also of product complexities and product demand elasticities. (p. 112) To re
call a discussion of the 1990s, the impact on competitiveness and growth of producing
potato chips is not identical to that of producing computer chips! In turn, the climbing up
Page 49 of 64
Perspective
is associated with the accumulation of technological and organizational capabilities, often
against the country’s comparative advantages. That is, absolute technological levels (and
not comparative ones) are a fundamental driver of trade performance, growth, and, ulti
mately, welfare. To clarify the point, Cimoli et al. (2009a) describe a thought experiment
of opening up trade between a ‘Stone Age economy’ and an ICT-based one. As Ricardo
would argue, the country coming from the Stone Age will be more likely to export ‘stone-
intensive’ products in which it has a comparative advantage (and vice versa for the ICT-
based economy with, say, computers). However, there could be no bilateral trade at all if
the more advanced ICT economy will end up producing almost anything worth trading ir
respective of the stone- or ICT-intensities of the products. What really matters for eco
nomic growth might ultimately be absolute levels of technological capabilities and how
they interact with world demand for products. An interesting measure in this respect of
the ‘fitness’ of a country in terms of the ‘complexity’ of the products in which it special
izes as a predictor of its growth potential is presented in Tacchella et al. (2012) and
Cristelli et al. (2015).4
Comprehensive historical overviews of successful cases of structural transformation and

industrialization is provided by Moses Abramovitz (1986) and Ha-Joon Chang (2002). The
most telling cases of successful latecomer industrialization are probably the United
States, Germany, and more recently Japan, South Korea, and China.
It was the First Secretary of the US Treasury, Alexander Hamilton, who systematically
elaborated the infant industry argument in 1791. In a nutshell, Hamilton argued that for
eign competition would have prevented domestic industries from becoming international
ly competitive, unless the state had intervened to compensate initial losses or to enforce
import duties (Hamilton, 1791). American industries ended up being literally the most
protected in the world until after the Second World War (Chang, 2002); needless to say,
this goes a long way in explaining the US pattern of structural change. Furthermore, the
role of the Federal Government in industrial development has been substantial even in
the post-war era, thanks to extensive defence-related procurement and mission-oriented
research (Mazzucato, 2013; Mowery, 2012). Similarly, List (1841) lucidly discusses the
shortcomings of simply adhering to comparative advantages: in his view, the true objec
tive of Britain trying to impose free trade on Europe was simply ‘kicking away the ladder’
that they themselves had climbed (Chang, 2002). The German experience also points to
the importance of ad hoc institutional innovations which facilitated catching-up and were
the basis of the successive forging ahead with respect to Britain. Of particular impor
tance was the introduction of the Humboldtian university model for the education of grad
uate engineers, which supplied human capital that proved essential for the diffusion of in-
house industrial R&D (p. 113) departments (Dosi et al., 1994). Another pillar of German in
dustrialization was the emulation of imported British machine tools (often thanks to
British craftsmen attracted to Prussia; Freeman, 1995). More recently, Japan (Freeman,
1987) and the Asian Tigers (Nelson and Pack, 1999) were able to reap the benefits of
rapidly growing markets for industrial machinery embodying ‘frontier’ knowledge. At the
heart of the Japanese success lay the explicit decision by Japanese political authorities to
neglect the ‘natural’ development path implied by comparative advantages (Freeman,
Page 50 of 64
Perspective
2004). In just a few years, Japan ceased being an importer of foreign technology and de
veloped important indigenous innovation capabilities, even surpassing the United States
in terms of R&D efforts. The secret of its success was building up one of the most suc
cessful innovation systems (which inspired the formulation of the concept itself, see Free
man, 1987), in which long-term planning by the Ministry of International Trade and In
dustry (MITI) fostered learning and spurred innovation in the export-led industrial com
plexes.
The classic works mentioned above are detailed case studies of single countries and their
historical experience. More recently, research leveraging natural quasi-experiments and
new estimation techniques has allowed the precise causal identification of the effects of
sectoral policies. For instance, China’s 11th Five-Year Plan (2006–10) promoted shipbuild
ing as a strategic industry for defence-related purposes. Kalouptsidi (2017) finds that the
reduction in production costs associated with the policy explains China’s massive gains of
global market share in ships: without the targeted subsidies, China’s production would be
cut to less than half. Lane (2017) studies the Heavy Chemical and Industry (HCI) policy
that South Korea enacted in 1973 as a response to the US troop withdrawal. Again, tar
geted industries were chosen for their military importance, and the comparison with oth
erwise similar industries shows that the policy promoted rapid development that lasted
long after the measures were removed. Interestingly enough, downstream sectors also
benefited from the lower prices induced by the policy, an instance of the policy-induced
industrial externalities that Hirschman (1958) termed ‘forward linkages’. The HCI en
tailed both industrial subsidies and targeted trade protection. Nonetheless, it must be
noted that in certain situations trade protection alone can be sufficient to change the pat
terns of trade and allow industrialization. Juhasz (2018) documents that the temporary
protection from British imports caused by the Napoleonic Blockade was fundamental for
the accumulation of technological capabilities in nineteenth-century France. The mecha
nized cotton-spinning industry rapidly developed in French départements that received
more sheltering, in accordance with the predictions of the infant industry argument. Han
lon (2019) complements this evidence by looking at production input advantages, instead
of output market protection. Using twentieth-century metal shipbuilding data, he shows
that even a temporary cost advantage can become the source of long-lasting competitive
advantage due to dynamic localized learning effects and learning-by-doing. Head (1994)
studies the effects of infant-industry protection in the rail steel industry. And the list could
be very long.
Let us turn to the role of policies. Here we end by simply noting that technologi
(p. 114)
cal catching-up (and of course straightforward innovation) goes hand in hand with organi
zational innovation.
Page 51 of 64
Perspective
4.6 The Crucial Role of Industrial Policies

Some general patterns can be distilled from these historical cases. (More in Cimoli et al.,
2009a, and especially 2009c and 2009d. For quite germane policy considerations cf. Paus,
2019; Reinert, 1999.)
4.6.1 Emulation and, sometimes, Leapfrogging as a General Principle

Inspiring Policies
Emulation—to borrow Reinert’s (2009) term—is the purposeful effort of imitation of ‘fron
tier’ technologies and production activities, irrespective of the incumbent’s comparative
advantage profile. It often involves explicit public policies aimed at doing what rich coun
tries are doing in terms of production, and it always involves microeconomic efforts—on
the part of individuals and, more so, firms—to learn how to do things others in frontier
countries are already able to do. It is a familiar story over the last three centuries. It
dates back at least to the case of England vis-à-vis the Low Countries in the period pre
ceding the Industrial Revolution, and it applies all the way to contemporary Chinese in
dustrialization. Emulation primarily concerns products and processes based on new tech
nological paradigms. At one time it meant mechanized textile production and the con
struction of related machines. Later it was steel production, electricity-based products
and machinery, and the internal combustion engine. Nowadays it has to do first and fore
most with information and telecommunication technologies, biotech, and nanotechnolo
gies.
It has sometimes happened that catching-up countries not only emulated the leading
ones, but ‘leapfrogged’ them in some of the newest, most promising technologies. This oc
curred in the nineteenth-century United States and Germany, which forged ahead of Eng
land in electromechanical engineering, consumer durables, and synthetic chemistry.
But why should everyone emulate frontier technologies in the first place, rather than be
ing guided by one’s own ‘comparative advantages’? Or, as the sceptics often put it, isn’t it
absurd to suggest that everybody should specialize in ICT production? We have answered
this question above. Typically, relatively backward economies display an absolute disad
vantage in everything, that is, they are less efficient in the production of (p. 115) every
commodity, and in fact the disadvantage in many commodities is likely to be infinite in the
sense that they are not able to produce them at all. Catching up entails closing the gap in
production knowledge and learning how to produce novel goods (which in the beginning
are generally novel only for the catching-up country, even if ‘old’ for the world). This is
particularly important with respect to new technological paradigms because such tech
nologies are most often general purpose: they influence most production activities direct
ly or indirectly. In the past this was the case for mechanical engineering and electricity as
it is today for ICT technologies. Moreover, goods and equipment based on the new tech
nological paradigms generally entail higher elasticity of demand and richer opportunities
for further technological advance (cf. Dosi, Pavitt, and Soete, 1990; Castaldi et al., 2009;
Cimoli et al., 2009b). Hence emulation of frontier countries in these activities implies,
Page 52 of 64
Perspective
other things being equal, higher growth possibilities and greater potential for productivi
ty growth and, eventually, domestic product innovation.
4.6.2 The Complementarity between Technological Learning and the

Development of Production Capacity
We have emphasized above the difference between technological knowledge and pure in
formation, with important implications in terms of ‘stickiness’ and difficulty in the trans
mission of the former—embodied as it generally is in specific people, organizations, and
local networks. One consequence of this is that learning rarely occurs ‘offline’, especially
in the initial phases of industrialization. Rather, it goes together with the acquisition of
production equipment, and with the efforts of learning how to use it and how to adapt it
to local conditions (more in Bell and Pavitt, 1993). In turn, this goes hand in hand with the
training of workers and engineers and the professional development of managers capable
of efficiently running complex organizations. These are also reasons why it is dangerous
to see industrialization, even in its early stages, simply as a matter of ‘diffusion’, or the
turnkey adoption and use of equipment acquired from abroad, all the more so when the
technologies are in the form of blueprints or licences requiring much painstaking local
learning. Of course, no policymaker is in a position to fine-tune the details of the produc
tion activities and patterns of learning which the economy has to exploit. The details of
the actual dynamics depend a good deal on the details of corporate strategies—and
chance. There was no way, for example, that the Korean policymakers could know or even
less ‘plan’, say, a learning push in semiconductor memories rather than microprocessors.
However, policymakers should be acutely aware of the fact that future capabilities build
upon, refine, and modify incumbent ones: hence the policy goal of building good path-de
pendencies (the point resonates with similar advice by Hausmann and Rodrick (2006) re
garding patterns of product diversification along the development process).
Two fundamental caveats must be kept in mind. First, a useful distinction can be
(p. 116)
made between production capacity—the knowledge and organizational routines needed to

run, repair, and incrementally improve existing equipment and products—and technologi
cal capabilities—the skills, knowledge, and organizational routines needed to manage and
generate technical change (Bell and Pavitt, 1993: 163). The kinds of activity that foster
the accumulation of the latter increasingly involve also specialized R&D laboratories, de
sign offices, production engineering departments, etc.
Second, and relatedly, ‘while various forms of “doing” are central to technological accu
mulation, learning should not be seen simply as a doing-based process that yields addi
tional knowledge simply as the by-product of activities undertaken with other objectives.
It may need to be undertaken as a costly, explicit activity in its own right: various forms of
technological training and deliberately managed experience accumulation’ (Bell and
Pavitt, 1993: 179). Interestingly, the transition from the production capacity phase to the
technological capabilities phase has been managed superbly by countries like Korea and
Taiwan and it is where, on the contrary, most Latin American countries got stuck.
Page 53 of 64
Perspective
4.6.3 The Necessity of Nurturing Infant Industries
Consider again the caricature of a Stone-Age economy and an ICT economy and allow
them to interact. Two properties are quite straightforward. First, the patterns of econom
ic signals will be biased in favour of stone-intensive product in one country, and ICT-inten
sive in the other (their current ‘comparative advantages’). Hence, if the former wants to
enter the ICT age, it has to purposefully distort market signals as they come from interna
tional exchanges (on the assumption that there are some: it could well be that the ICT
economy is unwilling to absorb any stone product!). Second, it is quite unlikely that the
stone producers, even under the ‘right kind of signal’, will be able to instantly acquire the
knowledge to competitively produce ICT products.
Certainly, all individuals take a long time to learn new skills. Turning violinists into foot
ball players and vice versa is rather hard, if possible at all. And this applies even more so
to organizations and organization-building. Even when transformations are possible, they
require time, nurturing, and care. If a newly born violinist, ex-footballer, is made to com
pete with professional violinists, he will make a fool of himself. If a catching-up company
is suddenly made to compete with world leaders it will most likely disappear. Often, it is a
daunting enough task to learn how to make—no matter how inefficiently—a product
which might indeed be rather standard in technologically more sophisticated economies:
demanding competitive efficiency as well is like asking the violinist to run 100 metres in
around ten seconds after some quick training rounds.
Safeguarding the possibility of learning is the basic pillar of the infant industry logic. On
the incentive side, market signals alone are often not enough, and indeed they frequently
discourage the accumulation of technological capabilities in so far as they (p. 117) ought
to occur in activities currently displaying significant comparative disadvantages and thus
also unfavourable current profitabilities. Incidentally note, also, that the existence of fi
nancial markets is a meagre instrument, if of any use at all, for translating a future and
uncertain potential for learning into current investment decisions (more in Stiglitz, 1994).
Thus, there are also sound learning-related reasons why, as the historical evidence shows,
just prior to industrial catching-up, average industrial import tariffs are relatively low;
they rise rapidly in the catching-up phase, and they fall after mature industrialization. In
deed, it is during the catching-up phase that the requirement to distort (international)
market signals is more acute, precisely because there are young and still relatively fragile
learning infants. Beforehand, there are no infants to speak of. Afterwards, there are
adults able to swim into the wild international ocean by themselves.
Doing so, however, involves more than just ‘signal distortion’. As many of the Latin Ameri
can experiences have shown, this is far from enough. Partly it has to do with the fact that
many forms of protection entail the possibility of learning but not, in the language of
Khan and Blankenburg (2009), the compulsion to innovate as distinct from the sheer in
centive to just exploit a monopoly rent, no matter how inefficient and lazy the potential
‘learner’. Partly, it has to do with the capabilities accumulation and the characteristics of
the actors involved. (An archetypical example of purposefully ‘getting signals wrong’ and
Page 54 of 64
Perspective
fostering capability accumulation is Korea: see Amsden, 1989.) After all, even with the
best of intentions and incentives, our violinist will not only take time to learn but will ac
tually be able to develop his/her football skills only as part of a team. The team, in turn,
will most often not be the making of sheer self-organization, especially when production
entails relatively complex products, as it usually does. At the same time, violinists might
not be the best candidates for playing football, irrespective of the incentive structure.
Moving away from the metaphor, industrialization might have little to do with simply the
award of property rights and with the establishment of firms as legal entities. It is quite
misleading to think that the world is full of evenly distributed sources of technological
knowledge just waiting to be exploited—the lag being due mainly to institutional and in
centive-related forces. On the contrary, irrespective of the opportunities for the entrepre
neurial exploitation of technological knowledge, which the ‘international knowledge fron
tier’ notionally offers, the fundamental gap is about the lack of capabilities to explore and
exploit them. This is a crucial bottleneck for development. Even casual visitors to devel
oping countries (whenever they walk out of hotels paid for by the IMF!) notice these gaps
—which in the early stages of development concern even rather basic activities such as
accessing the Internet or processing a credit card payment. They apply much more to
firm-level capabilities such as drilling an oil well (or, in the early stages, even keeping an
existing well working). As discussed in several contributions to Cimoli, Dosi, and Stiglitz
(2009a), ‘horizontal’ education and training policies, together with technical support giv
en to firms by public institutions, can go a long way to enhance capabilities. But even that
is not likely to be enough. Policies are often bound to get their hands dirty with respect to
the nature, internal structure, and strategies of a few (p. 118) corporate agents them
selves. Fostering the emergence of, and occasionally explicitly building, technologically
and organizationally competent firms are fundamental infant nurturing tasks. In fact,
even the most developed countries only boast a fraction of technologically dynamic orga
nizations within a much greater population of firms. (Note that all this applies to both
‘high tech’ and ‘low tech’ sectors as conventionally defined.) In a sense, industrialization
involves changing the distribution between ‘progressive’ and ‘backward’ firms.
Indeed, all this might not be enough: the state in the past often had to do more than just
‘pushing and pulling’ entrepreneurs into certain strategic sectors, and ended up acting as
‘entrepreneur of last resort’. We believe that this continues to be the case today.
Indeed industrial policies for development and catching up are likely to involve the fol
lowing ‘capital sins’ which the market faithful are supposed to avoid: (i) state ownership;
(ii) selective credit allocation; (iii) favourable tax treatment for selected industries; (iv) re
strictions (or some conditionalities) on foreign investment; (v) local context requirements;
(vi) special IPR regimes; (vii) government procurement; and (viii) promotion of large do
mestic firms. (Dahlman, 2009, discusses them for China and India, but the lessons are
more general.)
In a nutshell, this is the full list of the capital sins which the market faithful are supposed
to avoid!
Page 55 of 64
Perspective
4.6.4 Industrial Policies in a Sino-centric World: Some Conclusions
A combination of ‘infant nurturing’ measures has been a major ingredient of development

policies throughout the history of industrialization, and it continues to be so today. Histor
ically, the ‘infant learners’ had to be shielded in the domestic markets and helped in the
international ones in their interactions with the more efficient and more innovative firms
from ‘frontier’ countries. This happens to a large extent to also hold today. However, the
unique feature of the current ‘Sino-centric’ world—as Castro (2009) puts it—is that many
catching-up countries are caught between a rock and a hard place: the developed world
is still ahead of them, but at the same time China is rapidly eliminating its absolute disad
vantages across the board, both in more traditional productions and in activities based on
the newest technological paradigms. And it is doing so at a higher rate than it is catching
up in wages (notwithstanding the fast growth of the latter). The outcome is an absolute
cost advantage in an expanding set of goods, including those which were/are central to
the industrial production of many low- and middle-income countries. In that respect the
magnitude and speed of Chinese industrialization risks crowding out the industrializing
potential of many other countries. So, for example, Brazil—a country on the upper tail of
the distribution of industrializers in terms of its technological capabilities—is a very ‘high
wage’ country (p. 119) compared to China, but so also are other less developed Latin
America countries, and even African countries are losing cost-based international (and
domestic) competitiveness vis-à-vis China. Is this a reason to give up the ‘infant nurtur
ing’/capability accumulation philosophy? In our view it is not: on the contrary, it adds to
the reasons for practising various combinations of the ‘capital policy sins’ mentioned
above. And it ought to encourage a more explicit use of domestic or regional markets for
cultivating emerging national industries, even when the latter are being squeezed in the
international arena between ‘advanced producers’ and Chinese exports.
4.7 Appendix: Shift-and-Share Decomposition

Methodology
To measure the contributions of structural change to aggregate productivity growth, it is
crucial to distinguish between the contributions of shifts between sectors and the contri
butions of productivity growth within sectors. Notice that we use an aggregated dataset
based on our firm-level dataset for the decomposition method described below.
Recent studies using the shift-share technique include Ali Akkemik (2005), Timmer and
Szirmai (2000), and Kumar and Russell (2002). We adopt van Ark and Timmer’s (2003)
shift-share model, in order to be comparable with the results of Wang and Szirmai (2008).
The difference in aggregate labour productivity levels at time 0 and T can be written as
Page 56 of 64
Perspective
(1)
with and the labour productivity of sector i at year 0 and T; and the employ
ment share of sector i at year 0 and T; sector’s period average share of total employ
ment; and sector’s period average labour productivity. The growth of aggregate pro
ductivity can be decomposed into intra-sectoral productivity growth (the first term on the
right-hand side of equation (1), called ‘intra-effect’) and the effects of changes in the sec
toral allocation of labour (the second term, called ‘shift-effect’). Let Ci denote the contri
bution of sector i to the aggregate labour productivity growth. We have
(2)
Van Ark and Timmer (2003) reallocate all shift effects ( ) from sectors that experi
enced shrinking labour shares to sectors that expanded their share in total (p. 120) labour.
Suppose K is the set of sectors which expand their labour shares; J is the set of sectors
with declining labour share. For expanding sectors k and shrinking sectors j,
(3)
(4)
with average labour productivity overall shrinking sectors and averaging over years
(5)
Acknowledgements
This work draws significantly on Cimoli and Dosi (1995), Cimoli, Dosi, and Stiglitz
(2009a), Yu et al. (2015) and Dosi and Tranchero (2020). We thank Eva Paus for her in
sightful comments. Support from the European Union Horizon 2020 Research and Inno
vation programme under grant agreement No. 822781-GROWINPRO is gratefully ac
knowledged.
Page 57 of 64
Perspective
References
Abramovitz, M. (1986) ‘Catching up, Forging ahead, and Falling behind’, The Journal of
Economic History 46: 385–406.
Akkemik, K. Ali (2005) ‘Labor Productivity and Inter-sectoral Reallocation of Labor in Sin
gapore (1965–2002)’, Forum of International Development Studies 30: 1–22.
Amsden, Alice H. (1989) Asia’s Next Giant: South Korea and the Last Industrialization.
Amsden, Alice H. and Takashi Hikino (1993) ‘Staying behind, Stumbling back, Sneaking
up, Soaring ahead: Late Industrialization in Historical Perspective’, in William Baumol,
Richard Nelson, and Edward Wolff (eds) International Convergence of Productivity with
Some Evidence from History. Oxford: Oxford University Press.
Bell, Martin and Keith Pavitt (1993) ‘Technological Accumulation and Industrial Growth:
Contrasts between Developed and Developing Countries’, Industrial and Corporate
Change 2(2): 157–210.
Castaldi, Carolina, Mario Cimoli, Nelson Correa, and Giovanni Dosi (2009) ‘Technological
Learning, Policy Regimes and Growth: The Long Term Patterns and Some Specificities of
a Globalized Economy’, in Mario Cimoli et al. (eds) Industrial Policy and Development:
The Political Economy of Capabilities Accumulation. Oxford: Oxford University Press, pp.
39–76.
Castro, Antonio Barros de (2009) ‘The Impact of Public Policies in Brazil along the Path
from Semi-stagnation to Growth in a Sino-centric Market’, in Mario Cimoli et al. (eds) In
dustrial (p. 121) Policy and Development: The Political Economy of Capabilities Accumula
Chandler, Alfred D. (1990) Scale and Scope: The Dynamics of Industrial Capitalism. Cam
Chandler, Alfred D. (1992a) ‘Corporate Strategy, Structure and Control Methods in the
United States during the 20th Century’, Industrial and Corporate Change 1: 263–84.
Chandler, Alfred D. (1992b) ‘Organizational Capabilities and the Economic History of the
Industrial Enterprise’, Journal of Economic Perspectives 6, 79–100.
Chandler, Alfred D. (1993) ‘Learning and Technological Change: The Perspective from
Business History’, in Richard W. Thomson (ed.) Learning and Technological Change. Bas
ingstoke: Palgrave Macmillan, pp. 157–78.
Page 58 of 64
Perspective
Cimoli, Mario (1990) ‘Technology, International Trade and Development: A North South
Perspective’. PhD thesis, University of Sussex.
Cimoli, Mario and Giovanni Dosi (1988) ‘Technology and Development: Some Implications
of Recent Advances in the Economics of Innovation for the Process of Development’, in
Atul Wad (ed.) Science, Technology and Development: A Critical Perspective. Boulder, CO:
Westview Press.
Cimoli, Mario and Giovanni Dosi (1995) ‘Technological Paradigms, Patterns of Learning
and Development: An Introductory Roadmap’, Journal of Evolutionary Economics 5(3):
243–68.
Cimoli, Mario, Giovanni Dosi, and Joseph E. Stiglitz (2009a) Industrial Policy and Develop
ment: The Political Economy of Capabilities Accumulation. Oxford: Oxford University
Press.
Cimoli, Mario, Giovanni Dosi, and Joseph E. Stiglitz (2009b) ‘The Political Economy of Ca
pabilities Accumulation: The Past and Future of Policies for Industrial Development’, in
Mario Cimoli et al. (eds) Industrial Policy and Development: The Political Economy of Ca
pabilities Accumulation. Oxford: Oxford University Press, pp. 1–16.
Cimoli, Mario, Giovanni Dosi, Richard Nelson, and Joseph E. Stiglitz (2009c) ‘Institutions
and Policies Shaping Industrial Development: An Introductory Note’, in Mario Cimoli et
al. (eds) Industrial Policy and Development: The Political Economy of Capabilities Accu
mulation. Oxford: Oxford University Press, pp. 19–38.
Cimoli, Mario, Giovanni Dosi, and Joseph E. Stiglitz (2009d) ‘The Future of Industrial Poli
cies in the New Millennium: Toward a Knowledge-centered Development Agenda’, in
Mario Cimoli et al. (eds) Industrial Policy and Development: The Political Economy of Ca
pabilities Accumulation. Oxford: Oxford University Press, pp. 541–60.
Cohen, Avi J. and Geoffrey C. Harcourt (2003) ‘Whatever Happened to the Cambridge
Capital Theory Controversies?’, Journal of Economic Perspectives, 17: 199–214.
Cohen, Wesley M. and Daniel A. Levinthal (1989) ‘Innovation and Learning: The Two
Faces of R&D’, The Economic Journal 99: 569–96.
Constant, Edward (1985) The Turbojet Revolution. Baltimore, MD: Johns Hopkins Univer
sity Press.
Coriat, Benjamin (1990) Penser a l’envers. Paris: Bourgeois.
Cristelli, Matthieu, Andrea Tacchella, and Luciano Pietronero (2015) ‘The Heterogeneous
Dynamics of Economic Complexity’, PloS One 10(2): e0117174.
Dahlman, Carl J. (2009) ‘Growth and Development in China and India: The Role of Indus
trial and Innovation Policy in Rapid Catch-up’, in Mario Cimoli et al. (eds) Industrial Poli
Page 59 of 64
Perspective
cy and (p. 122) Development: The Political Economy of Capabilities Accumulation. Oxford:
Oxford University Press, pp. 303–35.
Dornseifer, Bernd and Jürgen Kocka (1993) ‘The Impact of the Pre-industrial Heritage:
Reconsiderations on the German Pattern of Corporate Development in the Late 19th and
Early 20th Century’, Industrial and Corporate Change 2(2): 233–48.
Dosi, Giovanni (1988) ‘Sources, Procedures, and Microeconomic Effects of Innovation’,

Journal of Economic Literature 26(3): 1120–71.
Dosi, Giovanni and Marco Grazzi (2006) ‘Technologies as Problem-solving Procedures and
Technologies as Input–Output Relations: Some Perspectives on the Theory of Production’,
Industrial and Corporate Change 15(1): 173–202.
Dosi, Giovanni and Richard Nelson (2010) ‘Technical Change and Industrial Dynamics as
Evolutionary Processes’, in Bronwyn H. Hall and Nathan Rosenberg (eds) Handbook of
the Economics of Innovation. Amsterdam: Elsevier, pp. 51–127.
Dosi, Giovanni and Matteo Tranchero (2020) ‘The Role of Comparative Advantage, En
dowments and Technology in Structural Transformation’, in Ludovico Alcorta, Adam Szir
mai, Bart Verspagen, and Neil Foster-McGregor (eds) New Perspectives on Structural
Change: Causes and Consequences of Structural Change in the Global Economy. Oxford:
Dosi, Giovanni, Christopher Freeman, and Sylvia Fabiani (1994) ‘The Process of Econom
ic Development: Introducing Some Stylized Facts and Theories on Technologies, Firms
and Institutions’, Industrial and Corporate Change 3(1): 1–45.
Dosi, Giovanni, Renato Giannetti, and Pier Angelo Toninelli (eds) (1993) Technology and
Enterprise in a Historical Perspective. Oxford: Clarendon Press.
Dosi, Giovanni, Keith Pavitt, and Luc Soete (1990) The Economics of Technical Change
and International Trade. London: Harvester Wheatsheaf.
Freeman, Chris (1987) Technology Policy and Economic Performance: Lessons from
Japan. Pinter.
Freeman, Chris (1995) ‘“The National System of Innovation” in Historical Perspective’,

Cambridge Journal of Economics 19(1): 5–24.
Freeman, Chris (2004) ‘Technological Infrastructure and International Competitiveness’,

Industrial and Corporate Change 13: 541–69.
Freeman, Chris (2019) ‘History, Co-evolution and Economic Growth’, Industrial and Cor
porate Change 28(1): 1–44.
Freeman, Chris and Francisco Louca (2001) As Time Goes by: From the Industrial Revolu
tions to the Information Revolution. Oxford: Oxford University Press.
Page 60 of 64
Perspective
Hamilton, Alexander (1791) Reports of the Secretary of the Treasury on the Subject of
Manufactures, in The Papers of Alexander Hamilton, edited by Harold C. Syrett et al., 26
vols. New York: Columbia University Press (1961–79).
Hanlon, W. Walker (2019) ‘The Persistent Effect of Temporary Input Cost Advantages in
Shipbuilding, 1850–1911’. NYU Stern Working Paper.
Hausmann, Ricardo and Dani Rodrik (2006) ‘Doomed to Choose: Industrial Policy as
Predicament’. CID Working Paper. Cambridge, MA: Center for International Development
at Harvard University.
Head, Keith (1994) ‘Infant Industry Protection in the Steel Rail Industry’, Journal of Inter
national Economics 37(3–4): 141–65. (p. 123)
Hidalgo, Cesar A. and Ricardo Hausmann (2009) ‘The Building Blocks of Economic Com
plexity’, Proceedings of the National Academy of Sciences 106: 10570–5.
Juhasz, Reka (2018) ‘Temporary Protection and Technology Adoption: Evidence from the
Napoleonic Blockade’, American Economic Review 108: 3339–76.
Kalouptsidi, Myrto (2017) ‘Detection and Impact of Industrial Subsidies: The Case of Chi
nese Shipbuilding’, The Review of Economic Studies 85: 1111–58.
Khan, Mushtaq H. and Stephanie Blankenburg (2009) ‘The Political Economy of Industrial
Policy in Asia and Latin America’, in Mario Cimoli et al. (eds) Industrial Policy and Devel
opment: The Political Economy of Capabilities Accumulation. Oxford: Oxford University
Press, pp. 336–77.
Kogut, Bruce (1992) ‘National Organizing Principles of Work and the Dominance of the
American Multinational Corporation’, Industrial and Corporate Change 1: 285–326.
Kogut, Bruce (1993) Country Competitiveness. Oxford: Oxford University Press.
Kumar, Subodh and R. Robert Russell (2002) ‘Technological Change, Technological Catch-
up, and Capital Deepening: Relative Contributions to Growth and Convergence’, The
American Economic Review 92(3): 527–48.
Landini, Fabio and Franco Malerba (2017) ‘Public Policy and Catching up by Developing
Countries in Global Industries: A Simulation Model’, Cambridge Journal of Economics
41(3): 927–60.
Lane, Nathan (2017) ‘Manufacturing Revolutions: Industrial Policy and Networks in

South Korea’. IIES Working Paper. Stockholm.
Page 61 of 64
Perspective
Lee, Keun and Franco Malerba (2017) ‘Catch-up Cycles and Changes in Industrial Leader
ship: Windows of Opportunity and Responses of Firms and Countries in the Evolution of
Sectoral Systems’, Research Policy 46(2): 338–51.
List, Friedrich (1841) The National System of Political Economy. London: Longmans.
Malerba, Franco and Richard Nelson (2011) ‘Learning and Catching up in Different Sec
toral Systems: Evidence from Six Industries’, Industrial and Corporate Change 20(6):
1645–75.
Mowery, David C. (2012) ‘Defense-related R&D as a Model for “Grand Challenges” Tech
nology Policies’, Research Policy 41(10): 1703–15.
Nelson, Richard and Howard Pack (1999) ‘The Asian Miracle and Modern Growth Theo
ry’, The Economic Journal 109: 416–36.
Paus, Eva (2019) ‘Innovation Strategies Matter: Latin America’s Middle-income Trap
Meets China and Globalization’, The Journal of Development Studies 56(4): 657–79.
Reinert, Erik S. (1999) ‘The Role of the State in Economic Growth’, Journal of Economic
Studies 26: 268–326.
Reinert, Erik S. (2007) How Rich Countries Got Rich and Why Poor Countries Stay Poor.
London: Constable.
Reinert, Erik S. (2009) ‘Emulation versus Comparative Advantage: Competing and Com
plementary Principles in the History of Economic Policy’, in Mario Cimoli et al. (eds) In
dustrial Policy and Development: The Political Economy of Capabilities Accumulation. Ox
ford: Oxford University Press, pp. 79–106.
Shaikh, Anwar (2016) Capitalism. Oxford: Oxford University Press. (p. 124)
Stiglitz, Joseph E. (1994) ‘Economic Growth Revisited’, Industrial and Corporate Change
3(1): 65–110.
Syverson, Chad (2011) ‘What Determines Productivity?’, Journal of Economic Literature

49(2): 326–65.
Tacchella, Andrea, Matthieu Cristelli, Guido Caldarelli, Andrea Gabrielli, and Luciano
Pietronero (2012) ‘A New Metrics for Countries’ Fitness and Products’ Complexity’, Scien
tific Reports on Nature 2: 723.
Timmer, Marcel P. and Adam Szirmai (2000) ‘Productivity Growth in Asian Manufactur
ing: The Structural Bonus Hypothesis Examined’, Structural Change and Economic Dy
namics 11: 371–92.
Page 62 of 64
Perspective
van Ark, Bart and Marcel Timmer (2003) ‘Asia’s Productivity Performance and Potential:
The Contribution of Sectors and Structural Change’. The Conference Board, Washington.
Vincenti, Walter G. (1992) ‘Engineering Knowledge, Type of Design, and Level of Hierar
chy: Further Thoughts about What Engineers Know’, in Peter Kroes and Martijn Bakker
(eds) Technological Development and Science in the Industrial Age. Dordrecht: Kluwer
Academic Publishers, pp. 17–34.
Wang, Lili and Adam Szirmai (2008) ‘Productivity Growth and Structural Change in Chi
nese Manufacturing, 1980–2002’, Industrial and Corporate Change 17(4): 841–74.
Yu, Xiaodan, Giovanni Dosi, Jiasu Lei, and Alessandro Nuvolari (2015) ‘Institutional
Change and Productivity Growth in China’s Manufacturing: The Microeconomics of
Knowledge Accumulation and “Creative Restructuring”’, Industrial and Corporate Change
24(3): 565–602.
Notes:
(1) We chose Italy and France as we have access to comparable micro data. Our informed
guess, based on smaller samples like COMPUSTAT and Orbis firm-level evidence, sup
ports the argument that the property applies to all advanced countries, including the
United States and Germany.
(2) Measures of ‘capital’ are at best biased by construction: witness the old ‘capital con
troversy’ between Cambridge, United Kingdom and Cambridge, MA (more in Cohen and
Harcourt, 2003 and Shaikh, 2016). And more so are measures simply obtained from bal
ance-sheets. In particular, ‘capital’ measures in the case of China (in firm’s balance-sheet)
are calculated as the value of fixed capital stock at original purchase prices (these book
values are the sum of nominal values for different years).
(3) We are currently exploring the conjecture that within the overall pattern of fast learn
ing by Chinese manufacturing, many ‘non-frontier’ firms found it easier to enter export
markets following the accession of China to the WTO.
(4) A germane attempt, characterized however by a few technical drawbacks, is in Hidal

go and Hausmann (2009).
Mario Cimoli
Mario Cimoli is deputy executive secretary of the United Nations Economic Commis
sion for Latin America and the Caribbean (ECLAC). In 2004, he was appointed co-di
rector (with Giovanni Dosi and Joseph Stiglitz) of two task forces: Industrial Policy
and Intellectual Property Rights Regimes for Development (Initiative for Policy Dia
logue, Columbia University, New York). Amongst his broad interests are economic de
velopment and its relationship to production structure, productivity growth, interna
tional trade, and structural change. His work analyses the linkages between industri
Page 63 of 64
Perspective
al policy, technology development and innovation, and their role in shaping develop
ment trajectories.
Giovanni Dosi
Giovanni Dosi is professor of economics at Sant’Anna School of Advanced Studies,

Pisa, Italy, and continental Europe editor of the journal Industrial and Corporate
Change. He is included in the ISI Highly Cited Research list of scholars who have
made fundamental contributions to the advancement of science and technology, and
is a corresponding member of the Accademia Nazionale dei Lincei, the first academy
of sciences in Italy. In 2016, he received the Wiley TIM Distinguished Scholar Award
from the Technology and Innovation Management Division of the American Academy
of Management. His major research areas, in which he is author and editor of several
works, include the economics of innovation and technological change, industrial eco
nomics, evolutionary theory, economic growth and development, and organizational
studies.
Xiaodan Yu
Xiaodan Yu is an assistant professor in the Department of Entrepreneurship and In

novation at Nottingham University Business School, China. Prior to joining NUBS,
she held research positions at the National Research Council and Scuola Superiore
Sant’Anna in Italy. She received her BEng from Beihang University in China, MSc
from University of Cincinnati, and PhD in economics from Scuola Superiore
Sant’Anna in 2014. Her research interests include the economics of innovation, in
dustrial dynamics, firm growth, and entrepreneurship.
Page 64 of 64
Neoclassical Economic Perspectives on Industrial Policy
Neoclassical Economic Perspectives on Industrial Poli

cy
John Weiss
Conventional welfare economics has always recognized that the real world is character
ized by market imperfections, which provide a rationale for policy intervention. The early
analysis of this focused on tax-subsidy solutions and the implications of domestic ‘distor
tions’ for the case for free trade. The full implications of market imperfections for policy
were brought out clearly in the early 1970s by Corden’s concept of the ‘policy hierarchy’.
Any given distortion could be addressed in a variety of ways, the preferred policy being
that which minimized by-product costs. In practical terms this translated into interven
tions that aimed directly at the source of the distortion and did not discriminate between
possible beneficiaries. Whilst un-influential during the structural adjustment era, this
type of framework underlay the revival of a neoclassical perspective on industrial policy
in the early 2000s. This chapter reviews this literature and the planning techniques and
policy recommendations arising from it.
Keywords: industrialization, industrial policy, market failures, distortions, trade
5.1 Introduction
IN his book on economic development, Ian Little, perhaps the most prominent neoclassi
cal author on development of the last century, characterized the neoclassical approach to
economic development as a belief in the power of the price system to allocate resources
efficiently and by implication to stimulate growth. Little labelled those who questioned
the ability of markets to function in this way ‘structuralists’ (Little, 1982). Theoretically,
neoclassical economics is associated with the concept of equilibrium and the achievement
of allocative optimality through the market mechanism, which in policy terms implies a
preference for laissez-faire and free trade. While much of modern economics is based on
these ideas, once one allows for exceptions to these simple policy prescriptions based on
the existence of market imperfections or ‘market failures’, the meaning of the term ‘neo
classical’ becomes less clear, and the most common use of the term is now as an alterna
Page 1 of 29
tive to heterodox positions (Colander, 2000).1 This chapter focuses on what can be seen
as ‘mainstream’ interpretations of industrial policy, which acknowledge that the existence
of market imperfections or (p. 126) ‘failures’ undermines simple policy prescriptions based
on the assumptions of the pure competitive model.
The limitation of markets as a means of fostering development was an assumption com

mon to most early writers on problems of economic development. In the development
context, key aspects of market failures included the inability of labour markets to clear at
a socially acceptable wage rate associated with open unemployment and underemploy
ment; lack of information and collateral in credit markets leading to unsatisfied demand
for credit, particularly from small and medium-size firms; shortages of skills with under
investment in training by individual producers because of the externality created for com
petitors; under investment in new technology because of lack of capital, financial and hu
man, and the externality for non-innovating competitors. In principle all of these justify
intervention to address the gap between market prices and economic or social values
(sometimes termed ‘shadow prices’).
Neoclassical industrial policy is usually interpreted as a minimalist approach to correc

tions to the functioning of markets through measures such as the provision of better in
formation, regulation of monopolies, investment in infrastructure, and the loosening of
government regulation. This is exemplified by the ‘business environment reform’ agenda
introduced as part of structural adjustment programmes in the 1980s and 1990s, when
the term ‘industrial policy’ did not figure in discussions around the Washington Consen
sus (Williamson, 1993, 2004). This chapter discusses the tradition in welfare economics
that can be used to justify a more interventionist version of industrial policy, and consid
ers more recent restatements of the case for industrial policy based on a framework ini
tially developed in the 1950s.2 The chapter begins with an overview of the neoclassical
approach before discussing its neglect during the structural adjustment era. It then con
siders two recent contributions that revive the case for a neoclassical version of industri
al policy. The application of such policies, in particular priority-setting techniques, is also
discussed. A final section draws some conclusions.
5.2 Theory of Domestic Distortions and the Pol

icy Hierarchy
Neoclassical versions of economics have been dismissed as dogmatic assertions of the
benefits of laissez-faire and free trade, but within the theoretical framework there is the
possibility of an interventionist, if still limited, role for public policy. Building on (p. 127)
Marshall’s idea of external effects, for which the initiating producers neither received
recompense nor paid costs, Pigou (1920) created the case for tax-subsidy interventions as
a means of addressing the failures of markets. With reference to the taxing of ‘smoke nui
sance’, he anticipated the case for environmental taxes by several decades.3 The applica
tion of these ideas to a broader policy context had to wait until the 1950s, when a theoret
ical literature on policy in inefficient (or ‘distorted’) markets emerged from analysis of the
Page 2 of 29
conditions under which free trade remained the best policy, where competitive conditions
did not hold domestically. Although generally argued in abstract terms, the issue was
highly relevant in the development context as trade protection was a relatively easy way,
at least in fiscal terms, for governments to compensate producers for factors like wage
rates above the opportunity cost of labour or for externalities created by learning effects.
This literature was termed the ‘theory of domestic distortions’, and seminal contributions
were made by Meade (1955), Bhagwati and Ramaswami (1963), and Johnson (1965).
However, it was Max Corden who generalized the approach beyond trade-related inter
ventions to cover a range of possible measures to address different market failures in
what he termed the ‘policy hierarchy’ (Corden, 1974: ch. 2). This offered the first general
statement of the principles behind what can be seen as a neoclassical version of industri
al policy, which in principle could be used to justify a relatively interventionist policy
stance.
Corden recognized that policy needs to address barriers to growth created by real-world
features of imperfect markets, but argued that there are different ways to do this associ
ated with different by-product costs in terms of their effects on incentives and prices in
related markets. Policy will involve a trade-off between addressing a specific market fail
ure and the costs arising from the intervention.4 The best solution for a given problem
was judged to be the most direct—in the sense of addressing the specific market failure
at source—as this would minimize distortionary by-product costs. Policy interventions
could thus be ranked by the number (or more accurately) the scale of the by-product dis
tortions they created. One of Corden’s original examples related to a labour-market dis
tortion (such as a minimum wage or union pressure) which kept wages above labour’s op
portunity cost. If labour-market reforms were ruled out as impractical, the most direct
policy would be a subsidy for employment (for example through tax credits). Other alter
natives, such as a subsidy to production, not employment, or import protection via tariffs
or export subsidies, created increasing levels of by-product costs. A production subsidy
would leave labour-intensity of production too low; an import tariff combined with an ex
port subsidy would have the same effect and distort consumer choice; and an import tar
iff alone would leave labour-intensity of production too low, distort consumer choice, and
bias sales against exports (Corden, 1974: 29). Corden was clear that similar analyses
could be applied to different policy problems and combinations of policy instruments. For
example, Corden (1985) anticipated more recent discussions on the role of the real ex
change rate as an instrument of industrial policy, contrasting ‘exchange rate protection’
with the use of import tariffs as means of encouraging tradable activities.5
Page 3 of 29
Table 5.1 Policy hierarchy: industry support by alternative measures
Rank Policy measures By-product distortions
1 Production subsidy None
2a Import tariff plus export sub Consumption distortion

sidy
2b Subsidy to inputs Factor intensity distortion
3 Import tariff Consumption distortion
Anti-export bias
4 Import quotas Consumption distortion
Anti-export bias
Public revenue effect
Note: Items at 2a and 2b are equivalent in ranking since they have the
same number of by-product distortions.
Source: Adapted from Cody et al. (1990: table 4.1).
Neoclassical economics as applied to development policy generally recognized infant in

dustry policy as a possible exception to free trade; such a policy was viewed as a form of
investment, with short-run costs to consumers as costs of imported goods are raised by
import tariffs and non-tariff barriers being compared with long-run gains in productivity
through successful learning (Baldwin, 1969).6 However, support for an infant can be pro
vided in different ways. In terms of supporting new activities, following the logic of the
policy hierarchy import tariffs were perceived as an inferior means of promoting a new
activity because they distorted consumer choice and created a bias against exports by
raising the relative profitability of home-market sales. The most direct intervention would
be a production subsidy financed by raising taxes in as non-distortionary a way as possi
ble.7 Table 5.1 ranks five alternative policy measures for supporting an infant activity,
with the hierarchy determined by the number of by-product distortions each creates. A
combination of an import tariff and export subsidy (p. 129) at the same rate would avoid
an anti-export bias but would create a consumption distortion by raising prices paid by
consumers. A subsidy to a factor input, such as labour, power, or credit, would not affect
consumer prices but would distort input selection. An import tariff would have an anti-ex
port bias and would distort consumer choice. Finally, physical non-tariff barriers like quo
tas, prevalent in many countries in the 1970s, were perceived as the most distorting of all
Page 4 of 29
forms of protection, creating not only biases in consumer choice, but also an anti-export
bias and a loss of government revenue.8
The policy hierarchy approach has been and remains influential, although it has clear lim
itations. It is in the tradition of the theory of the second best, but establishing exactly
what optimal measures will be in a realistic world full of market imperfections is theoreti
cally highly complex, and the policy hierarchy rule is a way of simplifying the problem
(Lipsey and Lancaster, 1956). Ranking inversely by number of by-product distortions a
measure creates assumes both that each measure is equally effective in terms of its im
pact on the chosen policy objective and that the costs of each distortion are equal. Logi
cally, where these conditions do not hold it is possible to prefer a more distortionary, but
more effective policy over a less distortionary, but less effective one (Cody et al., 1990). In
addition, the approach assumes away the fiscal costs and their associated distortionary
effects where taxes are raised to fund subsidies, by assuming subsidies are financed by a
tax package that minimizes tax distortions.9
The general point that the unexpected by-product effects of policies can be significant is
important, but the policy hierarchy approach alone cannot provide a definitive choice be
tween alternatives. Their relative effectiveness and their costs of implementation and
funding need also to be considered in a simple overall cost‒benefit comparison. One of
the key central arguments over trade protection, for example, has been that it can have a
direct effect in blocking imports into the home market and that it does not require fund
ing in the way that promotion by subsidies does. Despite these limitations, as discussed
further below, the influence of the hierarchy approach is still found in policy discussions,
which stress the importance of addressing a market distortion or failure as directly as
possible.10
5.3 Structural Adjustment Era

The significance of market failures was largely ignored during the structural adjustment
era of the 1980s and 1990s, with the counter-argument that, while markets can (p. 130)
fail, so can governments, in the sense that they can be captured by vested interests so
that interventions become a means of delivering rents to favoured activities. The avoid
ance of cronyism, it was argued, required the avoidance of any major interventions to ad
dress market failures (Krueger, 1974). Only minor interventions, for example, to improve
information flows, restrict administrative barriers, and improve infrastructure, were ac
ceptable.
In relation to foreign trade, there was unanimous support for removing non-tariff barriers
and lowering the mean and dispersion of tariff rates, with the possibility of retaining
above-average tariffs for activities with potential (the so-called infants). Low relatively
uniform import tariffs were advocated, largely for revenue purposes; for example, in his
summary of the Washington Consensus, Williamson (1993) refers to rates in the range of
10‒20 per cent, with the lower end more likely to be applied.
Page 5 of 29
In terms of wider industrial policy issues, along with opening economies to foreign com
petition through major trade liberalization programmes, the thinking was that policy re
form should aim at making markets work as competitively as possible by removing distor
tions directly. This meant that rather than accepting market failures and designing sec
ond-best policies based on taxes, subsidies, or other measures to work around them, the
failures themselves should be removed. Put simply, for example, this meant that if wages
were above the opportunity cost of labour, impediments to the free functioning of labour
markets, such as restrictive legislation on hiring and firing, should be removed; alterna
tively, if firms had difficulty accessing credit, the system of financial intermediation
should be improved by removing restrictions on the entry of new banks or by other regu
latory measures to increase competition. The imposition of this type of liberalization re
form across the major markets of economies as part of structural adjustment conditionali
ty was highly controversial. For the present discussion, what is important is that if imple
mented effectively it removes the need for the type of industrial policy implied by the poli
cy hierarchy discussion.
What replaced industrial policy at this time was a focus on barriers to private-sector ex
pansion as identified in ‘investment climate’ surveys. These were based on interviews
with firms who were asked to rank the importance of different types of constraint on their
growth, whether, for example, poor infrastructure, lack of finance, government regula
tion, or lack of skills. Specific measures to address these constraints, within the confines
of an overall policy of liberalization, sometimes described as ‘business environment re
form’ became the new industrial policy of this era (Weiss, 2013). The rationale for such
measures was based on empirical evidence collected by the World Bank and others on the
impact of various barriers to private-sector expansion (World Bank, 2005). The basis for
policy at this stage was to overcome these various obstacles by investing in infrastructure
and training, streamlining regulation, where possible lowering taxes, and improving the
system of financial intermediation by banking reform.
Improvements were as far as possible to benefit all producers equally through ‘horizon
tal’ interventions (as opposed to selective or ‘vertical’ measures). However, in practice a
truly level playing field approach was difficult to implement, since differential impacts
were inevitable as improvement in access to a given resource or service would (p. 131)
benefit more those who used the resource or service intensively. In addition, firm surveys
revealed that across firms from many countries there was a large gap between de jure
and de facto business environment conditions, which in many instances was explained by
corrupt or informal payments as means of evading controls. This meant that there could
be large differences in the business environment between firms in the same country.11
At this stage, therefore, the policy consensus was dominated by the primacy of market re
form and the need to remove market failures and distortions at source. Behind this ap
proach was a conception of the state as a facilitator for the private sector rather than as
an active participant in a process to support long-term development, with its perceived
risk of government failure and cronyism. However, a major omission of this version of pol
Page 6 of 29
icy, highlighted in more recent contributions to the neoclassical literature, is its failure to
address issues of innovation and technical change.
5.4 Revival of Interest in Neoclassical Industri

al Policy after 2000
If one believes that market failures in poor countries cannot be adequately addressed
simply by measures to improve the business environment, the market failure/policy hier
archy framework, while largely neglected in the period from the mid-1970s to 2000, can
be used to justify a considerably more active policy than that practised during that time.
From a neoclassical perspective, the revival of interest in industrial policy is largely due
to the work of the Harvard-based economists Ricardo Hausmann and Dani Rodrik. Al
though the authors themselves depart from conventional (and therefore neoclassical) in
terpretations of causes of growth, their work follows the traditions of the policy hierarchy
literature. Their theoretical approach is set out clearly in Hausmann and Rodrik (2003),
and its policy implications are explored in a series of subsequent papers. The authors re
ject the standard interpretation of the route to economic growth set out by the Washing
ton Consensus involving openness to foreign trade and capital flows, market liberaliza
tion, and improved governance and institutional development. They argue that while
some of these measures individually may have positive effects, standard reform packages
miss the key issue of dynamic entrepreneurship and its link with innovation and ‘product
discovery’. To address this, additional support measures are required to help firms dis
cover what they can produce competitively. This opens the door to a version of industrial
policy that is based on dialogue between public agencies and representatives of the pri
vate sector and has been labelled ‘modern industrial policy’ (Felipe, 2015).
The more ambitious side to this policy (what the authors term ‘in the large’) in
(p. 132)
volves taking strategic bets on new activities, products, or processes and providing the
risk capital—for example through development banks or venture capital funds—or loan
guarantees to allow them to be established.12 The authors base this on the externality ef
fect created by innovation, defined as doing things in different ways and creating new
products (or products new to an economy), since followers benefit from the cost incurred
by the first entrant. This gives a rationale for supporting and subsidizing innovation, and
while not referring explicitly to the policy hierarchy approach, their discussion of the
costs and benefits of different policy options largely mirrors the discussion in Corden
from thirty years earlier.13 In their analysis, import protection as a means of supporting
innovators is the least attractive option. It does not discriminate between innovators and
followers, raises prices to consumers, and has an anti-export bias. It will focus innovation
on domestic markets which is expected to lower the return to innovation given their small
size relative to the world market. Similarly, export subsidies may have weaker price ef
fects for domestic consumers than import tariffs, but also do not discriminate between in
novators, leading to over investment in the activity. As the key concern is to target inno
vators, while limiting any leakage of benefits to ‘copycats’, public-sector credit or loan
Page 7 of 29
guarantees is the preferred instrument for this. Loans or loan guarantees can transfer the
risk of failure to governments as the loan is only repaid if the investment is successful.
In subsequent papers it is clear that the authors also see the possibility of industrial poli
cy helping to remove constraints to growth in existing activities (what the authors term
‘in the small’) and fostering linkages between producers and suppliers (Hausmann and
Rodrik, 2006; Hausmann et al., 2005, 2008). Here dialogue between the government and
the relevant private-sector stakeholders is meant to identify both areas for expansion and
the obstacles to this growth that can be removed by the government. Co-funding by pri
vate partners to remove bottlenecks is seen as helpful to reinforce the public‒private
partnership element.14 Furthermore, Rodrik has made clear that he sees what Corden
(1985) called ‘exchange rate protection’ and others have (p. 133) called a ‘stable and com
petitive real exchange rate’ (Guzman et al., 2018) as an important component of industri
al policy. He and other authors have shown a link between real exchange rates and eco
nomic growth (Rodrik, 2008). This association has been built on in recent discussions of
industrial policy where setting a competitive rate (that is an undervalued rate for local
currency) is seen as an important complement to supply-side interventions. The criticism
that this implies supporting all tradable activities, not just those with learning externali
ties, can be addressed by a set of taxes on the non-externality-creating traded sectors.15
The focus on innovation and subsidizing innovating firms from Hausmann and Rodrik
links with both the endogenous growth literature (Romer, 1990) and the capabilities ap
proach to industrialization as seen, for example, in the discussion of national systems of
innovation in Nelson (1993) and the work of Lall (1992) on industrial capabilities.16 It has
been argued that if knowledge is firm specific or tacit then it will not be transferrable and
that Hausmann and Rodrik therefore overplay the importance of knowledge transfer ef
fects and underplay the importance of inter-sectoral linkages and systemic effects (An
dreoni and Chang, 2019: 140). The empirical significance of the spread of a knowledge
externality partly depends on what is to be transferred. The argument of Hausmann and
Rodrik about innovation and product discovery appears to be a broader one, in that firms
may follow an innovator in a particular product line or way of organizing production with
out having specific blueprints, as the key information that is transferred concerns what
can be produced competitively in an economy. In relation to inter-sectoral linkages and
systemic issues, such as fostering a culture of entrepreneurship and factory discipline,
the charge that these are neglected overlooks the process or dialogue aspect of the au
thors’ policy, that is intended to establish the constraints to growth in a particular area,
which can include lack of domestic input suppliers or a trained workforce. In so far as
these are real issues and governments can relieve these constraints, concerns over link
ages and aspects of the industrial culture should be addressed. The dialogue element of
their policy introduces a potentially more interventionist slant to industrial policy absent
from the ‘business environment reform’ approach.
The most valid criticism of the Hausmann and Rodrik approach is that offering credit or
credit guarantees to firms may not itself provide sufficient incentive for or (p. 134) sup
port to the development of indigenous technology. The national system of innovation per
Page 8 of 29
spective views firms’ technological capability as strongly influenced by their operating

environment, including the education system and research institutions, and policy on for
eign investment, technology transfer, and intellectual property. Here the role envisaged
for industrial policy is much wider than supplying credit, but requires governments to
fund and coordinate basic research, to support its dissemination and commercial develop
ment and support firms in their efforts at technological upgrading. Case studies from
technologically successful economies reveal the key common denominators as learning
within firms, access to foreign technology and skilled human capital, and an active gov
ernment policy (Malerba and Nelson, 2011: 1666). Similarly, evidence from East Asia on
the development of electronics and information communication technology highlights the
role of government support (Lee, 2013).
5.5 New Structural Economics

A separate body of work influencing the neoclassical revival of industrial policy comes
from the Chinese economist Justin Lin, who gained international prominence and influ
ence from his time as chief economist of the World Bank. Lin’s case on industrial policy is
part of a wider framework of ‘New Structural Economics’, which aims to marry a belief
that economic structures are critical for development with an emphasis on the use of the
price mechanism as the key to efficient resource allocation. The author himself terms this
‘a neoclassical approach to structure’, as opposed to traditional structuralist analysis with
its suspicion of the effectiveness of markets (Lin, 2011: 193). The significance of this work
for industrial policy is that it acknowledges the existence of market failures and thus re
vives the case for government intervention through a ‘facilitating state’. As in the Haus
mann/Rodrik work, the importance of informational externalities created by pioneer
firms, whose experience both positive and negative followers can learn from, is stressed.
In addition to ensuring that pioneers are compensated in some way for these effects, the
government’s roles cover provision of information to and coordination of private firms, in
vestment in both hard and soft infrastructure and the nurturing of new potentially com
petitive industries, chiefly through foreign investment.17 Each of these roles are said to
stem from a perceived market failure, so the argument is consistent with earlier neoclas
sical work.
In terms of practical advice on the direction of structural change, Lin draws in
(p. 135)
sights from the ‘flying geese’ pattern of migration within East Asia, as industries moved
between economies in the region in response principally to wage differentials: with rising
wages in Japan in the 1980s, firms migrated to lower-wage locations like Thailand and In
donesia, and more recently to China and Vietnam (Lin, 2013). This pattern of migration
lies behind his suggestion that when poor countries start to think about new sub-sectors
in which to invest, they should start by examining sub-sectors in which similar countries
have had export success in the recent past. A similar dialogue to that in Hausmann and
Rodrik of identifying obstacles to expansion by existing firms or barriers to entry by new
firms is recommended.
Page 9 of 29
In this approach new activities to be supported by governments should be those that are
in line with existing factor endowments of an economy, but which are not yet profitable
for private firms because of short-term market failures. The expectation is that the re
moval of these failures will be sufficient to create internationally competitive production,
so that the activities can be described as those where there is a latent, but not current,
comparative advantage.18 Hence investment in activities such as these is described as
‘comparative advantage conforming’. Investing in activities which require skills and capa
bilities that the economy does not yet possess is termed a ‘comparative advantage defy
ing’ policy by Lin, who sees this as a characteristic mistake of traditional structuralism
and by implication ‘old-style’ industrial policy.19 Investment in activities without latent
comparative advantage needs to be postponed until the underlying endowment structure
has changed sufficiently for them to be commercially viable with the support of the envis
aged market-correcting industrial policy.
The type of market-correcting industrial policy interventions envisaged by this approach

is illustrated in a detailed study on the obstacles to light manufacturing in Africa, which
looked at the constraints on development in a range of sub-sectors in several East African
countries, with Ethiopia taken as a case study (Dinh et al., 2012). This focus on con
straints can be interpreted as ‘industrial policy in the small’ in the terminology of Haus
mann and Rodrik. The analysis is based around a form of sub-sector growth diagnostics
highlighting what are identified as the key constraints and solutions to them based on
cross-country cost comparisons and evidence from interviews.20 (p. 136)
5.6 Doomed to Choose: How to Implement

Selectivity
Since it draws its rationale from the existence of significant market failures, neoclassical
industrial policy is usually associated with horizontal as opposed to vertical measures
available only selectively to priority activities. However, both Hausmann and Rodrik and
Lin indicate that some form of selectivity is inevitable to allocate limited resources ex
ante and both agree that perfect neutrality of outcome is impossible even if all policy is
based ex ante on neutral, horizontal interventions.21
In principle, horizontal measures supporting innovation, available to all but with an in-
built element of selectivity in that they offer benefits only to dynamic firms offering new
products or processes of production, could be said to offer the greatest scope for neutrali
ty. Hausmann and Rodrik are a little ambivalent on the balance between horizontal and
vertical measures, although in one paper (Hausmann and Rodrik, 2006) they argue that
due to limited resources and technical capacity governments are ‘doomed to choose’ the
areas or firms to support. In Hausmann and Rodrik (2006: 79) they suggest that ‘in prin
ciple, interventions should be as horizontal as possible and as sectoral as necessary’.
Their key point is that measures—like support for innovation—that enhance growth can
span a range of sectors and therefore access to these should not be restricted selectively.
However, they acknowledge that in some circumstances it may be easier to support spe
Page 10 of 29
cific sectors, both as a way of co-ordinating firms and because different sectors have dif
ferent needs.22
As noted above, a key argument against selectivity is the risk of cronyism and policy cap
ture, which mean that differential support is not offered on the grounds of potential for
competitiveness. While the neoclassical policy literature is cautious of the concept of pri
ority areas, its emphasis on quantification and comparison of costs and benefits offers
ways of guiding choice, which can be used in quite different policy settings with varying
degrees of ambition for industrial policy. A starting point could be existing trade special
ization as shown in revealed comparative advantage indicators. However, this has no dy
namic component and does not in itself help to highlight future areas for growth.
Of similarly limited value is the concept from the neoclassical growth literature of total
factor productivity (TFP). TFP is meant to reflect efficiency gains, interpreted as techni
cal progress. While used widely in growth-accounting exercises at the macro level, for
some countries estimates are available at the sub-sector level and firm level, which in
principle could be used to set priorities in resource allocation.23 It has been (p. 137)
known for some time that there is a wide variation in productivity levels between firms,
particularly in low- and middle-income economies, so that policies that can reduce the
gap have potentially high returns. However, there are empirical difficulties with estima
tion of TFP and controversy over what the unexplained residual actually captures. In
practice, in terms of priority setting, given the high correlation between value-added
growth and TFP, setting priorities on the basis of such estimates would be simply assum
ing that past growth will be continued into the future.24
Broad guidance might be found by looking at past patterns of development in the spirit of
initial work by Chenery and his co-authors (Chenery and Syrquin, 1975; Chenery et al.,
1986). This has been updated by more recent work at UNIDO, for example distinguishing
‘early’, ‘middle’, and ‘late’ manufacturing sub-sector branches by the level of real GDP
per capita at which they reach their peak share of GDP (Haraguchi, 2016: table 3.1).
However, as a guide to policy it may be of relatively little use to know that Food and Bev
erages peaks at a relatively low-income level or that Fabricated Metals is usually signifi
cant in middle-income economies.
More specific, but still relatively general, guidance is provided in the simple rule of
thumb put forward by Lin and Monga (2010). Any individual country should target activi
ties that a country with similar structural characteristics and resource endowments and
an income per capita of about double that of the country concerned (in purchasing power
parity) has been able to export successfully in the past fifteen to twenty years. The argu
ment is that the comparator country will have specialized in activities likely to require
production capabilities that the follower country has or can develop readily, so that these
are activities in which the follower has a latent comparative advantage. This is on the
grounds that income per capita provides a proxy for skill sets and incomes more than
double those of a country imply a currently unattainable skill set. As the comparator
economy grows its wages will rise, potentially freeing space in the global market that the
Page 11 of 29
follower economy can fill. Industrial policy is therefore a two-stage procedure: finding a
comparator economy with an income not too far ahead of the country concerned; and
then identifying sectors in the comparator from which the follower economy can export
successfully (Lin and Wang, 2015).
The rule of thumb from Lin and Monga has the advantage of simplicity, but it is not clear
how far it is possible to generalize this rule. It may make sense in relation to standard
ized, labour-intensive products, where wage costs dominate competitiveness. For exam
ple, in relation to garments, with rising wages in older-established production centres
and the ending of the Multi Fibre Arrangement, new important exporter firms from coun
tries such as Bangladesh and Cambodia have emerged. In sectors where technical change
is rapid, however, products exported twenty years ago may have undergone significant re
design. Their technical content and the (p. 138) required production capabilities may have
changed as a result, making it less clear that the comparator‒follower path is straightfor
ward as a guide to industrial targeting.
Furthermore, based on East Asian experience, the relevance of the guidelines of twice
per capita income and fifteen to twenty years’ export experience in the comparator coun
try has been questioned. It is argued that effective industrial policy made it possible for
firms in Japan and Korea to move into new industries which, on a simple reading of the
rule, would be viewed as too ‘distant’ from their existing capabilities.25 Such qualifica
tions do not invalidate targeting export industries that are successful elsewhere, but they
do imply that first, the simplicity of the rule needs to be qualified and second, even where
it can be said to exist, latent comparative advantage may need to be supported through
active policies, for example on technology and training, as well as significant financial
support for the nascent activities, whether these policies are identified by the policy hier
archy approach or by other means.
Other more detailed approaches to the identification of priorities are available from the
neoclassical literature. Here we elaborate further on two approaches that continue to be
used in policy planning: one traditional, based on measures of economic efficiency, and
the other newer, based on ideas relating to the capability required for the export of differ
ent types of good.
5.7 Efficiency Indicators

Different versions of an efficiency indicator have been used as a guide to potential com
parative advantage of either a particular branch of manufacturing, or of individual project
proposals. In the 1960s and 1970s, the neoclassical literature on industrialization formal
ized a version of economic cost‒benefit analysis based on the fact that manufacturing
goods are internationally tradable—with the option of importing or producing domestical
ly for either export or domestic sale. Thus, for any individual manufacturing investment,
benefits from foreign exchange savings by import substitution or generation by exports
could be compared with the opportunity costs of the resources associated with their do
mestic production (Little and Mirrlees, 1968, 1974; UNIDO, 1972). Adjusting prices used
Page 12 of 29
in the analysis to reflect opportunity costs (the use of shadow prices) meant that in princi
ple the effect of key market failures—such as an overvalued wage or an undervalued ex
change rate for foreign currency—would be removed from the analysis to allow an assess
ment of whether an investment was economically efficient over the operating life of the
project. Application of the (p. 139) methodology was part of a wider drive to shift industri
alization strategy away from import substitution behind protective barriers towards a
greater focus on exports. Projects with rates of return below a cut-off opportunity cost
discount rate were to be rejected with the implication that they were activities where
over the project’s life there was no comparative advantage.
In terms of empirical work on trade policy and industrialization, the most widely used
neoclassical technique of the 1960s and 1970s was the effective rate of protection (ERP),
which was used primarily to show the distorting effect of import-substitution policies (Lit
tle et al., 1970; Balassa, 1982). As this showed the effect of trade policy in inflating value
added above what it would have been in the absence of protection, it was strictly a mea
sure of incentive, not of efficiency.26
An offshoot of cost‒benefit analysis, the domestic resource cost (DRC) ratio is a more ap
propriate measure of trade efficiency that provides a relatively simple indicator to assess
comparative advantage (Schydlowsky, 1984). Developed originally for planning decisions
in Israel, it offers a direct measure of short-term trade efficiency and has been used to
demonstrate the high costs of some import-substitution activities (Bruno, 1972; Krueger,
1966). It estimates the domestic resources required per unit of foreign exchange earned
or saved by an activity, in an activity-specific exchange rate.27 Using the DRC, a test of
comparative advantage requires that the domestic resources involved when valued at
their economic opportunity cost are less than the value of net foreign exchange generat
ed by an activity. In policy discussions, relatively high DRC measures were also used to
justify policies of trade liberalization. Although the inverse ranking of activities by the
size of their DRC is only strictly valid under restrictive assumptions, such as constant re
turns to scale, activities with ratios well above unity or a DRC ratio well above the shad
ow exchange rate, will not be activities in which an economy has a current comparative
advantage. Unless there is strong evidence that their productivity (p. 140) can be im
proved markedly over the life of new investments, expansion of these will be comparative
advantage defying in the terminology of Lin (2011).28 The DRC ratio thus provides a tech
nique for distinguishing the short-term competitiveness of tradable activities.
The DRC analysis gives a snapshot picture of efficiency at a point in time, and where ac
curate data can be collected it highlights where there is current or short-term efficiency.
Its weakness as a guide to priorities is that ideally what is required is an indication of
longer-term potential that is dynamic, not static, comparative advantage. In principle an
element of dynamism can be introduced by projecting both sides of the DRC ratio into the
future; for example, learning or technical change may reduce the domestic resources in
volved (and thus lower the numerator) or improved demand prospects or design may in
crease the net foreign exchange effect (and thus raise the denominator). The exchange
rate used should be a projected equilibrium value and both future costs and benefits
Page 13 of 29
should be discounted to give present values. With further adjustments the analysis can be
converted into a full cost‒benefit calculation. Any benefits from employment creation are
picked by the lower valuation of surplus or underemployed labour either used directly in
production or indirectly though the production of non-traded inputs. If externalities are
involved, depending on their sign, they will add to or reduce the domestic resources in
the ratio.29 All of these adjustments are uncertain, particularly when big structural leaps
are taken into completely new areas. Thus, despite the possibility of projecting into the
future, DRC analysis as applied in practice is essentially static. Hence heterodox or struc
turalist authors tend to dismiss this approach as either impractical or having an in-built
bias towards unambitious interventions (Chang, 2009).
5.8 Product Space

Hausmann and Rodrik (2003) highlight the need for countries and firms to discover what
they are good at producing; however, practical policy requires some guidance on what
this is likely to be and where dynamic comparative advantage lies. An addition to their
work, ‘product space’ analysis, is intended to offer this. In the short term it is likely that a
country will move into the production of goods that require similar skills, technology, and
capital assets to those in which it is already competitive, but the authors argue that the
conventional approach of simply looking at areas where there is already a revealed com
parative advantage is insufficient. Within broad product (p. 141) categories (like textiles or
clothing) there can be significant diversity between individual products in terms of tech
nology content and branding. This implies that price, profit margins, and demand
prospects can differ between products within the same broad category, offering scope for
a country to upgrade into higher-value products. Patterns of trade specialization can af
fect growth and there is evidence that exporting higher-value or more sophisticated prod
ucts is associated with faster economic growth, so that countries whose export basket is
more sophisticated than expected for its income level experience faster growth (Haus
mann et al., 2007, 2011).
If more sophisticated products are to be targeted by policy, a measure of sophistication is

required at the individual product level to assess when changing the composition of ex
ports adds to growth potential. Initial measures of the sophistication of products used a
weighted average of the income per capita of countries which exported the goods, so the
average income of the exporters became a proxy for the technological depth, complexity,
and thus growth potential of a product.30
Use of an income-based product-level indicator to explain growth of income across coun

tries risks being tautological, hence a revised measure of product sophistication—termed
‘product complexity’—was developed (Hidalgo et al., 2009; Hausmann et al., 2011). This
index number combines information on diversification—the number of products a country
exports with revealed comparative advantage—and ubiquity—the number of countries
which also specialize in these products. Complex products with a high score on this index
are those where diversified economies with high levels of capability have a
Page 14 of 29
specialization.31 Rich countries are expected to export complex goods that few other
countries export, while poor countries are expected to export goods of less complexity ex
ported by many other countries. The most complex goods by this measure are machinery,
chemicals, and metal products and the least complex are raw materials and primary com
modities. As expected, there is a positive association between a country’s income per
capita and the share of complex goods in exports.
Product space analysis is based on the idea that there is a trade-off between moving to
‘close’ goods—which may not be highly sophisticated or complex—but which require ca
pabilities close to an economy’s current set, and ‘distant’ goods with a higher potential
value. This requires a measure of the economic distance between goods, which Haus
mann and Klinger (2007) base on the lower of the two probabilities that a good is export
ed with comparative advantage by pairs of countries.32 The argument is that close prod
ucts require similar production capabilities for competitive production and that a country
that specializes in goods that are close to other goods will find it easier to diversify its ex
port basket. Export targets will be goods which add to the (p. 142) complexity of a
country’s overall export basket but are not so far away in terms of capabilities required
for their production that producing them competitively is not feasible in a realistic time
frame.
Product space analysis is intended to identify potential export products at a disaggregate

level. The total product complexity of a country’s exports is the product complexity index
for each good weighted by the share of goods in total exports. Distance is determined by
the probability data on joint exports noted above, but what is not too far away is ambigu
ous. In practice crude approaches tend to be applied, for example, to require that target
products have a distance from the current export basket which is below the median dis
tance for all goods exported currently without revealed comparative advantage.33 An effi
cient frontier is defined by goods which add to the product complexity of the export bas
ket, meet the distance criteria, and add to the links with other complex products.34
The products identified in this way should require capabilities for competitive production
not too dissimilar from those required in areas where successful specialization has al
ready been achieved, so the implication is that these are areas where there is a potential
or latent comparative advantage. However, the authors stress that it is not designed to
‘pick winners’ as candidates for export subsidies or other forms of support (Hausmann
and Klinger, 2006, 2008). The idea is that the export priorities implied by this analysis
identify areas for consideration, usually at the four-digit level, in discussion between pub
lic agencies and potential investors, including foreign investors.
Product space analysis has been applied in a highly diverse set of countries and a limita
tion is that in some cases potential export products which emerge from this analysis cov
er a wide range of goods and simply match a priori expectations.35 For Rwanda, for exam
ple, using this approach for exports to global markets, seventy-two potential export prod
uct categories at the four-digit level are identified (Hausmann and Chauvin, 2015). Since
they use local materials, are generally labour intensive, and have below-average trans
Page 15 of 29
port costs, they are the type of goods that might be suggested by an a priori analysis of
‘early’ industries based on past international experience. Thus, in this case it is not clear
what the product space analysis has added.36 However, with the development of a large
database of trade statistics and software to derive the relevant indicators, it has become
much easier to apply this approach as a first screening of possible candidates for new ex
ports, some of which might be of interest to private investors.37
5.9 Do these Calculations have any Role?

Given the uncertainty attached to all future projections and the clear limitations of these
techniques, do they have any useful role in helping to choose areas to support in the con
text of an active industrial policy? Emphasis on quantification can be seen as the strength
of a neoclassical approach to policy, since in its absence decisions will inevitably be taken
on weaker grounds.38 As an efficiency indicator, the DRC will inevitably be approximate
and at best offers indicative guidelines for policy priorities. The most useful role of the
analysis is as an initial screening device providing a loose form of ranking by the inverse
of the DRC ratio—either between manufacturing branches or individual potential projects
—with activities with a high domestic cost per unit of foreign exchange treated as unlike
ly candidates for support unless convincing arguments can be provided to demonstrate
their long-term potential. Short-term costs associated with initially uncompetitive activi
ties should not be ignored for priority setting and the indicator is useful as a means of
gauging by how much productivity would need to improve for certain activities to justify
support and to allow a judgement on how best, if at all, policy can help to bring these
down. It has less of a role where genuinely new activities with no comparable reference
price are under consideration.
The product space analysis aims to link the capabilities needed for goods already export
ed successfully with those needed for new export target niches. Similarity in (p. 144) capa
bilities in itself does not guarantee that efficient production can be established. Efficiency
will be determined by a range of factors (such as access to technology, skilled labour, raw
material supplies, or marketing links). Hence there is a role for industrial policy in engag
ing with producers to establish the constraints that need to be overcome to create com
petitive exports in these areas. The data from a product space analysis should be seen as
a first step in a screening and identification process, with detailed discussions, plans, and
appraisals required prior to final decisions.39
5.10 Conclusions
Overall, the ideas from this newer neoclassical policy literature have been taken seriously
by both international agencies and governments in a number of countries, who see the
need for a form of industrial policy.40 This literature has served to remind policymakers
that government can have an important role in economic transformation that goes well
beyond that set out in the Washington Consensus of the 1980s and 1990s.
Page 16 of 29
As an example of the influence of this revival of interest, the Inter-American Development

Bank has developed an approach to ‘productive development policies’ based on principles
which draw heavily on the ideas behind neoclassical industrial policy, as defined here. De
tails of a variety of interventions are given in Crespi et al. (2014), while chapter 2 of that
volume sets out the conceptual framework involved. The emphasis is on correcting mar
ket failures, addressing these as directly as possible, and conducting an implicit (or at
times explicit) comparison of the costs and benefits involved. The influence of the policy
hierarchy is evident as two tests are set out for policy analysis: ‘What is the plausible
market failure that has been diagnosed to justify the policy? Is the alleged policy remedy
—whether it entails alleviating the failure or redressing its impact—a good match for the
diagnosis?’ (Crespi et al., 2014: 28). The work of Hausmann and Rodrik is also cited, par
ticularly in relation to aiding self-discovery through support for innovation and encourag
ing public‒private dialogue to identify constraints to growth. The examples in the book
can be seen as examples of what Ocampo and Porcile (Chapter 28 in this volume) de
scribe as the ‘timid return’ of industrial policy post 2000.41
There are different ways of justifying industrial policy and mainstream welfare
(p. 145)
economics has always offered one based on market failures. In a development context the
theoretical base for industrial policy was ignored during the era of structural adjustment,
despite earlier work pointing out that real-world features of markets could be corrected
in ways that minimized any negative allocative effects. Recent thinking has revived the
case for industrial policy, as the limitations of market-driven growth as the basis for inno
vation and dynamic structural change have become apparent, with a new emphasis on di
alogue between public agencies and the private sector to determine how the former can
best support the latter to become competitive. However, this neoclassical version of in
dustrial policy has its own limitations. By focusing on price adjustments through markets
it can ignore more effective—but more distortionary—direct measures like import quotas,
investment licences, or direct credit allocations. By highlighting individual markets, it
may miss the bigger picture that supportive macro policies combined with more effective
financial intermediation and an active policy on technology can be critical. By focusing on
short-term costs and thus on activities with latent comparative advantage relatively close
to competitiveness, where existing capabilities are not very different from those required
at the efficiency frontier, this approach misses the possibility of skipping between genera
tions of technology into new areas. Because of uncertainty the quantitative planning tools
of the neoclassical approach can have an in-built conservative bias in that they will select
activities with costs that can be estimated or policy interventions with predictable, if lim
ited, outcomes. In this view, none of the big leaps or breakthroughs associated with the
successful industrialization of the past would have been justified ex ante by neoclassical
criteria.
However, even ambitious ‘vision-based’ or ‘mission-oriented’ versions of industrial policy

which aim at creating winners rather than picking them from existing firms, or creating
markets as opposed to fixing them through tax-subsidy adjustments (Mazzucato and Kat
tel, Chapter 12 in this volume), still require some approximate comparison of benefits and
costs. Experience suggests that industrial success requires a combination of a vision,
Page 17 of 29
well-designed support, and the application of some form of economic criteria to aid deci
sion-taking. The neoclassical techniques referred to here, particularly cost‒benefit analy
sis and its offshoot the DRC ratio, have limitations. They may not be very effective in aid
ing thinking on the most dynamic new areas, but sensible application of the DRC indica
tor, for example, offers a way of weeding out obviously inefficient investments that have
little prospect of competitiveness. Thus, they provide a check on the short-term costs of
different investments associated with different policy options. In this sense there is scope
for a synthesis between competing versions of industrial policy with the focus of specifi
cation and quantification of benefits and costs from the neoclassical tradition combined
with ambitious programmes for industrial transformation that involve more substantial in
stitutional change. (p. 146)
References
Abdon, Arnelyn, Marife Bacate, Jesus Felipe, and Utsav Kumar (2010) ‘Product Complexi
ty and Economic Development’. Levy Economics Institute Working Paper No. 616.
ADB (2016) Tajikistan: Promoting Export Diversification and Growth. Manila: Asian Devel
opment Bank.
Ansu, Yaw (2013) ‘Industrial Policy and Economic Transformation in Africa: Strategies for
Development and a Research Agenda’, in Joseph E. Stiglitz, Justin Yifu Lin, and Ebrahim
Patel (eds) The Industrial Policy Revolution II: Africa in the 21st Century, International
Economic Association Conference Volume 151–11. Basingstoke: Palgrave Macmillan, pp.
492–528.
Balassa, Bela (1982) Development Strategies in Semi-Industrialized Economies.

Baltimore, MD: Johns Hopkins for the World Bank.
Baldwin, Robert (1969) ‘The Case against Infant Industry Tariff Protection’, Journal of Po
litical Economy 77(3): 295–306.
Baranson, Jack (1967) Manufacturing Problems in India. Syracuse, NY: Syracuse Universi
ty Press.
Bell, Martin, Bruce Ross-Larson, and Larry Westphal (1984) ‘Assessing the Performance
of Infant Industries’, Journal of Development Economics 16(1–2): 101–28.
Bhagwati, Jagdish and V. K. Ramaswami (1963) ‘Domestic Distortions, Tariffs and the The
ory of Optimum Subsidy’, Journal of Political Economy 71(2): 44–50.
Bruno, Michael (1972) ‘Domestic Resource Cost and Effective Protection: Clarification
and Synthesis’, Journal of Political Economy 80(1): 16–33.
Page 18 of 29
Chang, Ha-Joon (2009) ‘Should Industrial Policy in Developing Countries Conform to

Comparative Advantage or Defy It? A Debate between Justin Lin and Ha-Joon Chang’, De
velopment Policy Review 27(5): 483–502.
Chang, Ha-Joon (2013) ‘Comments on “Comparative Advantage: The Silver Bullet of In
dustrial Policy” ’, in Joseph E. Stiglitz and Justin Yifu Lin (eds) The Industrial Policy Revo
lution 1, IEA Conference Volume 151–1. New York: Palgrave Macmillan, pp. 39–42.
Chenery, Hollis and Moshe Syrquin (1975) Patterns of Development 1950–70. Oxford: Ox
ford University Press for the World Bank.
Chenery, Hollis, Sherman Robinson, and Moshe Syrquin (eds) (1986) Industrialization and
Growth: A Comparative Study. Oxford: Oxford University Press.
Cody, John, Richard Kitchen, and John Weiss (1990) Policy Design and Price Reform in De
veloping Countries. Hemel Hempstead: Harvester Wheatsheaf.
Colander, David (2000) ‘The Death of Neoclassical Economics’, Journal of History of Eco
nomic Thought 22(2): 127–43.
Corden, W. Max (1966) ‘The Structure of the Tariff System and the Effective Protection
Rate’, Journal of Political Economy 74: 221–37.
Corden, W. Max (1974) Trade Policy and Economic Welfare. Oxford: Oxford University
Press.
Corden, W. Max (1985) ‘Protection and the Real Exchange Rate’, in W. Max Corden Pro
tection, Growth and Trade. Oxford: Blackwell, pp. 302–10. (p. 147)
Crespi, Gustavo, Eduardo Fernando-Arias, and Ernesto Stein (2014) Rethinking Produc
tive Development: Sound Policies and Institutions for Economic Transformation. Inter-
American Development Bank. Basingstoke: Palgrave Macmillan.
Development Policy Review (2009) DPR Debate ‘Should Industrial Policy in Developing
Countries Conform to Comparative Advantage or Defy It? A Debate between Justin Lin
and Ha-Joon Chang’, Development Policy Review 27(5): 483–502.
Dinh, Hinh, Vincent Palmade, Vandana, Chandraand, and Frances Cossar (2012) Light
Manufacturing in Africa: Targeted Policies to Enhance Private Investment and Create
Jobs. Washington, DC: World Bank.
Felipe, Jesus (ed.) (2015) Development and Modern Industrial Policy in Practice.
Cheltenham: Edward Elgar.
Felipe, Jesus and John McCombie (2019) ‘The Illusion of Calculating TFP Growth and
Testing Growth Models’. Available at https://ssrn.com/abstract=3377758.
Guzman, Martin, José Antonio Ocampo, and Joseph E. Stiglitz (2018) ‘Real Exchange Rate
Policies for Development’, World Development 110: 51–62.
Page 19 of 29
Hallward-Driemeier, Mary and Lant Pritchett (2011) ‘How Business is Done and the Do
ing Business Indicators’. Policy Research Working Paper No. 5563. Washington, DC:
World Bank.
Hallward-Driemeier, Mary, Gita Klun Jush, and Lant Pritchett (2010) ‘Deal versus Rules’.
Haraguchi, Nobuya (2016) ‘Patterns of Structural Change and Manufacturing Develop

ment’, in John Weiss and Michael Tribe (eds) Routledge Handbook of Industry and Devel
opment. Abingdon: Routledge, pp. 38–64.
Hausmann, Ricardo and Jasmina Chauvin (2015) ‘Moving to the Adjacent Possible: Dis
covering Paths for Export Diversification in Rwanda’. CID Working Paper No. 294. Centre
for International Development, Harvard University.
Hausmann, Ricardo and Bailey Klinger (2006) ‘Structural Transformation and Patterns of
Comparative Advantage’. CID Working Paper No. 128. Centre for International Develop
ment, Harvard University.
Hausmann, Ricardo and Bailey Klinger (2007) ‘The Structure of the Product Space and
the Evolution of Comparative Advantage’. CID Working Paper No. 146. Centre for Inter
national Development, Harvard University.
Hausmann, Ricardo and Bailey Klinger (2008) ‘Achieving Export-led Economic Growth in
Colombia’. CID Working Paper No. 182. Centre for International Development, Harvard
University.

Journal of Development Economics 72: 603–33.
Hausmann, Ricardo and Dani Rodrik, D (2006) ‘Doomed to Choose: Industrial Policy as
Predicament’, mimeo Harvard. Available at www.ricardohausmann.com.
Hausmann, Ricardo, Cesar Hidalgo, Sebastian Bustos, Michele Coscia, Alexander Simoes,
and Muhammed A. Yildirim (2011) Atlas of Economic Complexity. Cambridge, MA: MIT
Press.
Hausmann, Ricardo, Jason Hwang, and Dani Rodrik (2007) ‘What You Export Matters’,
Journal of Economic Growth 12(1): 1–25.
Hausmann, Ricardo, Dani Rodrik, Jose Miguel Benavente, and Francisco Rodriguez
(2005) ‘Self-discovery in Development Strategy for El Salvador’, Economia 6(1): 43–101.
Hausmann, Ricardo, Dani Rodrik, and Charles Sabel (2008) ‘Reconfiguring Industrial Pol
icy: A Framework with an Application to South Africa’. CID Working Paper No. 168. Cen
tre for International Development, Harvard University. (p. 148)
Page 20 of 29
Hidalgo, Cesar A., Ricardo Hausmann, and Partha Sarathi Dasgupta (2009) ‘The Building
Blocks of Economic Complexity’, Proceedings of the National Academy of Sciences
106(26): 10570–5.
HMG (2017) ‘Industrial Strategy: Building a Britain Fit for the Future’. UK Government,
White Paper.
Johnson, Harry (1965) ‘Optimal Trade Intervention in the Presence of Domestic Distor
tions’, in Richard Caves, Harry Johnson, and Peter Kenen (eds) Trade, Growth and the
Balance of Payments. Chicago, IL: Rand-McNally, pp. 3–34.
Krueger, Anne (1966) ‘Some Economic Costs of Exchange Control: The Turkish Case’,
Journal of Political Economy October: 466–80.
Krueger, Anne (1974) ‘The Political Economy of the Rent-seeking Society’, American Eco
nomic Review 64(3): 291–303.
Lall, Sanjaya (1992) ‘Technological Capabilities and Industrialisation’, World Development

20(2): 165–86.
Lall, Sanjaya, John Weiss, and Jinkang Zhang (2006) ‘The Sophistication of Exports: A
New Trade Measure’, World Development 34: 222–37.
Lee, Keun (2013) Schumpeterian Analysis of Economic Catch-Up. Cambridge: Cambridge

University Press.
Lin, Justin Yifu (2010) ‘New Structural Economics: A Framework for Rethinking Develop
ment Policy’. Policy Research Working Paper No. 5197. Washington, DC: World Bank.
ment’ World Bank Research Observer 26(2): 193–221.
Lin, Justin Yifu (2013) ‘From Flying Geese to Leading Dragons: New Opportunities and
Strategies for Structural Transformation in Developing Countries’, in Joseph E. Stiglitz,
Justin Yifu Lin, and Ebrahim Patel (eds) The Industrial Policy Revolution II: Africa in the
21st Century. Basingstoke: Palgrave Macmillan, pp. 50–72.
Lin, Justin Yifu and Célestin Monga (2010) ‘Growth Identification and Facilitation: The
Role of the State in the Dynamics of Industrial Change’. World Bank Policy Research
Working Paper No. 5313. Washington, DC: World Bank.
Lin, Justin Yifu and Yan Wang (2015) ‘Catching up: Structural Transformation and Diversi
fication’, in Jesus Felipe (ed.) Development and Modern Industrial Policy in Practice. Chel
tenham: Edward Elgar, pp. 127–59.
Lipsey, Richard and Kelvin Lancaster (1956) ‘The General Theory of the Second Best’, Re
view of Economic Studies 24(1): 11–32.
Page 21 of 29
Little, Ian (1957) A Critique of Welfare Economics. 2nd edition. Oxford: Oxford University
Press.
Little, Ian (1982) Economic Development: Theory, Policy and International Relations. New
York: Basic Books.
Little, Ian and James Mirrlees (1968) Manual for Industrial Project Analysis. Paris: OECD.
Little, Ian and James Mirrlees (1974) Project Appraisal and Planning for Developing
Countries. New York: Basic Books.
Little, Ian, Tibor Scitovsky, and Maurice Scott (1970) Industry and Trade in Some Devel
oping Countries. Oxford: Oxford University Press.
Malerba, Franco and Richard Nelson (2011) ‘Learning and Catching-up in Different Sec
toral Systems: Evidence from Six Industries’, Industrial and Corporate Change 20(6):
1645–75.
Meade, James (1955) Trade and Welfare. Oxford: Oxford University Press. (p. 149)
Nelson, Richard (1993) National Innovation Systems: A Comparative Analysis. New York:
O’Mahony, Mary and Marcel Timmer (2009) ‘Output, Input and Productivity Measures at
the Industry Level: The EU India KLEMS Database’, Economic Journal 119(538): 344–403.
Pigou, Arthur (1920) The Economics of Welfare. London: Macmillan.
Robinson, Joan and John Eatwell (1973) An Introduction to Modern Economics. New York:
McGraw Hill.
on Economic Activity 2: 365–412.
Romer, Paul (1990) ‘Endogenous Technological Change’, Journal of Political Economy

98(5): S71–102.
Schydlowsky, Daniel (1984) ‘A Policy-maker’s Guide to Comparative Advantage’, World

Development 12(4): 439–49.
Teitel, Simon (1984) ‘Technology Creation in Semi-industrialised Economies’, Journal of

Development Economics 16(1–2): 39–61.
Tinbergen, Jan (1967) Economic Policy: Principles and Design. Amsterdam: North Hol
land.
UNIDO (1972) Guidelines for Project Evaluation. New York: United Nations Industrial De
velopment Organization.
Page 22 of 29
Weiss, John (2010) ‘Changing Trade Structure and its Implications for Growth’, World
Economy November: 1269–79.
Weiss John (2013) ‘Strategic Industrial Policy and Business Environment Reform: Are
They Compatible?’. Donor Committee on Enterprise Development Working Paper, June.
Weiss, John (2018a) ‘Lewis on Industrialisation and Industrial Policy’, Journal of Interna
tional Development 30(1): 61–79.
Weiss, John (2018b) ‘Implementing Industrial Policy: How to Choose’, Oxford Develop
ment Studies 46(1): 71–82.
Weiss, John and Hossein Jalilian (2019) ‘Competitiveness in African Manufacturing: Some
Evidence from Tanzania’, in David Potts (ed.) Tanzanian Development: A Comparative Per
spective. London: James Currey, pp. 182–98.
Williamson, John (1993) ‘Democracy and the Washington Consensus’, World Development
21(8): 1329–36.
Williamson, John (2004) ‘The Washington Consensus as Policy Prescription for Develop
ment’. Lecture at World Bank, January.
World Bank (2005) ‘A Better Investment Climate for Everyone’, World Development Re
port 2005.
Notes:
(1) Little himself, in his earlier authoritative work on welfare economics, when writing of
the marginal cost-pricing rule, brings out the policy ambiguity inherent in welfare theory:
‘We may sum up the present discussion by asking (as we have asked for the other “opti
mum” conditions) what is required for it to be sufficient for an improvement in welfare al
ways to adjust output wherever and whenever possible until price is equal to marginal
cost. Formally the answer is that it can be proved to be sufficient only if all the ‘optimum’
conditions are put into operation at once, and if the resultant redistribution of income is
considered to be not unfavourable, and if all external effects are absent. But we have
seen that it is manifestly impossible to put all the conditions into operation’ (Little, 1957:
161; emphasis added).
(2) An interesting link between current suggestions and thinking in the 1950s is found by
comparing some of the arguments reviewed in this chapter and the work of Arthur Lewis
in advising on the prospects for industrialization in the Gold Coast (now Ghana); see
Weiss (2018a).
(3) Robinson and Eatwell (1973: 310).
(4) Implicitly (and sometimes explicitly) this involves a comparison of the likely benefits
and costs of a policy intervention to address market failures (Tinbergen, 1967).
Page 23 of 29
(5) The original version of Corden (1985) was delivered at a seminar in 1978. Using the
hierarchy approach, he suggested both created distortions, which for selective import tar
iffs was a consumer distortion and for an undervalued real exchange rate was the shift in
producer prices for all tradables regardless of whether all needed support. In addition,
with exchange rate undervaluation there would be the need to sterilize any resulting cur
rent account surplus. On balance he appeared to favour conventional tariff protection.
(6) For example, one of the key neoclassical authors on trade policy, Bela Balassa, sug
gested a basic rate of 10 per cent protection for manufacturing in general with the possi
bility of a rate of 20 per cent for infant industries (Balassa, 1982: 69). Here rates refer to
effective, not nominal, protection. It was usually understood that gains were based on
‘learning by doing’ rather than on conscious efforts at accumulation of production capa
bilities, as identified in empirical work and stressed in structuralist literature (Bell et al.,
1984; Teitel, 1984).
(7) This explains the recommendation in Little et al. (1970) for support for new activities
through ‘promotion’ by production subsidy rather than ‘protection’ by import tariff.
(8) Little et al. (1970) is a classic analysis of the costs of protection from a neoclassical
perspective.
(9) Balassa (1982: 67) argues that ignoring budgetary considerations is unrealistic and for
that reason questions the feasibility of subsidy schemes.
(10) For example, in its discussion of possible interventions World Bank (2005: 162‒3)
suggests to ‘match the instrument to the rationale’, which is a restatement of the policy
hierarchy approach.
(11) Hallward-Driemeier and Pritchett (2011) describe ‘favoured’ and ‘disfavoured’ firms,
the former barely troubled by controls and regulations. In the African context Hallward-
Driemeier et al. (2010) show that variations in indicators like days to clear customs or
time spent dealing with officials vary more within countries than between them.
(12) In high-income economies it is now common for industrial policy to be rationalized by

reference to externalities from pioneer activities and the need to support these by ‘social
izing risk’. Thus, for example, to quote the most recent industrial strategy from the gov
ernment of the United Kingdom: ‘Governments in successful economies have recognized
their strategic power and leadership role, allowing them to coordinate and convene ef
forts to develop and assimilate new technologies and industries…Governments can make
long-term investments that no single commercial or academic player can take alone. The
modern nation state is the most powerful means we have of pooling risk’ (HMG, 2017:
22).
(13) ‘In attempting to promote innovation governments have used a variety of instruments
such as trade protection, public sector credit, tax holidays and investment and export
Page 24 of 29
subsidies…However, interventions typically create other distortions’ (Hausmann and Ro

drik, 2003: 630).
(14) In their policy advice to the government of South Africa, specific suggestions from the
authors included the creation of a special fund to finance public inputs suggested on the
basis of proposals from the private sector, with these public inputs defined broadly to cov
er infrastructure, training, administrative reform, R&D or regulatory reform, and the con
version of the existing Industrial Development Corporation into a risk-taking venture cap
ital fund (Hausmann et al., 2008).
(15) The distinguished authors of Guzman et al. (2018) who advocate this approach would
probably not wish to be described as neoclassical, but in so far as their policy recommen
dation on the real exchange rate is justified by learning externalities it falls under the def
inition of neoclassical industrial policy used here.
(16) Andreoni and Chang (2019: 139), reviewing the work of Hausmann and Rodrik, sug
gest that the focus on informational externalities is an example of a ‘clumsy translation of
old ideas by non-neoclassical schools into the neoclassical language’. The point that the
informational externality is an old argument is, of course, correct, but the concept of a
‘knowledge externality’ drawn on by Hausman and Rodrik was clearly cited in the earlier
literature (Corden, 1974: 262‒4; Johnson, 1965). The fact that many neoclassical authors
tended to underplay its significance as the basis for special support for new activities is a
different argument and illustrates the point that the potential for an active policy inher
ent in the market failure analysis was often not developed.
(17) ‘Historical evidence and economic theory suggest that while markets are indispens
able mechanisms to allocate resources to the most productive sectors and industries, gov
ernment intervention—through the provision of information, co-ordination of hard and
soft infrastructure improvement and compensation for externalities—are equally indis
pensable for helping economies move from one stage of development to another’ (Lin,
2010).
(18) Lin and Monga (2010: 304) argue that ‘by facilitating co-ordination and addressing
externality issues, industrial policy helps many domestic and foreign firms to enter sec
tors that are consistent with the country’s latent comparative advantage and turn them
into overt comparative advantages, and thereby intensifies competition within the indus
tries and enhances the economy’s competitiveness’.
(19) The debate between Lin and Chang (Development Policy Review, 2009) provides a
helpful insight into alternative visions of technological upgrading and structural transfor
mation in industrializing economies. While both agree on the need for government sup
port, they disagree on how far and how fast it is wise for countries to move into ‘distant’
products in the sense of goods that are far removed from current comparative advantage.
(20) For example, for apparel, specific recommendations include streamlining customs
procedures, eliminating all import tariffs on imported inputs, rehabilitating rail links to
Page 25 of 29
the port, and developing an industrial park near the port; for leather goods, recommenda
tions include removing the import ban on processed leather and providing technical assis
tance in factory operations and product design.
(21) The arguments in this section draw on Weiss (2018b).
(22) Andreoni (Chapter 14 in this volume) suggests a blurring of the division between con
ventionally defined sectors as a result of recent technical change, which makes sector-
based policies less relevant.
(23) The KLEMS database originally constructed for OECD countries, but now extended to
India and elsewhere, provides TFP estimates at the sub-sector level (O’Mahony and Tim
mer, 2009).
(24) Felipe and McCombie (2019) argue that TFP estimates in neoclassical growth models
are based on an initial accounting identity and that the unexplained residual is simply a
weighted average of the growth of wages and the rate of profit, with weights given by
labour and capital shares.
(25) Chang (2013) argues that Japan targeted its auto sector in the early 1960s when its
income was one-sixth of the market leader economy, the United States, and Korea en
tered semiconductors in the mid-1980s when its income per capita was one-seventh of
that of the United States. Lin argues that in the early 1960s Japan’s income was 40 per
cent of the United States’, so that the auto example fits his rule of thumb. Development
Policy Review (2009) sets out the contrasting views.
(26) The ERP for output i was usually calculated from the formula (ti – ∑aji.tj)/(1 – ∑aji),
where ti is the tariff on i (or tariff equivalent where direct controls were used), tj is the
tariff or tariff equivalent on input j, summation covers all inputs j to n, and aji is the share
of input j in the value of output i. Theoretical problems arise with this formula in relation
to the treatment of non-traded inputs into i, whose price is also raised by protection and
by the fact that strictly the coefficient aji should not be the observed coefficient at domes
tic prices, but the counterfactual coefficient at world prices under free trade (Corden,
1966).
(27) DRCj = VADPj/(WPj – ∑aijWPi), where WPj and WPi are the world prices for the trad
ed outputs j and inputs i involved, so for all traded inputs required (WPj – ∑aiWPi) is the
net foreign exchange effect per unit of j’s production after deducting the cost of traded
inputs used in j. If the value-added content of j covering labour and non-traded goods is
valued at their opportunity cost, VADPj gives the economic value of the domestic re
sources required to generate a given foreign exchange effect. DRC can either be ex
pressed as an exchange rate (where VADPj is in local currency and (WPj – ∑aiWPi) is in
foreign currency) or as a number (where WPj – ∑aiWPi is in local currency converted at
an exchange rate taken to reflect the long-run or equilibrium value of foreign currency).
Where the former approach of an exchange rate is used, an activity is efficient if the DRC
ratio is below the economic value of foreign currency (the shadow exchange rate). If the
Page 26 of 29
DRC is expressed as a number, the net foreign exchange in the denominator (WPj – ∑aiW
Pi) should be converted to local currency at the shadow exchange rate and efficiency re
quires a ratio of below unity.
(28) An early example of the use of the DRC ratio is the study on the high cost of automo
bile engine assembly in India (Baranson, 1967). Weiss and Jalilian (2019) give DRC esti
mates used in a more recent discussion of planning priorities in Tanzania.
(29) Little and Mirrlees (1974: ch. 18) briefly showed the equivalence between their ap
proach and the DRC analysis.
(30) The weights were determined either by revealed comparative advantage (Hausmann
et al., 2007) or share in world exports (Lall et al., 2006). Weiss (2010) contrasts the two
approaches.
(31) An intuitive explanation, along with product complexity scores for over 5,000 prod
ucts, is in Abdon et al. (2010).
(32) Thus, for two goods, for example pig meat and copper, it is the minimum probability
of all meat exporters exporting copper with revealed comparative advantage and all cop
per exporters exporting pig meat with revealed comparative advantage.
(33) This is used in Hausmann and Chauvin (2015). Hausmann and Klinger (2008) use dif
ferent standard deviations from the mean distance to show the sensitivity of the results to
the choice of distance.
(34) This last condition is met by a measure termed ‘opportunity gain’ by Hausmann and
Chauvin (2015), although different terminology was used in earlier studies. Ideally a
country should specialize in goods with the highest complexity, the shortest distance, and
the largest opportunity gain. However, given the trade-offs between distance (as a mea
sure of how feasible it is to move to new products) and the other two indicators, for con
sideration in the list of export targets, product j should have the following characteristics:
where PCI is the product complexity index, j is an individual product, PCIav is the aver
age PCI for the export basket, Dj is the distance between j and the current export basket,
Dav is the median distance between goods not currently exported with comparative ad
vantage and the current export basket and OG is opportunity gain (a measure of what j
adds to the links with other potential exports).
(35) See, for example, for Colombia (Hausmann and Klinger, 2008), Rwanda (Hausmann
and Chauvin, 2015), and Tajikistan (ADB, 2016). The Asian Development Bank has used
Page 27 of 29
the product space approach in its country diagnoses for industrial development in several
countries.
(36) The classification of these priority areas into three categories groups them as: (1)
processed agricultural products, beverages, tobacco, and agro-chemicals; (2) specialized
textiles and garments; and (3) construction materials, metal, and wood products. The au
thors comment that their priority list does not differ markedly from the priorities in the
National Export Strategy for non-traditional exports, which were derived from expert
judgements.
(37) The website www.atlas.cid.harvard.edu provides information by country with an ex

planation of the methodology.
(38) As Little and Mirrlees (1974: 349) wrote in the context of external effects: ‘Even a
back of the envelope calculation may serve to show either that the initial suspicion was
unjustified, or that further work might need to be done. If it is thought that the presence
of external effects will be strongly claimed by opponents or proponents of a project, every
effort to achieve a sensible, albeit rough quantification, should be made. Otherwise wild
exaggeration is all too easy.’
(39) Andreoni and Chang (2019: 140) criticize the methodology for focusing on the charac
teristics of products rather than on links through use of similar technology and produc
tion processes, which may be more relevant in determining ease of moving into new ar
eas. This point is valid but the product space analysis is meant as an operational guide
and data unavailability would preclude a similarly detailed analysis based on technology.
They also appear to imply that the product space analysis is not intended to be used as
part of a supportive industrial policy, but of course it could be.
(40) Ansu (2013) surveys approaches to ‘market-supporting industrial policy’ in five

African economies.
(41) Crespi et al. (2014: ch. 10) highlight the difficulties involved including lack of institu
tional capacity, the risk of capture and rent seeking, and the intrinsic difficulty of identify
ing the best steps to self-discovery. They also conclude, however, that some public-sector
institutions in the region have had success in working closely with the private sector.
John Weiss
John Weiss is emeritus professor of development economics at the University of Brad

ford, United Kingdom. He has written extensively on issues of cost‒benefit analysis,
industrialization, and industrial policy, publishing several textbooks, including
Project Analysis in Developing Countries (with Steve Curry, MacMillan, 2000), Indus
try in Developing Countries (Routledge, 1990) and The Economics of Industrial De
velopment (Routledge, 2011). He is the joint editor of the Routledge Handbook on In
dustry and Development (Routledge, 2016). He has been a consultant to, among oth
Page 28 of 29
ers, the United Nations Industrial Development Organization (UNIDO), the World
Bank, and the Asian Development Bank.
Page 29 of 29
Enterprises and Industrial Policy: Firm-based Perspectives
Enterprises and Industrial Policy: Firm-based Perspec

tives
Simon Roberts
This chapter considers the central role of firm strategies in industrial policy. It focuses, in
particular, on large firms which can make investments to realize economies of scale and
scope, coordinate clusters of suppliers, and build capabilities. The main schools of
thought addressing the theory of the firm and the development of productive capabilities
are reviewed, especially with reference to the resource-based theory of the firm. Large
firms generally have economic power and considerable political influence and, as such, a
critical aspect is firms as agents in political settlements and the implications for industri
al development. Substantive sections then cover the following key areas: market power,
competition, and enterprise strategies; the internationalization of production and global
value chains; and recent developments regarding digital platforms and the implications
for industrial policy.
Keywords: industrial policy, firm-based perspectives, market power, competition and enterprise strategies, digital
platforms
6.1 Introduction
THIS chapter considers the central role of firm strategies in industrial policy. It focuses,
in particular, on large firms which can make investments to realize economies of scale
and scope, coordinate clusters of suppliers, and build capabilities. The productive re
sources, organizational capabilities and ability of big businesses to shape the policy envi
ronment in their interests are central to a production-centric understanding of industrial
policy.
The enterprise is at the centre of industrialization and the strategies of large enterprises,
in particular, drive the development of productive capabilities. The theory of the firm is
thus an essential part of the conceptual framework for understanding industrial policy
and economic transformation. As emphasized by Ohno (2013), a dynamic private sector is
critical for catch-up and supporting this is perhaps the key industrial policy challenge fac
Page 1 of 31
ing developing countries. Four different dimensions to this may be distinguished (Lall,
1994): technological upgrading within industries; entry into more complex activities; in
creasing local content involving local innovations and design; and mastering more com
plex technological tasks within industries.
Neoclassical theories of the firm struggle to address these core concerns as their market-
centric views simply posit the firm as a set of arrangements to minimize transaction
costs. Firms are located in idealized perfectly competitive markets with many small ri
vals, and no economies of scale or substantial market power. Transaction-costs theories
influence approaches to understanding the role of businesses in industrial development
which abstract the firms from the political economy (p. 151) context and the reality of
market power, and support simplistic ‘investor-friendly’ policy frameworks lowering the
costs of doing business.
It is therefore essential to critically evaluate alternative theories of the firm which explain
the development of productive capabilities, and the ways in which businesses exert con
trol and hold power over economic activity. The market power which firms hold in setting
high prices if they have a monopoly position or collude with their competitors applies to
arms-length market exchanges. However, value creation involves a range of relationships
through which activities are organized in production networks and value chains. The co
ordination of the activities involves direct and diffuse power, which can be exerted in di
rect and indirect ways (Dallas et al., 2019). For example, international business organiza
tions can shape sustainability codes through diverse lobbying activities which affect the
competitiveness of industry in different countries and benefit some interests over others
(Ponte, 2019).
Business is increasingly transnational in nature as companies coordinate activities spread

around the world. Industrial policies and regulations remain predominantly national. The
ongoing digitalization of design, production, marketing, and logistics comes on top of the
advances in information and communication technologies that have enabled international
coordination by lead firms. It portends a further stepwise change to a hyper-globalized
world where companies transcend national borders. At the same time, the dramatic
growth of the Chinese economy and corporations stands as a counter-model, where the
state is intimately involved in the strategic orientation of corporations to ensure consis
tency with the development vision (as addressed in Oqubay and Lin, 2019; Oqubay and
Ohno, 2019).
I start by reviewing the main schools of thought in the theory of the firm and the develop
ment of productive capabilities. This takes into account the literature on transaction costs
and the resource-based theory of the firm. Large firms naturally have considerable politi
cal influence; a critical aspect is firms as agents in political settlements and the implica
tions for industrial development. Substantive sections then cover the following key areas:
market power, competition, and enterprise strategies; the internationalization of produc
tion and global value chains; and recent developments regarding the digitalization of pro
duction, digital platforms, and the implications for industrial policy.
Page 2 of 31
6.2 Theory of the Firm and Enterprises in In

dustrial Development
The theories of the firm in economics can essentially be divided into two main groups.
Firms can be understood either in terms of their organizational capabilities and produc
tive resources or as a set of coordinating arrangements which replace market exchange.
The transaction-costs theory of the firm, based on the seminal article by Coase
(p. 152)
(1937), posits that the firm may essentially be viewed as the unified governance structure
that results from the properties of different transactions. It thus takes exchange as its
starting point and defines firms as sets of non-market relationships. The primary factors
producing transactional difficulties include bounded rationality and imperfect informa
tion, opportunism, small-numbers bargaining, and asymmetric information. The costs as
sociated with transactions increase with uncertainty and asset specificity (Williamson,
1986).
Under the transaction-costs approach the firm is the result of fixed, structured, and open-
ended relational contracting which internalizes the transactions, removing or ameliorat
ing the conflict, and exerting the administrative control required for planning and coordi
nation. The internationalization of production can be seen as a particular subset of the
reasons for the formation of firms, in response to the costs of organizing cross-border
markets (Dunning, 1993). The transnational corporation (TNC) is therefore essentially
viewed as an efficient response to intrinsic market failures.
The firm is, however, treated as a ‘black box’ and ‘[w]hat happens between the purchase
of factors of production and the sale of goods that are produced by these factors is large
ly ignored’ (Coase, 1992). The firm is simply a nexus of contracts (Jensen and Meckling,
1976). When transaction costs are defined to subsume a broad range of issues with mar
ket relationships, it becomes unclear what exactly is being internalized. For example, we
need to understand costs associated with transactions in products due to asset specificity
or embodied technology (leading to intra-firm trade), the development of products
through different stages of the product cycle, as well as factors such as managerial capa
bilities and organizational techniques.
In fact, what are identified as market failures are intrinsic features of the organization of
economic activity—they are not characteristics which, if somehow ‘corrected’, will mean
an arms-length market relationship can exist. At the most basic level, there are
economies of scale and scope, and network effects. Technological improvements at the
firm level also need to be taken into account. These factors have effects on industrial de
velopment which depend on complementary changes, such as in management approach
es, that need to be understood in a systems framework incorporating inter-industry rela
tionships (Rosenberg, 1979).
Page 3 of 31
Institutional-based theories of the firm suffer from being primarily concerned with ex
plaining the types of contractual arrangements which govern exchange rather than un
derstanding the way in which the production is organized and capabilities are developed
over time. This relates to the factors which explain how firms are able to reconfigure
their productive assets and processes (their dynamic capabilities) to produce new and
better products and services (Teece, 2009; Teece and Pisano, 1998; Pisano, 2017). The in
stitutional approach also ignores the power and political influence of large firms (Zin
gales, 2017). The literature on the resource-based theory of the firm, including critiques
and extensions, has sought to address these issues head on.
(p. 153) 6.2.1 Resource-based Theory of the Firm
The resource-based theory of the firm builds on the essential contribution of Edith Pen
rose (1959) who focused on firms’ productive resources, organizational routines, capabili
ties, and competencies. There is a dynamic learning process through the interaction of
technology capabilities and market opportunities that is at the heart of the principles of
production and business organization.
Penrose argued that experience builds the knowledge of productive opportunities on the
part of firms and that ‘productive services’ (that is, capabilities) open up new productive
opportunities. Firms shape markets through decisions which alter the parameters (tech
nological, product, and organizational) of the market. The productive capabilities of the
firms are heterogeneous and evolve depending on how they are used. This means that
productive opportunities are unique for each firm and are shaped in dynamic processes
(Penrose, 1959). Firms are competitive in so far as they are able to establish technologi
cal or market ‘bases’ (a distinctive technological capability). There can be innovation-dri
ven increasing returns to scale in industry, which were identified by Babbage in the nine
teenth century in his book originally published in 1832 (Babbage, 2005), and are still
highly relevant to considering the implications of the digitalization of production in the
twenty-first.
In a similar vein, Chandler (1990) views the ‘modern business firm’ as a collection of ‘dy
namic organizational capabilities’ which are the source of the firm’s competitiveness (see
also Dosi, 1997). In this framework, the firm is an organization whose development de
pends primarily on its ability to continuously innovate and develop new products and
ways of working. These capabilities are based on a firm’s internal organization and the
extent to which the collective experiences from production are developed, shared, and
learnt from. Cooperation and collaboration are therefore the basis for competition which
is viewed not as aiming to be the lowest-cost producer, but as the ability to advance and
produce better products, more quickly, than other groups of firms. These factors together
describe differences in firms’ competitiveness, as well as the evolution of ‘business-enter
prise systems’ (Fujimoto, 1998).
The resource-based theory of the firm, however, insufficiently recognizes the extent to
which firms may represent a locus of control and use this control to condition the envi
Page 4 of 31
ronments within which they operate to extract rents. Large profits may be the reward for
investment and innovation to generate or maintain competitiveness; however, they may
also be due to the erection of barriers to entrants and strategies to entrench a position of
market power over potentially more innovative rivals. These considerations point to the
importance of understanding firms as part of the wider political economy, and to why
large business groupings in some countries have built productive capabilities while in
others they have simply extracted rents (as elucidated in different country/region chap
ters in this book). ‘Bringing production back-in’ (Amsden, 1997) requires locating a re
source-based theory of the firm in relation to the wider society.
The resource-based theory of the firm has been extended through a growing liter
(p. 154)
ature on firms understood in terms of their dynamic organizational capabilities (Chandler,

1990; Chandler et al., 1997; Dosi et al., 1998; Teece et al., 1997; Pisano, 2017). This in
cludes learning, innovation, skills, and organizational resources, and the construction of
competitive assets (Amsden, 2001; Lazonick, 2010; Khan, 2009, 2019). Firms have been
located within the evolution of business-enterprise systems and the institutional context
(Dosi, 1997; Chandler et al., 1997; Best, 2018).
Chandler drew on detailed analysis in business history to explain the rise in large indus
trial corporations in the United States, contrasted with the United Kingdom and Germany,
setting out the dynamics of industrial capitalism (Chandler, 1990). This emphasized the
central role of large industrial firms in the creation of wealth, and the organizational ca
pabilities of the enterprise as a whole. The firms could make investments large enough to
reap economies of scale. Firms are the bases for technological development and manage
rial skills. Large firms establish and coordinate a nexus of related small and medium-
sized firms to ensure the flows of materials and information necessary to maintain pro
duction and distribution.
The organizational capabilities were the collective physical facilities and human skills as
they were organized within the enterprise and their careful coordination and integration
to achieve the economies of scale and scope needed to compete locally and international
ly. These organizational capabilities have to be created and maintained (Chandler, 1990;
Lazonick, 2010; Teece, 2010).
Changes in technology, learning and skills development, and capital are constitutive of
the evolution of the firm itself. These are all associated with a reconceptualization of the
business entity around its organizational capabilities which determines a firm’s scale (de
fined as ‘throughput’) and scope, being the production of related goods. This focus on the
organizational aspects may be distinguished from the approach in much of the new insti
tutional economics which views institutions as a collection of relationships that can be
characterized by explicit or implicit contracts replacing market exchange (Amsden, 1997;
Khan, 2019).
The orientation of large firms is central to the nature of capitalism in different countries.
Hikino and Chandler (1997) contrast the US model, termed ‘competitive capitalism’ to re
flect the highly developed managerial nature of firms and the vigorous competition be
Page 5 of 31
tween them, with the experience of Germany which is termed ‘cooperative capitalism’. In
Germany it is argued that firms developed a management hierarchy capable of exploiting
the opportunities of economies of scale from investing in heavy industries such as iron
and steel, but cooperated closely together.
The resource-based theory of the firm focuses on the ability to strategically deploy its re
sources, on which there is now a vast business-school literature following Porter (1990).
However, the business-school adoption of the resource-based theory diverted attention
from the importance of taking into account innovation and learning (Lazonick, 2010;
Pisano, 2017) and the wider political economy instead of a narrow and functional focus on
organizational structure (Dosi, 1997). Comparative country studies (such as in Chandler
et al., 1997, 1998) highlight the importance of understanding (p. 155) different groupings
within big business, particularly finance, as well as the complex relationships of business
with the state, as addressed in other chapters in this Handbook. It is important to empha
size the diversity of outcomes, the inherent uncertainty, and that effects such as cumula
tive causation and path dependency need to be understood at the levels of production sys
tems, institutional frameworks, and wider society, beyond individual firms.
Teece and Pisano (1998) draw on the definition of the firm in terms of capabilities and
competencies to identify a firm’s ‘strategic dimensions’. These dimensions are identified
as: a firm’s managerial and organizational processes (including patterns of practice and
learning); its present position (including its technological and intellectual property, cus
tomer base, and relations with suppliers); and the paths available to it (the strategic alter
natives and opportunities available to the firm). Strategy and organization are crucial to
the choices firms make in identifying and selecting capabilities for future competitive ad
vantage, including general-purpose capabilities enabling functional integration across
specialist domains, and market-specific ones (Pisano, 2017).
The uncertainty inherent in innovation means that finance plays a particularly important
role in how some enterprises mobilize and deploy productive resources for the develop
ment of new products and services, as identified by Lazonick (2010) in the theory of the
innovative enterprise. This uncertainty includes: technological uncertainty relating to
whether the firm will be able to develop the products and processes it envisages; market
uncertainty in terms of the returns which will be made from the improved goods and ser
vices due to customer attraction; and competitive uncertainty, as rivals are engaged in
similar strategies and may be more successful. The financial commitment is the set of re
lations that ensures allocation of funds to sustain cumulative innovation processes, in
what has been termed ‘patient capital’. This is mainly about the availability of long-term
development finance and the ability of the firms to retain and deploy earnings.
The theory of an innovative enterprise must itself be embedded in a model of relations

among industrial sectors, business enterprises, and economic institutions. In addition to
the conditions for financial commitment (as opposed to financialization, discussed in oth
er chapters in this volume), Lazonick identifies two other social conditions which are es
sential for the innovative enterprise to thrive. First, strategic control is the set of rela
Page 6 of 31
tions that gives decision makers the power to allocate the firm’s resources to confront the
technological, market, and competitive uncertainties inherent in the innovation process.
Second, organizational integration is the extent to which there are the incentives for peo
ple to apply skills and efforts to the learning required at the heart of the innovation
process (see also Khan, 2019). There is an important interdependence of the organiza
tional capabilities with formal knowledge and skills (Khan, 2019).
While this conception of the firm grapples with issues of coordination and control, the
performance of firms is based on the extent to which they are effective in terms of
arrangements, given the competitive challenges of the time and industry in question.
Drawing on resource-based theories, Best (1990) identifies the ‘new competition’ as
(p. 156) being where performance rests on the ‘resources and experience of the parent
concern, including not only managerial and technical personnel but also the indefinable
advantage in its internal operations which an efficient going concern usually has over a
new one’. Firms also choose to collaborate (as well as to make or buy), and networked
groupings will outcompete competing groups of firms (Richardson, 1972). This is because
competition between firms does not support the emergence of specialized firms with net
worked complementarities as part of clusters.
The extensions of the resource-based theory of the firm require enterprises to be located
in production ecosystems if we are to take account of the social context and linkages
which underpin dynamic capabilities (Andreoni, 2018). A successful industrial policy re
quires the articulation of public policy with the detailed mechanics of change that occur
within businesses in a production-centric (rather than market-centric) framework (Best,
2018). Such a framework can be captured in a ‘capability triad’: business models; produc
tion capabilities, innovation, and skills; and clusters, networks, ecosystems, and learning.
This triad underpins the creation of dynamic increasing returns and the innovation dy
namics that underlie productivity growth.
Industrial policies to alter a country’s development path need to realize external

economies from stronger linkages, collective learning, and governance (Helmsing, 2001;
Kaplinsky, 2000). Processes of learning and technical change involve the firm and the
wider institutions of industrial policy (Lall and Teubal, 1998; Khan, 2019).
A proactive industrial policy requires private dynamism and improving policy capability to
ensure dynamic capacity development (Ohno, 2013). State policies reward value creation
and innovation, while disciplining unproductive rent seeking. The focus in firms and in
dustry clusters on knowledge, skills, and technologies needs to be mirrored by policy
practitioners in government keeping up with industry developments and building effec
tive engagement with industry.
I identify three key areas for further in-depth assessment. First, the interaction of enter
prises’ market power with the requirements of innovation and dynamic increasing re
turns is a key issue for understanding enterprises and industrial policy. Second, the evolv
ing internationalization of production has fundamental implications for countries’ indus
trial development and policies. Third, digitalization is rapidly changing the organization
Page 7 of 31
and coordination of economic activity, including across borders. These are now addressed
in turn. Other key issues, such as patterns of financialization, are equally important and
are addressed by other contributions to this volume.
6.3 Market Power, Competition, and Enterprise

Strategies
The significance of dynamic increasing returns to scale means that there will normally be
only a few firms in key industries and that concentration is an inevitable feature of
(p. 157) industrial development. For example, large and internationalized firms have had a
crucial role in developing key industries such as consumer electronics, automobiles, ship
building, steel, and basic chemicals around the world. Indeed, cross-industry regressions
find scale economies is a strong explanatory factor for concentration (Sutton, 2007,
2012). And smaller domestic markets relative to the minimum efficient scale of produc
tion in an industry means, other things being equal, that concentration will be higher in
many developing countries.
The importance of the large firms means they have considerable power over their suppli
ers and customers, as well as being important political actors. This has been emphasized
in seminal work on the lead East Asian industrializers, where state support was contin
gent on performance expectations which could be enforced through disciplining mecha
nisms to ensure that the outcomes were in line with performance expectations (Amsden,
1989; Chang, 1996; Wade, 1990; Sakakibara and Porter, 2001). Competition in export
markets was a key lever for discipline at the firm level and a way of measuring firms
against each other even while offering protection in the domestic market (Rodrik, 1995;
Singh, 2004). It also ensured that firms were forced to adopt and adapt more advanced
technologies, production, and marketing methods, as they were being pitched against the
international industry leaders in export markets.
More generally, competitive rivalry between big business groups is a very important disci
plining and motivating factor to ensure that state support does not lead to collusion and
rent extraction. The nature of the rivalry is also an important element of Chandler’s char
acterizations of different capitalisms, although quite different from the neoclassical mi
croeconomic elevation of competitive markets on the grounds of static allocative efficien
cy. Instead, it is part of the interaction of public policies and enterprise strategies to in
centivize and support innovative capabilities.
There is no ideal of maximum or perfect competition; rather, we can consider what is

meant by an ‘optimal competition’ in terms of performance (Amsden and Singh, 1994;
Singh, 2002). Management of the economic power of key companies requires reciprocal
conditionalities (Amsden, 1989, 2001). Rents are essential for inducing the effort and in
vestment required on the part of firms to develop innovative capabilities. Once again, a
neoclassical framing of markets which views market power as an aberration to be cor
rected is unable to take into account the real world of pervasive market power to differ
Page 8 of 31
ing degrees and, in particular, of those large firms required to undertake organization
and technological learning.
Competition and competitiveness are therefore a result of a successful industrial policy

which involves the creation of productive rents, and discipline regarding their access.
This can be equated with ‘performance competition’ where effort and innovation are re
warded rather than ‘handicap competition’ where firms seek to undermine their rivals
(Gerber, 2010).
There has been an increasing focus since the 2014 publication of Piketty’s Capital in the
Twenty-first Century on the implications of concentration of ownership and control for
economic development and for inequality. It is argued that the returns from the exertion
of market power go disproportionately to the wealthy and, on balance, market (p. 158)
power discourages innovation and reduces productivity, as observed in the United States
(Baker and Salop, 2015; Ennis et al., 2019; De Loecker and Eeckhout, 2017). Stronger
competition law is sometimes recommended as the answer (see Stiglitz, 2017; Atkinson,
2015).
There is little doubt that there has been a growing global concentration in the hands of
transnational corporations. The largest TNCs have bigger revenues than many govern
ments around the world. In 2015, sixty-nine of the largest 100 corporate and government
entities ranked by revenue were calculated to be corporations (Zingales, 2017).1 This is
not necessarily unprecedented. Under colonialism companies such as the British East In
dia Company controlled the economies of countries with monopoly rights over trade in
some commodities that lasted more than two centuries. Lobbying to maintain these rights
turned firms away from improving productivity in a ‘Medici vicious circle’ where money is
used to get power and power is used to make more money (Zingales, 2017). Control over
global profits in major industries is in itself one drive for global concentration (Hymer,
1976).
Network effects and the rise of digital platforms are one reason given for the increasing
concentration in recent years. I discuss the implications of the digitalization of industry in
section 6.5.
There are three main questions. First, under what terms will the concentration be accom
panied by improved productivity rather than simply rent extraction? Second, how is the
economic power of companies understood? Third, what is the nature of the political set
tlement with business which underpins learning and the development of dynamic capabil
ities?
6.3.1 Competition and Productivity
While the promise of high monopolistic profits can spur innovation (in a Schumpeterian
world), entrenched dominant firms have weaker incentives to invest, innovate, and im
prove productivity than if they are challenged by rivals (Schumpeter, 1942; Mathis and
Sand-Zantman, 2014; Arrow, 1962; Bloom et al., 2019). Firms with substantial market
Page 9 of 31
power can earn returns from exertion of this power and protect their position by exclud
ing rivals. As retained earnings are important for firms’ ability to make investments,
smaller firms are typically more constrained in terms of the liquidity they can use to in
vest (Audretsch and Elston, 2002; Goergen and Renneboog, 2001). High entry barriers to
rivals can weaken economy-wide investment, and can prevent new business models
emerging and products being introduced (Shapiro, 2012; Cohen and Levin, 1989; Federi
co et al., 2019).
Measuring and analysing productivity is complex precisely because of the different mech
anisms at the firm and industry level that influence it. Management practices are (p. 159)
one critical determinant. A major US study found substantial variations between different
plants in the same firm, with four key factors explaining plant-level productivity: product
competition; learning spillovers; education; and local business environment (Bloom et al.,
2017). Policies required to support innovation by industries and address the weakening
productivity performance in industrialized countries include support for R&D along with
skills in science, technology, engineering, and mathematics, and measures to increase
competitive rivalry (Bloom et al., 2019).
The issue comes back to ensuring the right kind of rivalry (the ‘optimal competition’ of
Amsden and Singh, 1994; Singh, 2004), which can spur large firms to continue to inno
vate and invest, rather than competition for its own sake (Van Reenen, 2018). A country
does not arrive at such markets by magic. Indeed, it seems obvious that market power,
imperfect competition, and market failures which can reinforce positions of market power
are intrinsic features of economic life, and competition policy is a necessary complement
to industrial policy (Roberts, 2013). We therefore need to understand how the process of
evolving competitive rivalry is related to the nature of economic opportunity and out
comes. This has been an explicit but largely ignored aspect of competition regulation in
South Korea, where the Korea Fair Trade Commission (KFTC) has monitored the conduct
of chaebols, for example in subcontracting relationships to protect against exploitation of
smaller firms (Hur, 2004; KFTC, 2011). Typically, in the short run such subcontracting
arrangements would lower prices and hence not harm consumers. In the longer term,
however, unfair subcontracting arrangements by large firms militate against the develop
ment of a dynamic base of small and medium firms able to invest in their own indepen
dent production capabilities.
6.3.2 Economic Power of Companies
The power of companies to shape markets and influence policies over time is more impor
tant than the market power to charge high prices in any given period. A simplistic focus
on market power measured in static terms as prices which are marked up over marginal
cost is therefore as naïve and unhelpful as modelling the firm as a black-box internaliza
tion of transaction costs and even risks distracting attention from the more fundamental
issues. The features of an industry, such as information asymmetries, scale economies
and network effects, can provide scope for strategic behaviour on the part of dominant
Page 10 of 31
firms to lock in advantages through a range of different strategies (Rey and Tirole, 2006;
Whinston, 2006; Vickers, 2005).
Persistant dominance can suggest that it is not effort and creativity which are being re
warded but rather that the legacy position of large firms continues to earn them rewards.
As Geroski and Jacquemin (1984: 22) caution, ‘when, however, small asymmetries can be
solidified into dominant positions that persist, the inequities they create become institu
tionalized, creating long-term problems in the performance of the (p. 160) economic sys
tem which cry out for policy attention’. In the context of small domestic markets, this is
consistent with the significance of economies of scale, dynamic effects related to technol
ogy, and the importance of production linkages in processes of industrialization (Gal,
2009; Hur, 2004).
The likelihood of entrenched dominant firms with extensive economic power depends on
the country’s conditions and history which are reflected in different approaches to com
petition and industrial policy. For example, analyses have highlighted the importance of
competitive discipline in the industrial development of East Asian countries such as Japan
and South Korea (Amsden and Singh, 1994; Sakakibara and Porter, 2001). The objectives
of South Korea’s KFTC have included promoting ‘balanced development’ in recognition
that the early stages of rapid industrialization were viewed as ‘unbalanced’, requiring an
active competition policy addressing dominant firms in that country (Fox, 2003; KFTC,
2011).
Firms’ power exerted within markets extends to shaping vertical and horizontal relation
ships with other firms. This is incorporated in the coercive, dyadic power between firms,
or between the state and firms, defined as ‘bargaining power’ by Dallas et al. (2019).
Firms with substantial market power as buyers, for example, large supermarket groups,
can undermine the ability and incentives of suppliers to reinvest and improve productive
capabilities (Inderst and Mazzarotto, 2008; Inderst and Valletti, 2011). Through vertical
arrangements lead firms are able to govern activities along value chains (Strange and
Humphrey, 2019).
Influence can also be more subtle at the horizontal level, as illustrated by findings on
common ownership. Even relatively small common ownership stakes by private equity
groups such as BlackRock and Vanguard in firms across economies have apparently influ
enced firm strategies to undermine investment and rivalry (Azar et al., 2016, 2018;
Newham et al., 2018; Seldeslachts et al., 2017). In industries such as airlines and pharma
ceuticals there have been findings that common shareholdings reduce head-to-head com
petition as it supports the alignment of incentives.
Large firms exert power to shape markets in broader ways through collective groups
such as business associations which lobby in direct ways, such as for a particular policy
platform or regulation that favours some interests over others (Vilakazi and Roberts,
2019). Power is also exercised in more diffuse ways, such as by influencing accepted stan
dards and best practices (Dallas et al., 2019; Ponte and Sturgeon, 2014). For example, the
provisions adopted in competition legislation on abuse of dominance, or the nature of
Page 11 of 31
support for public research in industry, reflect a conceptual framework about markets
and the role of the state in the economy. This is in turn an outcome of influence and ideas
which are promoted within and through political parties, the media, and academia.
This can be understood as part of the political settlement where elites exercise power
through ‘informal institutions’, setting the agenda (Khan, 2010). The nature of the politi
cal settlement determines whether longer-term capabilities formation is rewarded, or
coalitions of interests are made to enforce short-term rent extraction.
(p. 161) 6.3.3 Economic Power, Rents, and Political Settlement
The orientation of large enterprises is thus intimately related to the balance of power be
tween different interests in a country and the regulatory arrangements that are part of
the wider political settlement (Khan, 2015). A settlement is defined as a combination of
mutually compatible power relations and institutions that is sustainable in terms of eco
nomic and political viability (Khan, 2010). A given settlement depends on the coalitions of
interests within countries and, for industrial development, whether this coalition supports
the design and implementation of policies with incentives and conditions on firms to en
sure high levels of effort in learning and technological upgrading, while disciplining the
earning of rents from protection and subsidies (Khan, 2015).
The political settlements framework emerged as a criticism of institutional economics and

the position that competition with liberalized markets and independent institutions is the
primary requirement for economic development. The grounds for this are that competi
tion will erode rents which, along with liberalized markets, will move us to what North et
al. (2009) term ‘open access orders’. However, this supposes competition simply arises in
the absence of obstacles and it fails to recognize the need to address entrenched inequali
ty and economic power (Makhaya and Roberts, 2013, 2014). And it does not properly ex
plain what the underlying power arrangements and configuration of interests are that
support a given institutional configuration (Khan, 2010).
In addition, the institutional economics of North et al. (2009) fails to recognize the impor
tant role that industrial policies and tariffs have played in countries’ industrialization—in
other words, productive rents are required to induce investments in capabilities. We need
to understand what coalitions of interests are required to sustain a policy framework for
industrial development.
Powerful elites organize through ‘informal institutions’, especially during development

transitions, to sustain economic benefits for groups who would otherwise have lost out
(Khan, 2010). They are thus the mechanisms through which social and political stability is
maintained, helping to generate distributions of economic benefits that are more in line
with existing distributions of power. In doing so they also sustain these distributions of
power. The framework identifies four sources of ‘holding power’, which refers to the ca
pacity to engage and survive conflicts—in other words, the ability to inflict costs and ab
sorb costs inflicted by opposing groups. In addition to violence rights to exert control
Page 12 of 31
over populations, the sources of holding power include economic structure, ideology, and
rents.
The economic structure reflects a country’s economic history. This includes the construc
tion of markets and the main participants, including the large incumbent businesses. The
economic power of these companies, and how this influences the evolution of the political
settlement, is central to whether the industrial policy will support the growth of diversi
fied manufacturing sectors with higher levels of productivity (Khan and Blankenburg,
2009; Khan, 2017; Andreoni and Chang, 2019; Andreoni, 2016). (p. 162) The political set
tlement framework therefore allows us to assess how incumbent firms or groups of firms
are able to lobby and enter into arrangements with political actors to shape regulation
and the economic environment to ensure future rents. The evolving coalition of powerful
interests effectively sets the rules of the game. Whether a country adopts a competition
law appropriate for the country’s historical context, for example, with regard to the provi
sions on abuse of a dominant position, is part of this picture (Budzinski and Beigi, 2015;
Roberts, 2012).
6.4 Dimensions of the Internationalization of

Production
The increased transnationalization of economic activity over the last fifty years has been
driven by large international corporations. While big businesses were central to colonial
ism and earlier waves of globalization (Bairoch and Kozul-Wright, 1996), the internation
alization of production which now exists is much more multi-dimensional, encompassing
governance of trade, technology, value creation, brands, data, and the related rents. It is
associated with the ability to coordinate productive resources through a multiplicity of
governance arrangements beyond vertical integration of corporations. These blur the
boundary of the firm, although not as a ‘nexus of contracts’ but as a ‘locus of control’.
The exponential expansion in global trade (notwithstanding recent stalling) has far out
paced GDP growth. Information and communication technologies have enabled the global
dispersion of economic activity while being centrally governed by ‘lead firms’, as de
scribed in the literature on global value chains (GVCs) and global production networks
(Bair, 2009; Gereffi et al., 2005; Ponte and Sturgeon, 2014; Coe and Yeung, 2015). This
has included rapid growth in trade in services, including under digitalization of processes
within firms and along value chains (Baldwin, 2016).
Large internationalized firms are pursuing low costs combined with flexibility and speed
(Coe and Yeung, 2015). The digitalization of economic activity has further been associat
ed with a shift in the proportions of value added—with a higher share going to pre- and
post-production activities than to production activities (Baldwin, 2016; Rehnberg and
Ponte, 2017). There have been major advances in technology in the fields of bio-tech, de
sign, advanced manufacturing, and artificial intelligence, as well as the rise of digital
Page 13 of 31
platforms in e-commerce, online search, and social networking which merits a separate
discussion in section 6.5.2.
The term ‘transnational’ is more appropriate for large internationalized corporations

rather than ‘multinational’, as the corporate organization of economy activity transcends
nation states. Profits are attributed to subsidiaries and associated companies according
to the favourable tax treatment by different jurisdictions, while research activities, own
ership of intellectual property, manufacture, marketing, branding, and (p. 163) head office
functions may all be spread around the globe. Meanwhile regulation of TNCs and indus
trial policies remains predominantly national in scope, or at best regional in the case of
the EU.
There are notable exceptions to the transnational (as opposed to multinational) organiza
tion of production, in particular in the internationalization of Chinese corporations where
the state plays a pivotal role (see Oqubay and Lin, 2019).
The growing importance of intermediate goods in international trade, with prices being
administratively determined as transfer prices within firms or agreed between related
parties, reflects the importance of GVCs. Chapter 18 elsewhere in this book (Chapter 18,
this volume) assess the importance of GVCs, including the implications for industrial poli
cy to upgrade firms’ capabilities, create employment and support more inclusive growth.
The emerging evidence is, however, that the organization of production in GVCs has fur
ther polarized income and wealth distribution in both developed and developing countries
(UNCTAD, 2018). The reasons given for this include the nature of technological change
(Rodrik, 2018; Ford, 2015) and the concentrated market structure and dominance of
trade by large transnational corporations (UNCTAD, 2018).
The implications of TNCs for industrial development and the design and implementation
of industrial policy has been recognized for some time. For example, over twenty-five
years ago Lall (1993) identified five areas for future research with regard to TNCs: entre
preneurship and innovation; technological change; the conditions required to stimulate
TNC activity as part of paths of dynamic comparative advantage; the different character
istics of TNCs from different countries and the evolution of less developed-country TNCs;
and the relationship between FDI in services (such as communications and marketing)
and economic development. These all remain extremely relevant today, with a particular
extension to the implications of the international digitalization of economic activity, ad
dressed in section 6.5.
The international spread of businesses has been described in terms of ownership, loca
tion and internalization (OLI) factors in Dunning’s eclectic framework (Dunning, 1992;
Cantwell, 1994). This asks relevant questions related to the intra-firm location and orga
nization of activities across countries. However, it is rooted in the transaction-costs theo
ry of the firm (which drives internalization decisions, according to the framework), albeit
combined with other factors such as technological advantages and agglomeration
economies. In particular, it lacks a proper political economy understanding of large busi
ness and an assessment of the organization of production. The omission hides an appar
Page 14 of 31
ent leaning of many writers within the paradigm towards seeing the firm as essentially an
efficient response to conditions rather than a means to coordinate production and secure
profits. This fits with a business-oriented policy agenda for the de-politicization of eco
nomic policy and promotion of a technocratic approach, which is not consistent with the
observed reality, nor with businesses’ own behaviour (Vilakazi and Roberts, 2019; Hymer,
1976).
Understanding the internationalization of production requires a framework which ex

plains the evolving organization of production, drawing on resource-based theories
(p. 164) of the firm in terms of firms’ productive resources, organizational capabilities,
and competencies. The international competitiveness of business is determined to a sig

nificant extent by the position and power of the firm both with regard to competitors and
within the production chain, which in turn depends on a range of non-/semi-market rela
tionships, on dimensions of internationalization other than trade and ownership, and on
the position of groupings of business in the domestic economy.
The literature on GVCs analyses inter-firm linkages and vertical relations regarding de
sign, production, and marketing of products as layers of value addition (Gereffi and Fer
nandez-Stark, 2011). These linked activities span different enterprises often located in
both developed and developing countries. The geographic fragmentation of activities
means that developing countries can attract some activities in a value chain. However, for
this to be part of sustained productive capability building depends on local industrial clus
ters of producers of intermediate goods and services (Lee et al., 2018). The opening up of
opportunities for process and product upgrading in developing countries depends on the
overall governance and control of the value chain, and the ways in which industrial poli
cies alter the governance of the chain by key firms (Ponte and Sturgeon, 2014).
Lead firms coordinate activities to ensure the competitiveness of the end product and are
able to exploit often extremely asymmetrical power relations between firms within value
chains to control how, where, and by whom value is created and captured (Gereffi and
Fernandez-Stark, 2011). The influence of the lead firm depends on the nature of the sec
tor and the basis of competitiveness. Value chains such as automobiles have been identi
fied as producer driven due to the importance of technology and design (Gereffi and Fer
nandez-Stark, 2011). In buyer-driven chains, such as clothing, the owners of the brands
hold the power. The relationships range from arms-length market exchange, such as to
drive down costs through outsourcing, through close relational governance such as for
coordinating product-specific design, to fully internalized linkages (Gereffi et al., 2005;
Bair, 2009; Gereffi, 2014).
The digitalization of production networks and value chains enables greater coordination
of activities across borders and shifts power to the firms who collate and analyse the da
ta. The nature of governance over internationalized production is changing, with new
strategic industries and sectors which are the locus of control over data and underpin dy
namic capabilities (Petricevic and Teece, 2019).
Page 15 of 31
6.5 Digitalization of Production, Enterprises,

and Industrial Policy
A rapid digitalization of production is taking place associated with key technologies at dif
ferent stages of maturity, including advanced robotics and factory automation, data from
mobile and ubiquitous Internet connectivity, cloud computing, big-data analytics, (p. 165)
machine learning, and artificial intelligence. This is associated with the development of
new ‘platform’ business models and modes of value creation (UNCTAD, 2018; Sturgeon,
2017). The technologies and business models emerging in the digital economy have al
ready disrupted traditional industries and created entirely new ones.2
6.5.1 Digitalization of Production
Advanced manufacturing and automation can be understood as fully integrated, collabo

rative manufacturing systems that respond in real time to changing demands and condi
tions in the factory and the supply network. Though this typically relates to ‘smart facto
ries’, it is equally applicable in agriculture, commonly referred to as ‘precision farming’,
or even mining or construction. Advanced manufacturing is made possible by a combina
tion of technologies including robotics, sensors, machine learning, and the Internet of
things (IoT). Industrial robots linked to advanced manufacturing are automatically con
trolled and operate on their own. They are multipurpose (reprogrammable) and capable
of doing different kinds of tasks rather than simply repeating the same tasks.
The technological changes have profound implications for the coordination of economic
activity by firms and the division of returns (UNCTAD, 2018). Digital tools can improve
design, prototyping and customization processes, reducing scale economies, and opening
opportunities for businesses in developing countries; however, advanced economies have
tended to retain control of higher value-added activities. While, in principle, digital plat
forms can lower costs for smaller businesses to market their products and services, most
digital platforms are characterized by high levels of concentration. Many of the techno
logical changes are also skills biased and may undermine the ability of developing coun
tries to compete in traditionally labour-intensive industries that have supported their in
dustrialization (Rodrik, 2018).
Transnational corporations in the form of digital platforms are shaping global economic
activity and pose new challenges for industrial policy around the world. In addition, there
are substantial emerging regulatory challenges associated with the ownership and con
trol of the data that provide major platforms with their power and associated commercial
value (UNCTAD, 2018; Polson and Scott, 2018; McAfee and Brynjolfsson, 2017).
The nature of the changes can be incremental as well as disruptive. Incremental changes
include improving supply chain integration across firms through digital tracking of prod
uct flow and quality at every stage, which can be combined with (p. 166) machine learning
to improve processes and respond to changes in the environment. Disruptive changes
Page 16 of 31
mean fundamentally altering the way products and services are created and delivered, for
example, through 3D printing of products on demand.
Digitalization also enables a dynamic cycle of continuously improved efficiency that is in
creasingly being driven by the rapid advance of machine learning (artificial intelligence).
Maintaining competitiveness means adopting the incremental changes to build the capa
bilities of clusters of firms and, as such, engaging with the international technology lead
ers.
The generation of ‘big data’ is a characteristic feature of digitalization. Data can be col
lected through sensors in production, from users of the product or service, and from the
online search and purchasing activities of buyers. It is possible for firms to track perfor
mance across multiple production sites and distribution channels and to provide new ser
vices such as real-time monitoring of product use and/or performance using big-data ana
lytics, and increasingly, artificial intelligence (Sturgeon, 2017). This is often referred to as
the industrial Internet of things. It heightens the potential for centralized control across
dispersed production sites and along logistics chains, enabling a further transnationaliza
tion of production.
6.5.2 Digital Platforms
The more members or users on a platform, the more data is collated and the greater its
value to other users in respect of data aggregation and analysis. There are strong
economies of scale and scope and the positive network externality effects within and
across different user groups of the platforms imply they are likely to tip to the ‘winner-
takes-all’ quasi-monopoly (see Furman et al., 2019; Cremer et al., 2019; Committee for
Study of Digital Platforms, 2019). There are major barriers to smaller competitors at
tempting to enter the market and the dominant platforms can exert substantial market
power, requiring a rethink of the appropriate regulatory and policy framework.
Data itself is an asset and the ownership and control of data is an important determinant
of power relations in value chains and over markets. The data are collected partly in or
der to provide the service in question and partly as a by-product of the service. Moreover,
the combination of data from different sources enables additional value to be offered.
This underpins the substantial network effects where the aggregator of data is able to of
fer better services by virtue of having more users across the different ‘sides’ of their plat
forms (for example, suppliers of goods and consumers). First movers are therefore able to
lock in their advantages.
The routes to customers are increasingly through one of a very few ‘technology giants’,
whether in online search, social networks, or e-commerce. These online platforms bring
together consumer data and analysis, logistics, and payments. Firms such as Amazon and
Alibaba have immense scale, strong brand identity, supplier (p. 167) networks, and data
capabilities. Multi-sided online platforms are able to aggregate and analyse data from
groups of users.
Page 17 of 31
The technology giants have been able to grow so fast partly because they have been al
lowed to acquire small and innovative companies without much consideration given to the
possible harmful long-run effects on competition. Between 2008 and 2018 Google ac
quired 168 companies, Facebook acquired seventy-one companies and Amazon acquired
sixty companies (LEAR, 2019). The reviews carried out in 2018 and 2019 for various gov
ernments and competition authorities, including those in the EU, Australia, and the Unit
ed Kingdom, find that merger evaluation needs to recognize the competitive value of da
ta, the harm to potential competition, and the possible conglomerate effects of such
mergers.3 In other words, the assessment needs to extend beyond static analysis of
whether there will be a lessening of competition from the merger due to firms being in
horizontal or vertical relationships.
The convergence of platforms which operate in and across payments systems, retailing,
logistics, customer information and marketing, and telecoms requires regulation through
a flexible and responsive regime. In addition, an enabling environment for the digitaliza
tion of business, including more reliable and cheaper connectivity and an urgent upgrad
ing of the skills pool in big-data analytics, is urgently required to open up markets to local
entrepreneurs.
The evolving nature of economic power that arises from the digitalization of economic ac
tivity and the importance of digital platforms has huge implications for industrial develop
ment and industrial policy. The governance of value chains and production networks, as
well as the very determination of what constitutes the value (including for tax purposes)
is controlled by the aggregator of the core data in the form of the online platforms. The
growth and competitiveness of local businesses depends on the terms on which they can
insert themselves into, and/or interface with, the dominant international platforms.
Europe has led competition enforcement of digital platforms with three decisions against
Google as of September 2019. These cases highlight the mechanisms by which power is
exerted to shape markets. The European Commission’s Google Shopping case addressed
the ability to preference partners in search results, restricting consumer choices, and
raising prices paid by consumers.4 Similar concerns relating to consumers’ online search
ing behaviour have been identified in UK competition and consumer (p. 168) protection
matters related to digital comparison tools and online hotel booking.5 India has taken a
decision regarding Google preferencing searches for flights.
The EC’s Google Android case related to Google extending and protecting market power
with regard to restrictive terms applied to the Google Playstore for apps on mobile
phones, as well as to the development of alternative Android operating systems.6 The
Google Adsense case relates to Google extending power over advertising space on third-
party websites.7 These cases emphasize the evolving ways in which routes to market now
depend increasingly on the online reach to consumers including online advertising.
The competition policy recommendations that have been made in the various reviews in
clude changing the standards in merger control to consider a ‘balance of harms’ test, and
designating companies as having ‘strategic market status’ with greater obligations not to
Page 18 of 31
distort competition. To ensure that there is consistency of regulation and competition en
forcement some reviews have proposed establishing a regulatory ‘data unit’ with powers
to obtain information, and timeously make and enforce orders. This recognizes the cen
tral role that control over, and access to, data has for economic participation.
6.5.3 Industrial Policy Implications
The rise of digital platform businesses as the largest global corporations, and the impor
tance of data for the competitiveness of firms of all sizes, poses fundamental challenges
for industrial policy. The major technological changes in the ‘new digital economy’ re
quire business models that can integrate, coordinate, and control value creation across
businesses.
Industrial policy and economic regulation similarly need to be consistent and support
public organizations that set the standards required to support the interconnection and
integration of components and subsystems into larger systems. Three business models
have been identified as key to these goals (Sturgeon, 2017):
• Modularity describes a business model based on interchangeability, where sub-com

ponents can be added or subtracted without redesigning entire systems. On the facto
ry floor, different sub-assemblies with shared interfaces can be substituted in the as
sembly of larger products. In product design, off-the-shelf or lightly customized modu
lar components can be designed in as elements of larger systems. In supply chains,
standards and protocols allow for complex information about (p. 169) products, produc
tion, and logistics to be exchanged across organizational and geographic boundaries.8
• Open innovation refers to the pre-competitive pooling of R&D activities and design
criteria, either through consortia, or through the voluntary ‘crowdsourcing’ efforts of
engineers and technologists interested in creating free resources for their communi
ties. For example, nearly all the world’s major computer programming languages, such
as Python, are open source and free. Like modularity, open innovation helps companies
‘vertically specialize’, that is, develop a strategic focus on a specific bundle of compe
tencies, while still providing customers with a rich set of fully functional products and
solutions.
• Platforms provide services for networks of users, bringing together those using the
platform to sell goods and services with consumers looking to purchase, effectively
shaping and creating the markets themselves. The platform owners benefit from ag
gregating user data (including from key components such as geo-location and map
ping) and charging for advertising, and can vertically integrate into the suppliers of
the goods and services. The platforms can then channel consumers to the platform’s
preferred services.
Digitalization means that a coordinated industrial policy which prioritizes key sectors and
cross-cutting activities is even more important. The spillover effects mean that industry-
specific and cross-cutting technology and skills support are both critical. This includes
the key enablers of skills in software engineering, data science, and related ICT skills,
Page 19 of 31
along with intermediate industries such as electronic control systems, logistics, design,
and additive manufacturing. Development finance for investments in productive capabili
ties, upgrading of capital equipment, and supply-chain integration remains essential.
The significance of digital platforms which straddle communication, media, publishing,

search, e-commerce, finance, and payments requires a rethink of competition and regula
tion to establish rules for platforms to ensure local businesses are not discriminated
against. Legal provisions asserting ownership of data on behalf of citizens are a neces
sary step at the national level, while countries such as Rwanda and India have taken
steps to ensure local businesses are not undermined by transnational e-commerce busi
nesses. Effective regulation is an industrial policy imperative when control over data is a
barrier to being able to compete. This needs to be coupled with vertical policies to sup
port stronger local linkages.
While policy design and the governance framework are critical, the effective implementa
tion and enforcement of any industrial policy will depend on government capacity and
laws to ensure access to data. The implementation of a digital industrial policy requires
the establishment of high-level coordination capacity, institutionalized private-sector in
puts, and appropriate monitoring and evaluation systems. New (p. 170) coalitions for
change forged around better aligned productive interests are necessary for the success
ful implementation of industrial policy, enabling sector-specific and cross-sectoral inter
ventions.
6.6 Conclusion
Enterprises are at the centre of understanding the processes of industrialization, and the
potential for, and constraints on, industrial policy. A dynamic private sector is critical for
catch-up by developing countries (Ohno, 2013). The transaction-costs theories of the firm
which have dominated economics are, however, effectively unable to come to terms with
the most important questions of dynamic organizational capabilities—the strategic orien
tation of large internationalized businesses and the nature of power relations and the po
litical settlement.
The resource-based theory of the firm, developed through detailed empirical study within
and across countries, provides a robust framework for analysing the development of ca
pabilities. It has been extended in terms of the ways in which power is understood and
the internationalization of businesses. The implications of digitalization are subject to in
tense current research pointing to major changes in the nature of firm-level governance
of economic activity and the appropriate industrial policies.
Substantial economic power on the part of firms is an intrinsic feature of industrial devel
opment given the dynamic economies involved. There is therefore a natural tension be
tween the short-term objectives of firms in securing and exploiting rents, and the longer-
term investments required to build dynamic productive capabilities taking into account
the uncertainties involved. This tension is at the heart of an effective industrial policy
Page 20 of 31
which has analysis of firms located in a production-centric framework. A critical objective

of industrial policy is therefore to shift the calculus on the part of large firms, to be able
to set and enforce conditionalities for the support required. The effective engagement
with large firms includes locating them in clusters, with long-term relationships with
smaller firms as subcontractors, through policies supporting shared research and devel
opment, skills improvement, ongoing capabilities upgrading, and product development.
The greater emphasis which is now being placed on concentration of ownership and con
trol in many countries is welcome. It is part of a rethink of competition policy which rec
ognizes competition as a dynamic process of rivalry underpinning ongoing improvements
in productivity. An effective framework for competition and economic regulation, which
sets the rules for markets, complements industrial policies. Assessing the debates about
concentration and the appropriate policy responses requires attention to the influence of
coalitions of large business interests over the policy agenda in indirect and diffuse ways,
as well as alternative coalitions which can underpin progressive industrial policies.
The internationalization of businesses in the context of the digitalization of eco

(p. 171)
nomic activity raises important questions about the ability of regulations and industrial
policies at a national level to discipline the conduct of businesses. Large developing coun
tries have greater scope to take appropriate measures, including supporting alternative
digital platforms, in which China is leading. Countries with smaller economies, such as
those in Africa, are in relatively weak positions and depend on a broader international de
velopment coalition to underpin effective regulation of the global digital platforms. At the
national level, countries need to ensure access to data for local businesses as part of a
digital industrial policy platform to support local capabilities in targeted sectors and
cross-cutting measures for the critical skills and investment required to adapt to the digi
talization of economic activity.
References
Amsden, Alice (1989) Asia’s Next Giant—South Korea and Late Industrialization. New
York and Oxford: Oxford University Press.
Amsden, Alice (1997) ‘Bringing Production back in: Understanding Government’s Eco
nomic Role in Late Industrialization’, World Development 25(4): 469–80.
Amsden, Alice (2001) The Rise of the ‘Rest’: Challenges to the West from Late-industrial
izing Economies. New York: Oxford University Press.
Amsden, Alice and Ajit Singh (1994) ‘The Optimal Degree of Competition and Dynamic Ef
ficiency in Japan and Korea’, European Economic Review 38: 940–51.
Andreoni, Antonio (2016) ‘Varieties of Industrial Policy: Models, Packages, and Transfor
mation Cycles’, in Akbar Noman and Joseph E. Stiglitz (eds) Efficiency, Finance and Vari
eties of Industrial Policy. New York: Columbia University Press, pp. 245–305.
Page 21 of 31
Andreoni, Antonio (2018) ‘The Architecture and Dynamics of Industrial Ecosystems: Di
versification and Innovative Industrial Renewal in Emilia Romagna’, Cambridge Journal of
Economics 42(6): 1613–42.
Arrow, Kenneth (1962) ‘Allocation of Resources for Invention’, in The Rate and Direction
of Inventive Activity: Economic and Social Factors, National Bureau of Economic Re
search. Princeton, NJ: Princeton University Press.
Atkinson, Anthony (2015) Inequality: What Can Be Done? Cambridge, MA: Harvard Uni
versity Press.
Audretsch, David B. and Julie A. Elston (2002) ‘Does Firm Size Matter? Evidence on the
Impact of Liquidity Constraints on Firm Investment Behavior in Germany’, International
Journal of Industrial Organization 20(1): 1–17.
Australian Competition and Consumer Commission (ACCC) (2019) Digital Platform In
quiry Final Report.
Azar, José, Sahil Raina, and Martin C. Schmalz (2016) ‘Ultimate Ownership and Bank
Competition’, SSRN Electronic Journal. 10.2139/ssrn.2710252. (p. 172)
Azar, José, Martin C. Schmalz, and Isabel Tecu (2018) ‘Anticompetitive Effects of Common
Ownership’, The Journal of Finance 73: 1513–65.
Babbage, Charles (2005 [1832]) On the Economy of Machinery and Manufactures.

Cirencester: Echo Library.
Bair, Jennifer (2009) ‘Global Commodity Chains’, in Jennifer Bair (ed.) Frontiers of Com
modity Research. Palo Alto, CA: Stanford University Press.
Bairoch, Paul and Richard Kozul-Wright (1996) ‘Globalization Myths: Some Historical Re
flections on Integration, Industrialization, and Growth in the World Economy’. Discussion
Paper No. 113. Geneva: UNCTAD.
Baker, Jonathan and Steve Salop (2015) ‘Antitrust, Competition Policy, and Inequality’,
The Georgetown Law Journal Online 104: 1.
Convergence. Cambridge, MA: Harvard University Press.
Best, Michael H. (1990) The New Competition. Cambridge: Polity Press.
Page 22 of 31
Bloom, Nicholas, Erik Brynjolfsson, Lucia Foster, Ron Jarmin, Megha Patnaik, Itay Sapor
ta-Eksten, and John Van Reenen (2017) ‘What Drives Differences in Management?’. NBER
Working Paper No. 23300. Cambridge, MA: NBER.
Bloom, Nicholas, John Van Reenen, and Heidi Williams (2019) ‘A Toolkit of Policies to Pro
mote Innovation’, Journal of Economic Perspectives 33(3): 163–84.
Budzinski, Oliver and Maryam H. A. Beigi (2015) ‘Generating Instead of Protecting Com
petition’, in Michal S. Gal, Mor Bakhoum, Josef Drexl, Eleanor M. Fox, and David J. Ger
ber (eds) Economic Characteristics of Developing Jurisdictions: Their Implications for
Competition Law. Cheltenham: Edward Elgar, pp. 223–47.
Cantwell, John (1994) ‘The Relationship between International Trade and International
Production’, in David Greenaway and L. Alan Winters (eds) Surveys in International
Trade. Oxford: Blackwell, pp. 303–28.
Chandler, Alfred Jr. (1990) Scale and Scope: The Dynamics of Industrial Capitalism.
Chandler, Alfred Jr., Franco Amatori, and Takashi Hikino (eds) (1997) Big Business and
the Wealth of Nations. Cambridge: Cambridge University Press.
Chandler, Alfred Jr., Peter Hagström, and Örjan Sölvell (eds) (1998) The Dynamic Firm.
Chang, Ha-Joon (1996) The Political Economy of Industrial Policy. New York: St. Martin’s
Press.
Coase, Ronald (1937) ‘The Nature of the Firm’, Economica 4: 386–405.
Coase, Ronald (1992) ‘The Institutional Structure of Production’, American Economic Re
view 82(4): 713–19.
Coe, Neil M. and Henry W. Yeung (2015) Global Production Networks: Theorizing Eco
nomic Development in an Interconnected World. Oxford: Oxford University Press.
Cohen, Wesley M. and Richard Levin (1989) ‘Empirical Studies of Innovation and Market
Structure’, in Richard Schmalensee and Robert D. Willig (eds) Handbook of Industrial Or
ganization, Vol. II. Amsterdam: North Holland, pp. 1059–107.
Committee for the Study of Digital Platforms (2019) ‘Report of the Market Structure and
Antitrust Subcommittee’, Stigler Centre Committee for the Study of Digital Platforms
Market Structure and Antitrust Subcommittee, chaired by Fiona Scott Morton. (p. 173)
Cremer, Jacques, Yves-Alexandre de Montjoye, and Heike Schweitzer (2019) Competition

Policy for the Digital Era. Brussels: European Commission.
Dallas, Mark, Stefano Ponte, and Timothy Sturgeon (2019) ‘Power in Global Value
Chains’, Review of International Political Economy 26(4): 666–94.
Page 23 of 31
De Loecker, Jan and Jan Eeckhout (2017) ‘The Rise of Market Power and the Macroeco
nomic Implications’. NBER Working Paper No. 23687. Cambridge, MA: NBER.
Dosi, Giovanni (1997) ‘Organizational Competencies, Firm Size, and the Wealth of Na
tions: Some Comments from a Comparative Perspective’, in Alfred Chandler, Jr., Franco
Amatori, and Takashi Hikino (eds) Big Business and the Wealth of Nations. Cambridge:
Cambridge University Press, pp. 465–79.
Dosi, Giovanni, David J. Teece, and Joseph Chytry (eds) (1998) Technology, Organization,
and Competitiveness: Perspectives on Industrial and Corporate Change. Oxford: Oxford
University Press.
Dunning, John (1992) Multinational Enterprises and the Global Economy. Wokingham: Ad
dison-Wesley.
Dunning, John (1993) The Globalization of Business: The Challenge of the 1990s. London:
Routledge.
Ennis, Sean, Pedro Gonzaga, and Chris Pike (2019) ‘Inequality: A Hidden Cost of Market
Power’, Oxford Review of Economic Policy 35(3): 518–49.
Federico, Giulio, Fiona Scott Morton, and Carl Shapiro (2019) ‘Antitrust and Innovation:
Welcoming and Protecting Disruption’. NBER Working Paper No. 26005. Cambridge, MA:
NBER.
Ford, Martin (2015) Rise of the Robots: Technology and the Threat of a Jobless Future.
New York: Basic Books.
Fox, Eleanor (2003) ‘We Protect Competition, You Protect Competitors’, World Competi
tion 26(2): 149–65.
Fujimoto, Takahiro (1998) ‘Reinterpreting the Resource-capability View of the Firm: A

Case of the Development-Production Systems of the Japanese Auto-Makers’, in Alfred D.
Chandler, Peter Hagström, and Örjan Sölvell (eds) The Dynamic Firm: The Role of Tech
nology, Strategy, Organization, and Regions. Oxford: Oxford University Press, pp. 15–44.
Furman, Jason, Diane Coyle, Amelia Fletcher, Derek McAuley, and Philip Marsden (2019)
‘Unlocking Digital Competition’. Report of the Digital Competition Expert Panel, March.
London: Crown Publications.
Gal, Michal (2009) ‘Antitrust in a Globalised Economy: The Unique Enforcement Chal
lenges Faced by Small and Developing Jurisdictions’, Fordham International Law Journal
33(1): 156.
Gerber, David (2010) Global Competition: Law, Markets, and Globalisation. Oxford: Ox
Page 24 of 31
Gereffi, Gary (2014) ‘Global Value Chains in a Post-Washington Consensus World’, Review
of International Political Economy 21(1): 9–37.
Gereffi, Gary and Karina Fernandez-Stark (2011) ‘Global Value Chain Analysis: A Primer’.
Center on Globalization, Governance & Competitiveness (CGGC), Duke University, North
Carolina.
Gereffi, Gary, John Humphrey, and Timothy Sturgeon (2005) ‘The Governance of Global
Value Chains’, Review of International Political Economy 12: 78–104. (p. 174)
Geroski, Paul and Alexis Jacquemin (1984) ‘Dominant Firms and their Alleged Decline’,
International Journal of Industrial Organisation 2(1): 1–27.
Goergen, Marc and Luc Renneboog (2001) ‘Investment Policy, Internal Financing and
Ownership Concentration in the UK’, Journal of Corporate Finance 7(3): 257–84.
Helmsing, Albert H. J. (2001) ‘Externalities, Learning and Governance: New Perspectives

on Local Economic Development’, Development and Change 32: 277–308.
Hikino, Takashi and Alfred Chandler, Jr. (1997) ‘Historical and Comparative Contours of
Big Business’, in Alfred Chandler, Jr., Franco Amatori, and Takashi Hikino (eds) Big Busi
ness and the Wealth of Nations. Cambridge: Cambridge University Press, pp. 3–23.
Hur, Joseph Seon (2004) ‘The Evolution of Competition Policy and its Impact on Economic
Development in Korea’, in Philippe Brusick, Ana Maria Alvarez, Lucian Cernat, and Philip
Holmes (eds) Competition, Competitiveness and Development: Lessons from Developing
Countries. Geneva: United Nations Conference on Trade and Development, pp. 227–49.
Hymer, Stephen (1976) The International Operations of National Firms: A Study of Direct
Foreign Investment. Cambridge, MA: MIT Press.
Inderst, Roman and Nicola Mazzarotto (2008) ‘Buyer Power in Distribution’, in W. D.

Collins (ed.) ABA Antitrust Section Handbook: Issues in Competition Law and Policy,
American Bar Association.
Inderst, Roman and Tommaso M. Valletti (2011) ‘Buyer Power and the “Waterbed Effect”’,
Journal of Industrial Economics 59(1): 1–20.
Jensen, Michael C. and William H. Meckling (1976) ‘Theory of the Firm: Managerial Be
havior, Agency Costs and Ownership Structure’, Journal of Financial Economics 3(4): 305–
60.
Kaplinsky, Raphael (2000) ‘Globalisation and Unequalisation: What Can Be Learned from
Value Chain Analysis?’, Journal of Development Studies 37(2): 117–46.
Khan, Mushtaq (2009) ‘Learning, Technology Acquisition and Governance Challenges in

Developing Countries’. Research Paper Series on Governance for Growth. School of Ori
Page 25 of 31
ental and African Studies, University of London. Available at http://eprints.soas.ac.uk/

9967/1/Learning_and_Technology_Acquisition_internet.pdf.
Khan, Mushtaq (2010) Political Settlements and the Governance of Growth-enhancing In
stitutions. London: SOAS.
Khan, Mushtaq (2015) ‘Industrial Policy Design and Implementation Challenges’, in Jesus
Felipe (ed.) Development and Modern Industrial Policy in Practice: Issues and Country
Experiences. Cheltenham: Edward Elgar, pp. 94–126.
Khan, Mushtaq (2017) ‘Introduction: Political Settlements and the Analysis of Institu
tions’, African Affairs Special Issue: 1–20.
Khan, Mushtaq (2019) ‘Knowledge, Skills and Organizational Capabilities for Structural
Transformation’, Structural Change and Economic Dynamics 48: 42–52.
Khan, Mushtaq and Stephanie Blankenburg (2009) ‘The Political Economy of Industrial
Policy in Latin America’, in Giovanni Dosi, Mario Cimoli, and Joseph E. Stiglitz, Industrial
Policy and Development: The Political Economy of Capabilities Accumulation. Oxford: Ox
Korea Fair Trade Commission (KFTC) (2011) ‘30 Years of KFTC, Looking back, Moving
forwards’, Annual Report 2011, Special Edition. Seoul: KFTC.
Lall, Sanjaya (1993) ‘Introduction: Transnational Corporations and Economic Develop

ment’, in Sanjaya Lall (ed.) Transnational Corporations and Economic Development. Lon
don: Routledge. (p. 175)
Lall, Sanjaya (1994) ‘The East Asian Miracle: Does the Bell Toll for Industrial Strategy?’,
World Development 22(4): 645–54.
Lall, Sanjaya and Morris Teubal (1998) ‘“Market-Stimulating” Technology Policies in De
veloping Countries: A Framework with Examples from East Asia’, World Development
26(8): 1369–85.
Lazonick, William (2010) ‘The Chandlerian Corporation and the Theory of Innovative En
terprise’, Industrial and Corporate Change 19: 317–49.
LEAR (2019) ‘Ex post Assessment of Merger Control Decisions in Digital Markets’, Report
for the UK Competition and Markets Authority.
Lee, Keun, Marina Szapiro, and Zhuqing Mao (2018) ‘From Global Value Chains (GVC) to
Innovation Systems for Local Value Chains and Knowledge Creation’, European Journal of
Development Research 30(3): 424–41.
Makhaya, Gertrude and Simon Roberts (2013) ‘Expectations and Outcomes—Considering

Competition and Corporate Power in South Africa under Democracy’, Review of African
Political Economy 138: 556–71.
Page 26 of 31
Makhaya, Gertrude and Simon Roberts (2014) ‘The Changing Strategies of Large Corpo
rations in South Africa under Democracy and the Role of Competition Law’. CCRED Work
ing Paper No. 2014/2.
Mathis, Jérôme and Wilfried Sand-Zantman (2014) ‘Competition and Investment: What Do
We Know from the Literature?’. Institut d’Economie Industrielle Working Paper. Toulouse:
Institut d’Economie Industrielle.
McAfee, Andrew and Erik Brynjolfsson (2017) Machine, Platform, Crowd. New York: W. W.
Norton.
Newham, Melissa, Jo Seldeslachts, and Albert Banal Estañol (2018) ‘Common Ownership
and Market Entry: Evidence from Pharmaceutical Industry’. DIW Berlin Discussion Paper
No. 1738.
North, Douglass C., John Wallis, and Barry Weingast (2009) Violence and Social Orders.
Ohno, Kenichi (2013) Learning to Industrialize: From Given Growth to Policy-aided Value
Creation. Oxford: Routledge.
Oqubay, Arkebe and Justin Yifu Lin (eds) (2019) China–Africa and an Economic Transfor
mation. Oxford: Oxford University Press.
Oqubay, Arkebe and Kenichi Ohno (eds) (2019) How Nations Learn: Technological Learn
ing, Industrial Policy, and Catch-up. Oxford: Oxford University Press.
Penrose, Edith (1959) The Growth of the Firm. Oxford: Blackwell.
Petricevic, Olga and David Teece (2019) ‘The Structural Reshaping of Globalization: Im
plications for Strategic Sectors, Profiting from Innovation, and the Multinational Enter
prise’, Journal of International Business Studies 50(9): 1487–512.
Piketty, Thomas (2014) Capital in the Twenty-First Century. Cambridge: The Belknap
Press of Harvard University Press.
Pisano, Gary (2017) ‘Towards a Prescriptive Theory of Dynamic Capabilities: Connecting

Strategic Choice, Learning and Competition’, Industrial and Corporate Change 26(5):
747–62.
Polson, Nick and James Scott (2018) AIQ: How Artificial Intelligence Works and How We
Can Harness its Power for a Better World. London: Bantam Press.
Ponte, Stefano (2019) Business, Power, and Sustainability in a World of Global Value
Chains. London: Zed Books. (p. 176)
Ponte, Stefano and Timothy Sturgeon (2014) ‘Explaining Governance in Global Value
Chains: A Modular Theory-building Effort’, Review of International Political Economy
21(1): 195–223.
Page 27 of 31
Porter, Michael (1990) The Competitive Advantage of Nations. London: Macmillan.
Rehnberg, Martha and Stefano Ponte (2017) ‘From Smiling to Smirking? 3D Printing, Up
grading and the Restructuring of Global Value Chains’, Global Networks 18(1) doi.org/
10.1111/glob.12166.
Rey, Patrick and Jean Tirole (2006) ‘A Primer on Foreclosure’, in Mark Armstrong and
Robert H. Porter (eds) Handbook of Industrial Organization III. Amsterdam: North Hol
land, pp. 2145–220.
Richardson, George B. (1972) ‘The Organization of Industry’, The Economic Journal

82(327): 883–96.
Roberts, Simon (2012) ‘Administrability and Business Certainty in Abuse of Dominance

Enforcement: An Economist’s Review of the South African Record’, World Competition
35(2): 269–96.
Roberts, Simon (2013) ‘Competition Policy, Industrial Policy, and Corporate Conduct’, in
Joseph E. Stiglitz, Justin Yifu Lin, and Ebrahim Patel (eds) The Industrial Policy Revolu
tion II: Africa in the 21st Century. New York: Palgrave MacMillan, pp. 216–242.
Rodrik, Dani (1995) ‘Getting Interventions Right: How South Korea and Taiwan Grew
Rich’, Economic Policy 20: 53–108.
Rodrik, Dani (2018) ‘New Technologies, Global Value Chains, and the Developing
Economies’. CESifo Working Paper No. 7307.
Rosenberg, Nathan (1979) ‘Technological Interdependence in the American Economy’,

Technology and Culture 20(1): 25–50.
Sakakibara, Mariko and Michael E. Porter (2001) ‘Competing at Home to Win Abroad: Ev
idence from Japanese Industry’, Review of Economics and Statistics 83(2): 310–22.
Schumpeter, Joseph A. (1942) Capitalism, Socialism, and Democracy. New York: Harper
and Brothers.
Seldeslachts, Jo, Melissa Newham, and Albert Banal-Estanol (2017) ‘Changes in Common
Ownership of German Companies’, DIW Research Bulletin 7(30): 303–11.
Shapiro, Carl (2012) ‘Competition and Innovation: Did Arrow Hit the Bull’s Eye?’, in Josh
Lerner and Scott Stern (eds) The Rate and Direction of Inventive Activity Revisited.
Chicago, IL: University of Chicago Press.
Singh, Ajit (2002) ‘Competition and Competition Policy in Emerging Markets: Internation
al and Developmental Dimensions’. ESRC Centre for Business Research, Working Paper
No. 246. University of Cambridge.
Singh, Ajit (2004) ‘Multi-lateral Competition Policy and Economic Development’. UNC
TAD Series on Issues in Competition Law and Policy. Geneva: United Nations.
Page 28 of 31
Stiglitz, Joseph E. (2017) ‘Towards a Broader View of Competition Policy’, in Tembinkosi

Bonakele, Eleanor Fox, and Liberty Mncube (eds) Competition Policy for the New Era: In
sights from the BRICS Countries. Oxford: Oxford University Press, pp. 4–21.
Strange, Roger and John Humphrey (2019) ‘What Lies between Market and Hierarchy?
Insights from Internalization Theory and Global Value Chain Theory’, Journal of Interna
tional Business Studies 50(8): 1401–13.
Sturgeon, Timothy (2017) ‘The “New” Digital Economy and Development’. UNCTAD Tech
nical Notes on ICT for Development No. 8. Geneva: UNCTAD. (p. 177)
Sutton, John (2007) ‘Market Structure: Theory and Evidence’, in Robert H. Porter and
Mark Armstrong (eds) Handbook of Industrial Organization, Volume III. Amsterdam: Else
vier, pp. 2301–68.
Sutton, John (2012) Competing in Capabilities: the Globalisation Process. Oxford: Oxford
University Press.
Teece, David J. (2009) Dynamic Capabilities and Strategic Management: Organizing for
Innovation and Growth. Oxford: Oxford University Press.
Teece, David J. (2010) ‘Alfred Chandler and “Capabilities” Theories of Strategy and Man
agement’, Industrial and Corporate Change 19(2): 297–316.
Teece, David J. and Gary Pisano (1998) ‘The Dynamic Capabilities of Firms: An Introduc
tion’, in Giovanni Dosi, David J. Teece, and Josef Chytry (eds) Technology, Organization,
and Competitiveness: Perspectives on Industrial and Corporate Change. Oxford: Oxford
Teece, David J., Gary Pisano, and Amy Shuen (1997) ‘Dynamic Capabilities and Strategic
Management’, Strategic Management Journal 18(7): 509–33.
UNCTAD (2018) Trade and Development Report. Geneva: UNCTAD.
Van Reenen, John (2018) ‘Increasing Difference between Firms: Market Power and the
Macro-economy’, Proceedings of the 2018 The Federal Reserve Bank of Kansas City Eco
nomic Policy Symposium: Changing Market Structures and Implications for Monetary Pol
icy, Jackson Hole, WY, August.
Vickers, John (2005) ‘Abuse of Market Power’, The Economic Journal 115: 244–61.
Vilakazi, Thando and Simon Roberts (2019) ‘Cartels as “Fraud”? Insights from Collusion
in Southern and East Africa in the Fertilizer and Cement Industries’, Review of African
Political Economy 46(161): 369–86.
Whinston, Michael (2006) Lectures on Antitrust Economics. Cambridge, MA: MIT Press.
Page 29 of 31
Williamson, Oliver (1986) Economic Organisation: Firms, Markets, and Policy Control.
New York: Harvester Wheatsheaf.
Zingales, Luigi (2017) ‘Towards a Political Theory of the Firm’, Journal of Economic Per
spectives 31(3): 113–30.
Notes:
(1) These include corporations controlled by the Chinese state.
(2) This section draws on discussions with Rashmi Banga, Justin Barnes, Anthony Black,
Parminder Jeet Singh, Stefano Ponte, and Tim Sturgeon, for the Digital Industrial Policy
Issues paper produced by the Industrial Development Think Tank at the University of Jo
hannesburg.
(3) The UK and EU reviews are respectively Furman et al. (2019) and Cremer et al.
(2019). For Australia see Australia Competition and Consumer Commission Digital Plat
form Inquiry Final Report, 2019. For Germany see Report by the Commission ‘Competi
tion Law 4.0’, ‘A New Competition Framework for the Digital Economy’, report for the
Federal Ministry for Economic Affairs and Energy, available at https://www.bmwi.de/
Redaktion/EN/Publikationen/Wirtschaft/a-new-competition-framework-for-the-digital-
economy.html.
(4) European Commission, Decision of 27 June 2017, case AT.39740–Google Search (Shop
ping).
(5) UK Competition and Markets Authority: press release of 13 September 2019, available
at https://www.gov.uk/government/news/major-overhaul-of-hotel-booking-sector-after-
cma-action; report on Digital Comparison Tools Market Study, available at https://
assets.publishing.service.gov.uk/media/59c93546e5274a77468120d6/digital-comparison-
tools-market-study-final-report.pdf.
(6) European Commission, Decision of 18 July 2018, Case AT.40099–Google Android.
(7) European Commission, Decision of 20 March 2019, AT.40411–Google Search

(AdSense).
(8) The standards and protocols supporting value-chain modularity are often embedded in
digital ICT systems such as CAD/CAM and ERP.
Simon Roberts
Simon Roberts is a professor of economics at the University of Johannesburg and an

economics director at the UK Competition and Markets Authority. He founded and di
rected the Centre for Competition, Regulation and Economic Development (CCRED)
at the University of Johannesburg. Prior to this he was chief economist at Competi
tion Commission South Africa. He has worked extensively on issues of industrial de
Page 30 of 31
velopment, trade, regional value chains, competition and economic regulation in

Southern and East Africa, advising governments, competition authorities, and regula
tors. He has also testified as an expert witness in a number of major competition cas
es.
Page 31 of 31
Radical Perspectives on Industrial Policy

Sam Ashman, Susan Newman, and Fiona Tregenna
This chapter surveys, evaluates, and develops radical perspectives on industrial policy.
We examine the differences between mainstream and heterodox approaches to economic
and industrial development, and we look at the similarities and differences between
structuralist and Marxist approaches. We argue that Marx’s concern is not with sectors
but with value and the overall circuit of capital. Radical industrial policy foregrounds
class and capitalism, and integrates a distinctive conception of the state. We argue fur
ther that radical industrial policy operates at two levels, the analytical and the prescrip
tive, and we use the climate crisis to illustrate our argument. Radical industrial policy has
broad objectives, including fundamentally altering productive structures and dynamics to
wards labour-centred development. Finally, we briefly survey some experiences of radical
industrial policy, loosely dividing them between statist, co-operative, and participatory
planning approaches. We also reflect on the limits of industrial policy, especially under
globalized and financialized capitalism.
Keywords: industrial policy, industrialization, Marxism, radical, heterodox economics, developmental state, capi
talism, neo-liberalism
7.1 Introduction
INDUSTRIAL restructuring has been taking place in the Global North since the 1980s.
Ongoing de-industrialization, downsizing, outsourcing, off-shoring, and just-in-time pro
duction have had a profound impact on the conditions of work and on workers, marked by
worsening precarity and lower wages (Lazonick and O’Sullivan, 2000; Milberg and Win
kler, 2009; Tregenna, 2009, 2016). The last forty years has seen a dramatic rebalancing of
power in favour of capital, exemplified by the increasing share of income distributed to
capital and stagnant real wages for labour (Giovannoni, 2010; Stockhammer et al., 2008).
This has been most pronounced in state responses to the financial crisis, which have
amounted to bailouts for capital, austerity for workers, and anti-union legislation.
Page 1 of 29
This economic malaise, together with impending climate catastrophe, has generated
greater calls for inclusive and sustainable economic growth and development. Can
growth be made to be more equitable and socially and environmentally sustainable
through industrial restructuring and industrial development? Are there viable alternative
ways of organizing production so that it is more democratic? Can industrial policy con
tribute to shifting power relations—internationally in favour of peripheral countries, and
domestically in favour of the working class as well as in favour of women and other op
pressed groups. These are among the key concerns of radical industrial policy that go be
yond just capturing a greater share of global manufacturing production, and/or changing
the identity of the owners of industrial capital (for example, from former colonialist to in
digenous).
Radical industrial policy, we argue, emphasizes class structures and relations, including
the importance of exploitation, and the subjugation of labour to capital as (p. 179) central
to capitalist accumulation. It should be noted that class structures and relations are also
racialized and gendered, as expressed in the division of labour. Industrial development is
thus not gender neutral and occurs historically in interaction with prevailing structures of
oppression in a co-constitutive manner. For example, the gendered division of labour and
low wages for ‘feminized’ work is premised upon, and reproduces, social structures that
place the burden of unpaid social reproduction on women in the home. Industrial policy
choices can contribute to deepening or displacing patriarchal social structures. Chapter
16 in this volume, by Seguino, focuses on a gendered approach to industrialization and in
dustrial policy.1 Analytically, radical industrial policy must necessarily take into account
gendered social and economic relationships. Prescriptively, radical industrial policy would
also actively seek to promote a trajectory of industrial development in which women are
empowered and gender inequality is reduced.
This chapter aims to survey, evaluate, and develop radical perspectives on industrial poli
cy. Radical industrial policy is both analytic—understanding how capitalism works—and
prescriptive—suggesting what should guide industrial policy. We consider radical indus
trial policy as also aiming at fundamentally changing productive structures and dynamics
of accumulation in the direction of labour-centred economic and political restructuring.
‘Labour-centred’ refers not only to the distribution of gains from development in the in
terest of labour, but also to the central participatory and determining role of labour in the
process.
We begin the chapter by discussing radical approaches to the role of the manufacturing
sector and to industrialization, reflecting on how these approaches differ from both other
heterodox and mainstream approaches. We go on to analyse radical approaches to indus
trial policy, focusing on the conceptual and theoretical levels, with particular attention to
the inherent contradictions of capital‒labour relations under capitalism, the contradicto
ry role of the state, and radical industrial policy with respect to the case of the climate
crisis. We next discuss historical experiences of radical industrial policies with specific
reference to statist, local co-operative, and participatory planning models of industrial
policy and industrial development. By way of conclusion, we consider what is distinctive
Page 2 of 29
about radical industrial policy and what value such an approach adds. We also reflect on
the limitations of industrial policy, especially under globalized and financialized capital
ism. The effects of industrial policy on the balance of power are likely to be contingent
and conjuncturally specific, and we discuss the factors affecting these outcomes. Finally,
we consider the possibilities for radical industrial policies in current global and national
conditions. (p. 180)
7.2 Radical Approaches to the Role of Manufac

turing and Industrialization
Here we discuss radical approaches to the role of manufacturing in growth and develop
ment at a theoretical level. Rather than attempting to survey all ‘radical’ approaches to
these issues, we focus on the structuralist and Kaldorian traditions, and on a Marxist ap
proach.
The approach to sectors is one of the important dimensions along which mainstream and
heterodox economics can be distinguished. Mainstream economic approaches are gener
ally sector neutral in the sense that the marginal effects of an additional unit of value
added or of labour on growth do not vary by sector. Sectors are seen primarily as a way of
classifying economic activities. While there is recognition of common characteristics with
in sectors, the growth process itself is not seen as sector specific.
In contrast, heterodox approaches tend to view growth as sector specific. This is in the
sense of the composition of output and employment affecting the potentiality of growth,
and marginal increases in value added or labour affecting growth differently depending
on the sector in which these increases occur. More specifically, the manufacturing sector
is deemed to have a special role to play as an engine of economic growth, as discussed in
section 7.2.1 below.
7.2.1 Kaldorian and Structuralist Approaches: Manufacturing as a

Special Engine of Growth
Among the various heterodox schools of thought, it is in Kaldorian and structuralist ap
proaches that the sector specificity of growth and the special role of manufacturing come
out most clearly.2 The structuralist approach to development was first developed by econ
omists associated with the Economic Commission for Latin America and the Caribbean
(ECLAC). ECLAC identified the dual problems of foreign exchange constraints and the
tendency towards deterioration in the terms of trade for commodity exports as holding
back sustainable growth in developing countries. These problems derive both from the
structure of production in developing countries, and from the nature of articulation be
tween ‘periphery’ countries on the one hand, and ‘core’ or ‘centre’ economies on the oth
er hand. In structuralist thought, manufacturing matters and growth is product specific
both on the supply side and on the demand side. From this perspective, industrialization
Page 3 of 29
is the key—or in fact the only viable—path (p. 181) through which developing countries
can catch up with advanced economies (Blankenburg et al., 2008).
The importance of manufacturing as an engine of growth was theorized and formalized

through Kaldor’s three laws (Kaldor, 1966, 1978, 1980). In brief, the first of these laws
states that faster output growth in manufacturing is associated with faster aggregate eco
nomic growth. The second law, also known as Verdoorn’s Law (see Verdoorn, 1993
[1949]), states that productivity growth in manufacturing is endogenous to the growth of
manufacturing output. According to the third law, aggregate productivity growth is posi
tively related to the growth of manufacturing output and employment. The Kaldorian
school of thought thus emphasizes the special role of manufacturing in driving economic
growth. Kaldor’s views of increasing returns build on the ideas of Young (1928), who em
phasized the importance of dynamic increasing returns as distinct from firm-level static
increasing returns. Cramer and Tregenna (2020) argue that one of the ways in which
mainstream and heterodox approaches differ is in the recognition and treatment of in
creasing returns. In neoclassical thought, increasing returns are typically either ignored
or are dealt with in an extremely limited fashion (see Best, 2018).3 Superior scope for dy
namic increasing returns in manufacturing, relative to other sectors, is one of the key
premises for a special role for manufacturing and for the emphasis on industrialization.
Schumpeter’s wide-ranging contributions to economics also have implications for under

standing the dynamics of accumulation, including around business cycles and long waves,
innovation, and entrepreneurship. While Schumpeter’s approach is fundamentally differ
ent from Marx’s, he does draw on insights from Marx, including around the instability of
capitalism and the inevitability of crises (see Rosenberg, 2011). Schumpeter places less
emphasis on class and exploitation, and more on the role of the entrepreneur. (See
Cimoli, Dosi, and Yu, Chapter 4 in this volume, on evolutionary and neo-Schumpeterian
perspectives on technological change and industrial policy.)
Associated more or less directly with these traditions within heterodox economics have
been various applied studies, especially over the past three decades or so, of country and
regional industrialization experiences. Seminal contributions on East Asian industrializa
tion include Amsden (1989, 2001), Chang (2003), and Wade (1990). Oqubay (2015) is a
recent study of the case of Ethiopia, from a similar theoretical standpoint. These studies
bear out the ongoing relevance of industrialization.4 We return below to the implications
of different theoretical perspectives for industrial policy and to an appraisal of experi
ences of radical industrial policy. (p. 182)
7.2.2 A Marxist Approach to Manufacturing and Industrialization
Marxist economics takes a fundamentally different starting point from these other hetero
dox approaches and uses different categories of analysis. While Kaldorian/structuralist
traditions differ from mainstream economics in their approach to the sector specificity or
sector neutrality of growth, they share a view of economic activity as categorized in the
first instance in sectors. In Marx, by contrast, the basic issue is whether or not an activity
Page 4 of 29
is productive of surplus value (Tregenna, 2011, 2018). Surplus value is generated in the
(capitalistic) production of commodities. The primary basis for classifying economic activ
ities in a Marxist schema is thus according to their position in the circuit of capital, and
this does not fully coincide with sectoral distinctions.
While sectors are not the starting point in a Marxist approach, such an approach is also
not sector neutral. Marx does discuss sectors at length in various works, and identifies a
special role for manufacturing in the processes of accumulation and growth. In fact, Tre
genna (2013) argues that the special characteristics of manufacturing that are associated
with the structuralist and Kaldorian traditions (see earlier discussion) can be traced to
Marx’s own writings.
What Marx terms ‘modern machine industry’ is central to what he sees as the progressivi
ty and dynamism of the capitalist accumulation process. He regards mechanization and
industrialization as closely linked with the overcoming of human limitations in the pro
duction process, with competition between firms, with the concentration of ownership
and control and the associated emergence of large-scale enterprises, and with the inter
nationalization of production (Marx, 1976 [1867], 1984 [1894]; Marx and Engels, 1938
[1845/6]). Marx’s analysis was animated by the dramatic transformations rendered by the
First Industrial Revolution; he observed both the exploitation and social misery occa
sioned by this, as well as the unprecedented pace of accumulation and of technological
and economic progress.
More specifically, Marx identifies the features of manufacturing that lend it dynamism
and provide the basis for sustained growth in the rate of relative surplus value as includ
ing: scope for both the division of labour and the socialization of labour; technological ad
vancement; mechanization; learning by doing; increasing returns to scale; and overall,
stronger potential for cumulative productivity increases (Tregenna, 2013, 2018; see also
Ricoy, 2003). We can perceive here the commonalities between Marx’s view of manufac
turing and the Kaldorian and structuralist traditions that emerged in the second half of
the twentieth century.
For Marx, cumulative causation in manufacturing provides the basis for sustained growth
in the productivity of labour, and thence in the rate of relative surplus value. With the
rate of relative surplus-value production being central to the rate of accumulation, we can
then see the sector specificity of growth in a Marxist approach (see Tregenna, 2014).
As with non-Marxist radical approaches, Marx sees industrialization not only as

(p. 183)
central to accumulation and growth, but as transformative of wider social relations (see,
for example, Marx and Engels, 1938 [1845/6]). Of course, an important difference be
tween Marxist and other radical approaches is that, in Marxism, the analytical framework
is part of a broader political project for revolutionizing the way that production is orga
nized and the very nature of society, as discussed elsewhere in this chapter. Furthermore,
while Marx is concerned with accumulation and growth and with the dynamism of capital
Page 5 of 29
ism, it is exploitation and class relations that take centre stage rather than economic
growth and catching up between countries.
Both Marxist and structuralist/Kaldorian perspectives recognize the sector specificity of

growth, but a Marxist analysis is arguably less ‘sector fundamentalist’ than in the struc
turalist and Kaldorian traditions. From a Marxist stance, there is a great deal of hetero
geneity within sectors, as sectors comprise activities with totally different relationships to
the production and distribution of surplus value (see Tregenna, 2011, 2014). Manufactur
ing is seen as having a special role to play in accumulation and growth, yet not all manu
facturing activities necessarily have greater scope for producing relative surplus value
than all (capitalist commodity-producing) activities in other sectors.
In this section, we have discussed different approaches to sector specificity and to the
role of manufacturing and to industrialization. We have drawn out the commonalities be
tween different radical approaches (in contradistinction to mainstream economics), as
well as some pertinent differences between them. Different conceptualizations of manu
facturing and industrialization discussed here are part of fundamentally different political
projects, with different concerns and aspirations. Moreover, these conceptualizations lead
to different policy perspectives, in particular around the role and nature of industrial poli
cy. This is the focus of section 7.3.
7.3 Marxism and Industrial Policy: Beyond Sec

tors and Development
Marx’s analysis of capital accumulation and of capitalism goes beyond what might be
called a technical analysis of the role of manufacturing in industrialization or a sector-
specific conception of growth. Marx’s concern is with the overall dynamics of accumula
tion and how they are founded upon, and reproduce, class differences and entail process
es of class exploitation. As such, Marx is concerned with issues of the creation, circula
tion, and distribution of value, the power relations involved in this and the struggles that
ensue, as well as with crises, as central dynamics of accumulation. In this section we
look, first, at how Marx goes beyond the technical in ways that are both similar to, and
different from, structuralist analyses. Second, we argue that radical approaches to indus
trial policy can be understood at both analytical and prescriptive (p. 184) levels. Third, we
use the example of the climate crisis to argue the case for radical industrial policy today.
7.3.1 Value, Exploitation, and Power in Capitalism
Through his critique of classical political economy, Marx developed an analysis of capital
accumulation that demonstrates its dynamic and radical nature, particularly when com
pared to earlier modes of production, but also its exploitative and destructive character.
Competition between rival units of capital drives capitalists to innovate and to restruc
ture production ‘on pain of death’, producing a tendency to drive up productivity and for
capital to expand across the globe. This makes technical change far more rapid and sys
Page 6 of 29
tematic than in previous modes of production. At the same time, capitalist accumulation
is based upon the subjugation of labour to capital, as property-less labour is forced to sell
its labour power to capital in order to survive, and competition compels capitals to seek
new ways to increase surplus-value extraction.
Capitalist accumulation is therefore inherently unequal and contradictory, and it is not

only the market and the sphere of exchange, but also the relations of production, that are
critical for Marx (Brenner, 1985; Weeks, 2011). It follows that an equitable society re
quires a profound reorganization of society and the reorientation of work and production
to meet human needs and not simply the goal of profit (Spencer, 2009).5 The problematic
for Marx, then, is that of capitalist industrial development, not industrial development per
se. The state, the subject of some debate within Marxism (see, for example, Clarke, 1991),
is also viewed as part of capitalist social relations.
Marx’s analysis is inherently concerned with power relations and dynamics in under
standing any economy, and in understanding world economy. A concern with power rela
tions is also true of structuralist approaches. Prebisch, for example, highlights several
reasons for the declining terms of trade for developing-country exports that are political
in nature and/or relate to industrial organization and industrial relations; he underscores
the power of the centre to maintain control over the benefits of technology; he points to
the need for ‘social legislation’ to ensure real wage increases (in tandem with industrial
development); and posits the overall goal of industrialization to be to ‘increase the mea
surable well-being of the masses’ (Prebisch, 1950: 6).
Hirschman, too, grasped such dynamics, was a committed Marxist in his youth (Adelman,
2013), and remained in critical dialogue with Marxism. Hirschman developed the idea
that development would be accelerated through investment (either private or public) in
projects and industries with strong forward or backward linkages because of the induced
investment effect or the ‘push factors’ which come from the (p. 185) product side, arguing
that it is not simply incomes and demand that govern investment (Hirschman, 1958,
1977). We noted in section 7.2.1 the importance of growth pulling backward linkages:
‘backward linkages lead to new investment in input-supplying facilities and forward link
ages to investment in output-using facilities’ (Hirschman, 1977: 72). To these production
linkages he added consumption and fiscal linkages and what he called inside and outside
linkages.
Hirschman’s conception of linkages is worth discussing further as it demonstrates both

strengths and weaknesses in structuralist analysis. Hirschman points to how different
‘constellations’ of linkages can profoundly shape a specific pattern of economic develop
ment, be it in terms of the balance between the different types of linkages, the speed and
strength of linkage development, the sequencing of the development of linkages, or how
‘technological gaps’ can hinder the development of production linkages. For example, he
argues that strong fiscal linkages in the context of a lack of productive investment (or
weak production linkages) can lead to a growth in bureaucracy, wasteful spending, such
as on armaments, and/or a surge in imported consumer goods with concomitant impact
Page 7 of 29
on the balance of payments. In another context, customs duties on imported manufac

tures used in staple exports may seem desirable, but this is self-defeating if funds are
simply used for infrastructure that only further aids staple exports.
Hirschman’s insights continue to be of contemporary relevance for industrial policy and

development more broadly. Indeed, he argues that his approach could be seen as an in
stance of ‘micro-Marxism’ (1977: 92), an attempt to show how ‘the shape of economic de
velopment, including its social and political components, can be traced to the specific
economic activities a country undertakes’ (1977: 92). Countries will have different experi
ences according to ‘different linkage constellations’ (1977: 73). Industrialization can be
viewed as a ‘changing structure of linkages’ (Fine and Rustomjee, 1996: 39) with atten
dant implications for industrial policy.
Yet, from a radical perspective, there are important limits to Hirschman’s conceptualiza
tion, specifically a lack of emphasis on class relations and dynamics in understanding in
dustrialization processes. Fine and Rustomjee (1996), in their account of the industrial
ization of South Africa, provide both an account of industrialization in a particular place,
and a methodology grounded in Marxism for examining particular processes of industrial
ization. They argue that Hirschman tends, not without contradictions, largely to suggest
that a potential linkage will call forth the corresponding agency necessary to bring it
about, be it public or private. They argue instead that which classes forge linkages, and
how particular class interests are brought to bear in these processes, is critical to under
standing how industrialization develops (see also Sender and Smith, 1986). And so, in ac
counting for development in South Africa, Fine and Rustomjee emphasize linking class
and economic structure together with the role played by the state and industrial policy.
The result of this analysis is Fine and Rustomjee’s idea of South Africa being dominated
by a (changing) ‘minerals‒energy complex’, which is an analysis of industrial develop
ment that empirically identifies input‒output linkages between sectors. This analysis ties
these patterns of linkage to the (p. 186) different dimensions of interdependencies be
tween fractions of capital, industrial sectors, and the state. Their analysis involves eco
nomic and political scrutiny (including of the state) through an emphasis on evolving
class relations and conflicts and how these are reflected in patterns of accumulation and
economic and social reproduction over time.
7.3.2 Industrial Policy: The Analytic and the Prescriptive
Marxist industrial policy can be seen to operate at two levels: the analytic and the pre
scriptive. We discuss each in turn.
The analytic can be said to be the analysis of ‘actually existing’ capitalism, both the his
toric centres of accumulation and empire, and the late developers. This entails a critique
of mainstream accounts of capitalism, with its emphasis on self-regulating, dynamic, mar
ket-based processes. There is, therefore, considerable overlap with the perspectives ad
vanced by the ‘developmental state’ (DS) approach, which has focused on the reality of
state intervention to bring about rapid rates of capital accumulation and structural
Page 8 of 29
change in East Asia (see, for instance, Amsden, 1989; Chang, 2003; Wade, 1990), with
South Korean macroeconomic policy, for example, understood as ‘investment
management’ (Chang, 1993: 139). This approach emerged out of a critique of neo-liberal
ism and, in contrast to the sector neutrality of neo-liberal theories of economic growth,
the DS approach recognizes the specific role of manufacturing in economic dynamism
and growth. This work, with its emphasis on state intervention in, and state guidance of,
late industrial development, has provided rich insights from history (see, for example,
Chang, 2010). Following Gerschenkron (1962), ‘late’ development (late in the context of
an already developed world market), is seen as requiring more effort by the state, not
less. This brings forth a variety of institutional innovations, not least in the financing of
industrialization. This literature both critiques the mainstream and discusses important
aspects of industrial development, such as strategic mixing of incentivizing and disciplin
ing capital (Amsden, 1989, 2001; Di Maio, 2013); ‘embedded autonomy’ (Evans, 1995);
and the role played by political settlements and political economy more broadly in pat
terns of economic development, including, for example, corruption (Khan, 2018; Behuria,
Buur, and Gray, 2017).
There are also important Marxist criticisms of the DS approach (Ashman, Fine, and New
man, 2010; and see the essays collected in Fine, Saraswati, and Tavasci, 2013). Utilizing a
Marxist approach to the state and accumulation, to the theorization of both state and
market, and the relationship between the two, these perspectives place greater and more
systematic emphasis on issues around labour exploitation and struggle and the frequent
instances of labour repression within the processes of late industrialization.
The major issues can be summarized as follows. First: the tendency towards a technical-
industrial conception of broader capitalist development which downplays (p. 187) the so
cial antagonisms inherent in capitalist development, and which tends to see the state in
neutral terms (Chang, 2009; Selwyn, 2011). Second: a certain teleology, if at times only
vestigial, about development and industrialization and what is necessary to bring it about,
rather than a recognition of the possibilities for regression, marginalization, or of its
skewed, partial, or ‘combined and uneven’ nature (Ashman, 2012).
Third: lack of recognition of capitalism as a world system with distinctive phases of devel
opment that shape and condition the development of its constitutive parts. Specificities of
conditions in a particular period may not apply at another time (Chibber, 2014) and today
the financialization of the world economy and the internationalization of production net
works pose new challenges. The interconnections of the world capitalist system mean
that the development of one affects the prospects for the development of others, with
acute competition limiting the prospects for all to succeed, and endemic crises pointing to
the impermanence of gains. Manufacturing is not specially protected (despite knowledge
gains) from crises, radical restructuring, or ‘spatial fixes’, pointing to the limitations of
nation-centric views and of national(ist) strategies for development (Song, 2013). Fourth,
as already noted: the labour repression of late development and the tendency to neglect
or downplay labour exploitation.
Page 9 of 29
Fifth, and finally: the question of whose standpoint is adopted. Selwyn (2009) argues that
the DS approach generally takes the view of an elite strategy for catch-up, and advocates
policies that may give rise to faster rates of accumulation, but may not necessarily benefit
society as a whole. Burkett and Hart-Landsberg (2003) argue that catch-up views tend to
see labour as passive, merely an instrument of accumulation and growth, that mass move
ments are merely disruptive of accumulation, and that ‘popular forces’ cannot be con
ceivers of, and contributors to, development.
We should note at this point that state intervention is not without contradictions. Analysts
of welfare have argued rightly that the universal provision of welfare has the potential to
reduce the inequality involved in market-based provision, and in socializing the costs of
caring (for children, for the aged, for the sick and disabled), effectively decommodifies
provision and promotes collective over individualized values, and collective over individ
ual or market-based solutions to both social needs and social problems (Esping-Andersen,
1990). The nationalization of industries, similarly, while often forced by economic crisis
and war, has also provided gains in terms of labour conditions and trade union represen
tation and protection from the market.
In a sense, radical industrial policy can be seen as exploiting the contradictions of the
state, by, on behalf of, and (potentially) with the active involvement of, labour, women,
and the poor. This is important, as the same policies can be very different in different con
texts according to the social forces pursuing them, for example, capital controls in Chile
(Soederberg, 2002) compared with South Africa (Alami, 2018).
This leads us to the prescriptive dimensions of radical industrial policy. At the level of the
prescriptive, or of advocacy, we suggest that industrial policy needs yardsticks to assess
its efficacy. We suggest a broad conception of industrial policy designed to meet basic so
cial needs (not to be confused with a basic-needs approach) and to generate (p. 188) em
ployment. A broad-based conception of industrial policy seeks out the connections be
tween industrial and social policy with health provision, mass public transport, and hous
ing, for example, simultaneously meeting basic social needs, generating employment, and
stimulating backward and forward linkages (MERG, 1993). Public health provision re
quires skilled and unskilled labour as well as infrastructure and there are potential con
nections with pharmaceutical policies. Radical industrial policy would also contribute
broadly to a more egalitarian distribution of income and wealth, and to transforming so
cial relations along gender, class, racial, and other dimensions.
Diversification to a more broad-based and therefore equitable economic structure can be

connected with defending labour and socializing production to meet these basic social
needs. In addition, strengthening the redistributive capacities of the state, not least
through taxation, restriction on tax avoidance, and base erosion profit shifting in order to
fund collective rather than private provision of essential services, can offer protections
against the market and commodification, and potentially go beyond the market. But radi
cal industrial policy can only assist progressive change to the extent that it strengthens
Page 10 of 29
labour, civil society, and social movements, and promotes democracy and has a view to
moving towards socializing the economy under democratic control.
7.3.3 Climate Change: The Potential and Necessity of Radical Indus

trial Policy Today
The United Nations warned in 2018 that the world has twelve years to take the action
necessary to keep global warming to a maximum of 1.5°C, beyond which intense heat,
flooding, drought, and poverty will follow for many (Watts, 2018). Others suggest there is
even less time to radically decarbonize the world economy in order to avoid a disastrous
level of global warming. What has become clear is that climate change is moving faster
than previously thought, and its effects will be more rapid and severe. We use a brief dis
cussion of the climate emergency to illustrate some of the broader arguments of the chap
ter.
The roots of the climate crisis lie precisely in the industrial development being discussed
in this volume, and in capital’s dependence upon particular commodities. The CO2
emissions that arise from coal, oil, and gas—what Malm (2016) calls ‘fossil capital’—are
not the only drivers of climate change, but they are by far the largest. Fossil fuels are crit
ical for the production of surplus value across an enormous spectrum of commodities.
Coal, in particular, has been the ‘the fuel transmitting motion to the labour
process’ (Malm, 2016: 325). The climate crisis is an unintended consequence of industrial
development and industrial policy in many parts of the world. Now, industrial policy, and
more besides, is necessary to address it. But as a consequence of the very power rela
tions we have discussed, fossil capital continues to grow (alongside the growing green
economy), and continues to be funded by the big banks (Rainforest Action Network,
2019). Energy (p. 189) is increasingly financialized, with private equity, hedge funds, and
the whole shadow banking sector involved in the speculative trading of fossil capital. This
financialization exists now across every aspect of everyday life, but power and nature
have become increasingly intertwined.
While there is a certain irony in radical industrial policy to address climate change, the
consequences of failure to do so will be severe. Non-linear processes set in train by the
warming of the planet mean that once we reach a certain point, the impact will be beyond
human control and we will face ‘runaway climate change’ (Bendell, 2018: 6). This means
we must accept ‘collapse as inevitable, catastrophe as probable, and extinction as
possible’ (Bendell, 2018: 20). Social collapse—by which Bendell means ‘starvation, de
struction, migration, disease, and war’—is already with us.6
The scale of the problem can lead to despair, as encapsulated in the remark, attributed
both to Jameson and to Žižek, that ‘it’s easier to imagine an end to the world than an end
to capitalism’ (Fisher, 2009). But Bendell argues that recognizing the possibilities of col
lapse, catastrophe, and extinction can also produce something positive: ‘a shedding of
concern for conforming to the status quo, and a new creativity about what to focus on go
Page 11 of 29
ing forward’ (Bendell, 2018: 20). The climate strikes by schoolchildren and the discussion
of the need for a Green New Deal may support such a view.
There is growing awareness of this within industrial policy circles (see Pollin, Chapter 15
in this volume, as well as Altenburg and Assmann, 2017; Altenburg and Rodrik, 2017; Ro
drik, 2014). Yet it is tempting for developing-country governments to implement industri
al policy that, in trying to break out of commodity export dependence, disregards the im
pact of policy on climate change. UNCTAD (2019: 11) shows that some energy export-de
pendent countries have increased their non-commodity exports by adding value to down
stream energy sectors—a ‘good news’ story. Yet this has been achieved by increasing the
share of chemicals in these countries’ exports. Between 1998 and 2017, Trinidad and To
bago increased the share of (petrol-based) chemicals in its exports by nearly 10 per cent,
from 19.4 per cent to 27.7 per cent. Oman increased the share of chemicals in its exports
from 1 per cent to 9.5 per cent during the same period, mainly through fertilizer manu
facture. The UAE, Qatar, and Saudi Arabia have increased production of petroleum and
gas-based products, and Bahrain utilized their rich energy resources and began alumini
um production which is highly energy intensive (UNCTAD, 2019: 11). Instead of (p. 190)
diversifying out of fossil fuels, these policies reinforce both their centrality and their cli
mate impact.
There is insufficient space in the present chapter for a thorough discussion of the policy
measures needed to address climate change beyond pointing to areas of potential. Tack
ling climate change requires nearly all the burning of coal, oil, and gas to be ended. This
cannot be achieved through market-based solutions, innovative entrepreneurial efforts to
drive ‘green growth’, or the kind of mainstream approaches that tend to emphasize incen
tivizing firm and individual consumer behaviour (Klein, 2014). It requires radical state in
tervention, the creation of millions of public-sector climate jobs, and the extension and
deepening of economic democracy and planning. The Green New Deal under discussion
in the United States and elsewhere, though not without its limitations, proposes to simul
taneously tackle climate change and inequality. This is intended to be achieved through
economic planning and industrial policy measures to bring about mobilization for the en
vironment, in a way that resources were previously mobilized for war. For many advo
cates of a Green New Deal, planning and democracy need to go together, to encompass
labour and trade union involvement, while others advocate increased ‘energy democracy’
to bring about the necessary democratization of energy policy.
Meeting basic needs in a green way, reorienting priorities towards local production to
meet local needs—including more local food production and so less packaging and trans
port—would result in a degree of ‘de-globalization’ of the world economy. ‘Climate jobs’,
which seek to minimize greenhouse gas emissions and maximize employment, have been
advocated to address simultaneously the crises of unemployment and climate change.
Such climate jobs could contribute to meeting the basic needs of communities in an equi
table and sustainable way, with linkages focusing on jobs created in renewable energy
(especially wind and solar power), retrofitting of buildings, new construction methods, ex
panded public transport (run increasingly on renewable energy), and re-employment or
Page 12 of 29
support for those in coal and other sectors who would lose their jobs. Relating back to the
earlier discussion, it could be argued (adding a Kaldorian twist) that we can contrast the
extraction of energy under diminishing returns (coal, oil, gas) with the harvesting of ener
gy under increasing returns (solar, wind power).
In this section, we have emphasized the distinctiveness of a radical approach to industrial

development. Marx’s holistic account of capitalist industrial development moves beyond
issues revolving around the role of sectors and development, though overlaps exist. We
have argued that radical industrial policy operates at both the analytic and prescriptive
levels. Capitalism creates interconnected crises: economic crises such as the recent glob
al financial crisis, crises of war and global competition, crises of food, of health, of re
sources, and sustainability (Tabb, 2008). The climate crisis, we argue, reveals the possi
bility and the necessity for radical industrial policy today (and radical action more gener
ally), though there are limits to what can be achieved within the constraints of the cur
rent social relations of production, which we explore in section 7.5. (p. 191)
7.4 Experiences of ‘Radical’ Industrial Policies

This section discusses the ‘radical’ content of industrial policies that, to different extents,
have been informed by a desire to transform the organization of production and capital‒
labour relations. While by no means comprehensive, examples have been selected to illus
trate key features of, and challenges to, ‘radical’ industrial policy. In particular, we high
light the ways in which tensions between economic imperatives for rapid industrial devel
opment on the one hand, and the balance of social forces and the development of partici
patory democracy as they relate to the state on the other, have been dealt with. These his
torical experiences have been loosely categorized here as statist approaches that are cen
tred on state planning and national development, labour-centred co-operative approaches
that focus on workers’ democracy in the local organization of production, and examples of
industrial policy under participatory planning systems. There are very broad categoriza
tions to organize our discussion, with considerable diversity within each as well as some
commonality among these loose categories. For example, we discuss the Soviet Union and
East Asia together, despite not only the fundamental differences between their industrial
policies but also the broader political and economic characteristics of these models, and
we discuss Cuba and Yugoslavia as examples of industrial policy under participatory plan
ning systems. There are some relevant commonalities between these approaches (Cuba’s
in particular) and that of the Soviet Union that we discuss in a different category.
7.4.1 Statist Models of Industrialization
Statist models of industrialization include the—very different—models of ‘top-down’ in

dustrial development in the Soviet Union and the earlier developmental states of East
Asia, as well as more recent examples from China and Vietnam. Many technical aspects of
radical industrial policy discussed above were present in statist industrialization experi
ences. Economists such as Kritsman, who worked in the Soviet State Planning Commis
Page 13 of 29
sion during the 1920s, were early pioneers of the ideas of iteration and linkage as a way
of understanding sectoral interrelationships and the growth-pulling potential of building
industries with significant backward linkages. Leontief, who later developed the input‒
output production model, worked in the Soviet planning establishment in the early 1920s.
Stalin’s priority sectors, according to later assessment by Chenery and Watanabe (1958),
appeared to be those with maximum linkage effects, and there was considerable success
in developing domestic linkages (backward linkages in particular) and the minimization of
bottlenecks through highly effective mobilization of resources and high rates of fixed cap
ital investment.
Statist models prioritized the development of productive forces over the develop
(p. 192)
ment of democratic participation by labour. In different ways in the Soviet Union and the
East Asian developmental states, the focus was on state‒capital relations and the need
for the state to dominate and/or command capital. In the Soviet Union, this took the form
of direct state ownership over the means of production and the state’s ability to command
and deploy surpluses in the economy. In East Asia, the state‒capital relation took the
form of the role of the state in regulating, directing, and disciplining private business.
In line with the radical ideas on the role of manufacturing in economic development we
have already discussed, development under Stalin was, ‘identified with industrialization,
industrialization would be impossible without capital-intensive techniques, which meant
high savings/investment ratios’ (Dyker, 1985: 3). In the absence of foreign credit, Stalin
saw it as necessary to ‘pump over’ surpluses from the peasantry to fuel accumulation in
the modern sector. This reflected a crude version of Preobrazhensky’s theory of ‘primitive
socialist accumulation’ and entailed a campaign of collectivization (Dyker, 1985: 1). Indus
trialization was directed, top down, according to heavy centralization and the command
principle, via a series of five-year plans that prioritized the development of heavy industry
and the de-prioritization of light labour-intensive industry that had been the focus of earli
er Soviet industrialization strategy.
While the East Asian model can also be characterized as statist, this was under capitalism
with strong developmental states that pushed a strong industrialization agenda. The
strength of these states, combined with a high degree of bureaucratic autonomy, were
seen as key to the success of state-directed investment to promote the concentration of
capital and rising productivity and competitiveness via incentive structures and discipli
nary mechanisms. The state protected and supported domestic business, including
through subsidies and through captive domestic markets, yet by disciplining capital en
sured that the rents from this were reinvested in investment, upgrading, and expansion.
The role of the state in pushing firms to compete in export markets was one of the disci
plining mechanisms used to ensure that this reinvestment materialized, facilitating rapid
growth in productivity and an extraordinary record of economic growth sustained over a
long period of time. This transformed not only living standards but also social relations in
East Asian countries.
Page 14 of 29
One condition for state domination that has been identified by DS theorists has been that
of ‘weak labour’ (Amsden, 1989: 147). Authoritarianism and the repression of labour has
been an enduring feature of statist approaches to industrial development. It has been ar
gued that the success of catch-up industrialization in East Asia is associated in part with
the isolation and suppression of workers and social movements and the political geogra
phy established by the Cold War and the US foreign aid inflows that this engendered
(Chang, 2013).
In the Soviet Union, workers had limited scope for democratic participation either politi
cally or economically. Furthermore, mass collectivization and the shift in priority from
light to heavy industry under Stalin destroyed the (unofficial and voluntary) co-operative
organization of production, based upon principles of democratic (p. 193) participation that
had been emerging under Lenin’s leadership. Lenin had a cautious attitude towards the
establishment of communes and equalization and collective consumption in advance of
developments in production and the creation of abundance. Rather than supporting de
velopment of the communes towards wider social provisioning and consumption, the pri
oritizing of rapid industrial development entailed the expansion of agricultural surplus by
intensification of production and downward pressure on consumption.
Successful industrial development did raise average living standards in the Soviet Union
and East Asia, but this success was in the context of very limited scope for workers’ de
mocratic participation. This raises difficult questions around the compatibility of statist
models of industrialization with vibrant participatory democracies, especially in the twen
ty-first century.
7.4.2 Labour-centred Industrial Development: Worker Co-operatives
The neo-liberal assault on labour organization and the subsequent decline in the structur
al power of labour vis-à-vis capital manifest in capital’s rising share in the functional dis
tribution of income across both the Global North and South, has inspired greater atten
tion to alternative economic organization conducive to more equitable and sustainable
distribution in the context of austerity. Industrial policy at the local level, based upon citi
zen cooperation and shared wealth, has sprung up in the wake of austerity in, for exam
ple, Cleveland, United States and Preston, United Kingdom, where new models of munici
pal socialism have been developed that draw from the experiences of successful co-opera
tives—in particular the Mondragon co-operatives of Spain’s Basque region. These local
approaches stress the role of development in meeting human needs as well as the scope
this offers for green transitioning.
Participatory models of industrial organization have a long history and have co-existed
within a wider context of capitalist relations of production. Studies of participatory forms
of economic organization have variously identified the economic and social benefits of
workplace democracy, which include: improved productivity; greater contribution to so
cial development than purely private enterprise; improved conditions of work; improved
health outcomes associated with better work‒life balance and greater autonomy; a more
Page 15 of 29
equitable distribution between profit and wages/managers and workers; and greater re
silience (Foley and Polanyi, 2006; Pencavel, 2013; NEF, 2018). Yet, co-operative owner
ship of businesses is marginal compared with capital ownership. Where co-operatives
have endured, such as the Mondragon Group, they have undergone significant restructur
ing that has included a gradual degeneration from democratic forms to capitalist forms of
organization in the face of globalization and capitalist competition. This suggests that, in
the absence of supportive policies, in (p. 194) particular those that redress the power im
balance between capital and labour in society, the expansion of workers’ participation in
production will be limited.
The Mondragon complex, which consists of a network of linked productive operations,

was founded in 1956 on three institutional bases. First, a system for technical education.
Second, the League of Education and Culture, which played a central role in linking the
education system to co-operative firms and in the management of any legal and political
issues with local and national government. And third, the credit union the Caja Laboral
Popular, which was founded to support the expansion of industrial co-operatives and thus
operated on the principle of long-term productive investment and the reinvestment of
profits (Errasti et al., 2017). Strong input‒output linkages, together with a multi-layered
institutional structure that amassed workers’ pension savings and surpluses for long-term
productive investment, and facilitated education, research, and design, ensured that the
dynamics of cumulative causation were not hampered by limited access to capital or tech
nical expertise.
Since the 1980s, rising pressures associated with globalization and the increasingly com
petitive environment that it engendered, have led to major changes in firms’ management
structures. The complex was transformed into a uniform corporate identity, the Mondrag
on Co-operative Corporation, with a centralized management structure and more ‘ratio
nalized’ organization of constitutive elements that involved several mergers as well as the
opting out of several co-operatives (Clamp, 2003). Rationalization and corporatization of
the co-operative complex had ensured its survival and growth in an increasingly intensive
and internationally competitive environment, but at the expense of the degree of worker
participation in management and decision-making. The growth of Fagor Electrodomésti
cos, a leading co-operative in Mondragon, for example, entailed a process of acquisition
of capital-owned local and foreign firms, and the group ultimately collapsed in 2013.
These challenges raise questions around the viability of sustaining and expanding co-op
erative forms of production, and maintaining democratic forms of decision-making within
them, within wider national and international economies that are market based.
Examples of enduring workers’ co-operatives in industrial production can also be found in

Kolkata, India, where such groups were formed in the 1970s and 1980s in shipbuilding,
aluminium cables and conductors, hosiery, printing works, and textiles through workers’
takeover of industries in apparent decline (Bhowmik and Sarker, 2002). The enduring de
mocratic participation of workers in these co-operatives can be attributed to the lead tak
en by a strong trade union movement in their formation. The union movement could both
draw upon the support of wider membership for workers’ co-operatives and, to an extent,
Page 16 of 29
negotiate with government. The high level of internal democracy and collectivity has also
been credited for their survival in light of constrained access to finance and markets
(Bhowmik and Sarker, 2002). These co-operatives survived and maintained employment
but did not flourish in the competitive sense.
The existential tensions between market competition and the preservation of the egalitar
ian and participatory ethos of co-operative models of production under (p. 195) prevailing
capitalist social relations are widely acknowledged (Gunn, 2006). At the local level, gov
ernments have begun to experiment with forms of ‘municipal socialism’ in a few localities
in the United States and the United Kingdom, with policies that support local enterprise
and workers’ co-operatives.
The extent to which such programmes can contribute to the wider sustainable structural
transformation of production will depend upon the technical and political content of poli
cies, among other factors. Radical industrial policies require a political economy of labour
that is generated by collective action. The promotion of co-operatives therefore needs to
go hand in hand with policies that maximize workers’ ability to generate and deploy their
agential power. Challenges are also posed by competitive and other pressures exerted by
the wider market environments, beyond the local levels, within which these projects oper
ate. Rather than the imperative to generate additional capital through the exploitation of
workers, a labour-centred development approach sees the reproduction of labour as the
main objective, rather than the subsumption of labour to the objectives of capital.
7.4.3 Models of Industrial Policy with Participatory Planning
In this discussion, we present examples of industrial policy that have engaged with the
contradictions and tensions between economic development of productive forces and so
cial development in terms of the extension and deepening of democratic participation in
economy and society. Some socialist states outside the Soviet Union, including Cuba from
1959 and Yugoslavia after 1948, pursued radical industrial policies that can be deemed
‘developmental’. While recognizing the imperative for industrial development, these
states did not narrowly prioritize development of the forces of production.
In 1948, Yugoslavia split from Stalin and abandoned the Soviet centralized model of in
dustrial development in favour of a form of central planning that aimed at greater re
liance on the domestic and world market and workers’ control over enterprise (Estrin,
1991). Between 1952 and 1974, a series of reforms replaced vertical command planning
with ‘direct horizontal relations between more autonomous enterprise through a regulat
ed market’—a model that became known as market socialism (Estrin, 1991: 188). In con
trast to industrial development in the Soviet Union, institutional structures were set up to
promote the self-management of enterprise by workers with high levels of participation
by workers in decision-making, particularly over questions of welfare, employment, and
pay, while managers retained much of the decision-making power over investment (Es
trin, 1991).
Page 17 of 29
Further reforms between 1974 and 1988 sought to extend democratic participation by
workers beyond decision-making at the level of the enterprise. This involved the breaking
up of enterprises into their constituent establishments or workshops, called Basic Organi
zations of Associate Labor (BOALs), each with its own apparatus of self-management that
replaced the enterprise in law. Individual BOALs then worked (p. 196) collaboratively
based upon mutual agreements with the effect of replacing competition (between enter
prises) with wider cooperation in the system. Further, the devolution of state power to the
level of the republics represented a shift to a federal political structure that promoted the
involvement of all interested parties in the process of policymaking and implementation
(Estrin, 1991).
In the case of Cuba, the development process has been characterized by ‘the need to in
dustrialize and the imperative of social participation in the development process’ (Cole,
2002: 40). Even though these objectives were at times in conflict, their resolution marked
the evolution of socialist development in Cuba, which has entailed the deepening of par
ticipation in the organization of society. The ‘dialectic between the political and the eco
nomic has been the central dynamic of policy and change’ in Cuba (García Brigos quoted
by Cole, 2002: 49).
In this way, Cuban development strategies developed through various iterations reflect
ing these tensions. A period of ‘idealistic spontaneity’ (1959‒63) was followed by ‘central
ized pragmatism’ (1963‒70) and Soviet-style central planning (1970‒86), reflecting eco
nomic imperatives that were not met in the earlier phases. Central planning saw subse
quent adaptations to promote popular participation that had been marginalized by Soviet-
type planning focused on accelerated accumulation. The Rectification Campaign (1986‒
90) saw the culmination of political sentiments in the building of institutional structures
for greater citizen participation. While industrial development has been linked to social
provisioning by leveraging public procurement, notably in healthcare (Tancer, 1995), cap
ital accumulation remained centred on the agro-industrial complex and commodity ex
ports.
With the collapse of the Soviet Union, Cuba opened a window to the world economy, al
lowing foreign investment as a source of capital, resources, expertise, and markets. In
dustrial strategy focused on biotechnology, pharmaceutical, and medical equipment in
dustries that had developed to a level that could compete with developed countries. De
velopment of the Cuban biotechnology industry, for example, reflects elements of radical
industrial policy outlined above, in the linking of industrial development with public pro
visioning and also in the organization of production units strategically linked to a multi-in
stitutional system that included universities and research institutions fostering collabora
tion in innovation (Cárdenas, 2009). In addition, there was a shift from large-scale
monocrop industrial agriculture that had been heavily dependent on imported inputs to
small-scale, organic, or semi-organic agriculture organized via co-operatives that boosted
yields and food independence (Cole, 2002). The period after 1991 also saw the deepening
of forms of democratic participation and involvement in decision-making (Cole, 2002). In
this way, economic liberalization of the early post-Soviet era and the institution of greater
Page 18 of 29
economic specialization provided the opportunity for further socialist development via
deeper participation (Cole, 2002).
The aim of this cursory discussion of radical industrial policies in Yugoslavia and Cuba is
not to paint an unduly rosy or linear picture of economic and political development, or to
downplay the external and internal challenges these nations have (p. 197) faced. Mistakes
were made, and these development models have their own serious limitations and con
straints. Successive policy stages reflect an iterative approach that sought to calibrate
the dynamic relationship between economic and social imperatives and attempts at deep
ening democracy.
Overall, the experiences of industrialization processes and the range of models discussed
in this section help to illuminate the radical content of industrial policies. There are both
economic and political dimensions to radical industrial policies. Radical policies are class-
conscious and do not focus on the narrow interests of elites. Rather, they both stem from
and are directed at a class project that places labour at its centre, not only in the distribu
tion of gains from development, but through direct economic and political participation of
workers. Radical industrial policy, then, entails both the development of democracy and
technical innovations in the organization of production, the precise form of which itself
evolves and is subject to ongoing contestation.
7.5 Conclusion
In our survey, evaluation, and discussion of radical industrial policy, we have examined
the differences between mainstream and heterodox approaches to economic and industri
al development. The focus of the debate between the mainstream and the heterodox has
been around the question of whether or not growth is sector neutral and, to a lesser ex
tent, activity neutral. Heterodox approaches have argued that the sectoral basis of output
and employment will affect the pattern of growth and development. We looked in particu
lar at the similarities and the differences between Kaldorian/structuralist approaches and
Marxist ones. It is possible to identify points of common ground, and indeed Marx can be
seen to have provided analytical antecedents for later approaches in his analysis of nine
teenth-century industrial capitalism. Structuralists have been primarily concerned with
the balance-of-payments constraint on growth (reflecting the structure of production) and
the deteriorating terms of trade for developing economies (the nature of the international
system). Combined, these hold back development. A radical perspective recognizes both
the value and limits of such arguments.
We see important differences between Kaldorian/structuralist approaches and Marxist ap

proaches. Marx’s concern is not with sectors but with value and the overall circuit of cap
ital. This view of a circuit of capital, integrating the different forms which capital takes, is
rooted at a more abstract level in a broader framework for understanding capitalism as a
mode of production that sees accumulation as founded upon exploitative class relations
between labour and capital, and where competition between rival units of capital forces
capital both to innovate and to increase the production of surplus value. Marx’s problem
Page 19 of 29
atic, for all his insights into modern machine production, is concerned with the capitalist
basis of production, not primarily with industrial production. We have argued, therefore,
that radical industrial policy foregrounds class (p. 198) and capitalism, not sectors, and in
tegrates a distinctive conception of the state: the state as central to accumulation but
which can also, in particular circumstances, be leveraged to aid radical transformation if
connected to broader social forces and movements.
Connected to this, we have argued that radical industrial policy operates at two levels,
the analytical and the prescriptive. The analytical level refers to the understanding of
capitalist industrial development in particular contexts, integrating the role of the state
and diverging class interests, including patterns of linkages and the nature of industrial
policy interventions. Prescriptively, in radical industrial policy, the contradictions of the
state are exploited and there is a concern with meeting basic social needs, maximizing
employment-generating linkages, strengthening labour and democracy, and, urgently,
mitigating climate change. These are among the yardsticks by which to assess industrial
policy from a radical perspective.
Finally, we surveyed briefly some experiences of industrial policy, loosely dividing them
between statist and labour-centred. The economic imperatives for national strategies of
development have frequently entailed labour subordination to capital, as the experience
of the former Soviet Union attests: the construction of a centralized command economy
prioritizing ‘modernization’ and rapid accumulation, an approach that was replicated in
the five-year plans inspired by the Soviet Union in a number of developmental states.
Discussing labour-centred industrial policy based on co-operative models of organization

raises issues seemingly out of fashion in industrial policy circles and literature: the eco
nomic and social benefit of workplace democracy in terms of improved conditions of
work; better work‒life balance; greater autonomy; a more equitable distribution between
profit and wages; and hence a reduction in inequality. It is difficult for workers’ co-opera
tives to survive without supportive policies from the state. Yet if we are to avoid/reverse
the ‘race to the bottom’ and competition between workers, there are models from which
we could learn.
What, then, is the ‘value added’ of radical industrial policy? First, we argued that it
breaks out of the technical. It is not simply about economic growth and catch-up by devel
oping economies, nor is it simply about ‘sector fundamentalism’; indeed there are pro
foundly different political projects embodied in Marxist vis-à-vis Kaldorian and structural
ist approaches to economic development and industrial policy, which stem from their di
verging assumptions.
Second, radical industrial policy has a distinctive analysis rooted in understanding the dy
namics of capitalism: accumulation, exploitation, and the state. Class-based processes de
termine the nature and form of industrial development. Class is made central, rather than
sectors and activities, and both class and the state are critical to understanding specific
Page 20 of 29
‘constellations’ of linkages. These factors are brought to bear in critiques of the develop
mental state approaches.
Third, radical industrial puts emphasis upon both economic and political dimensions, and
the avoidance of economic reductionism (a partner of technical conceptions). We stressed
this particularly in the necessity of labour’s active participation, and the radical (p. 199)
extension of democracy necessary for greater labour participation in defining the priori
ties and methods of production, and in meeting the basic needs of the majority.
Finally, standpoint. Radical industrial policy does not take the standpoint of elites, but is a
class-based project with labour at its centre. In our discussion of radical industrial policy
and the climate crisis, we argued that only a radical extension of state involvement and
democracy can bring about the scale and speed of intervention necessary to address this
impending catastrophe.
Given this analysis, there are necessarily limits to what can be achieved by industrial poli
cy. Gains for labour are frequently temporary, ‘islands of socialism’ do not survive long in
seas of capitalism and genuine participatory projects remain utterly marginal to global
accumulation. The scale of the climate crisis can induce paralysis, yet it demands imagin
ing and striving for a more habitable and just economy, including through public trans
port, democratic control over energy and water supplies, local production to meet local
needs, and the opening of borders to migrants displaced by climate impacts (Klein, 2014).
Finally, what are the prospects for radical industrial policy today? While there is growing
mainstream acceptance of the need for industrial policy, there are widely differing con
ceptions of what the appropriate goals and tools of industrial policy should be. The finan
cialization of the global economy has not been the major focus of this chapter, yet this
conditions the prospects for industrial policy today, be it radical or otherwise. The present
phase of capitalism has witnessed the unprecedented (albeit uneven) internationalization
of productive capital, and its restructuring as driven by finance. ‘Downsize and
distribute’ (Lazonick and O’Sullivan, 2000) has radically altered the structure of many
economies (along with the rise of China) and radically shifted the balance of power to
wards capital. The neo-liberal state has been central to this restructuring, and ‘finance-in
duced’ policy orthodoxy has meant price stability, and high real interest rates continue to
be given priority (Fine et al., 1999). Policies which ‘risk’ higher inflation or higher taxes
are fiercely resisted, as are interventionist industrial policies to restructure capital, not
least to ward off the possibility that similar state intervention may be extended to the fi
nancial sector. Radical industrial policy, as we have argued, places power at its centre,
and addressing the power of the financial sector is an absolute necessity.
References
Adelman, Jeremy (2013) Worldly Philosopher: The Odyssey of Albert O. Hirschman.
Page 21 of 29
Alami, Ilias (2018) ‘Capital Accumulation and Capital Controls in South Africa: A Class
Perspective’, Review of African Political Economy 45(156): 223–49.
Altenburg, Tilman and Claudia Assmann (eds) (2017) Green Industrial Policy: Concepts,
Policies, Country Experiences. Geneva, Bonn: UN Environment, German Development
(p. 200) Institute. Available at https://www.un-page.org/files/public/
green_industrial_policy_book_aw_web.pdf.
Altenburg, Tilman and Dani Rodrik (2017) ‘Green Industrial Policy: Accelerating Structur
al Change towards Wealthy Green Economies’, in Tilman Altenburg and Claudia Assmann
(eds) Green Industrial Policy: Concepts, Policies, Country Experiences. Geneva, Bonn: UN
Environment, German Development Institute, pp. 2–20.
Amsden, Alice (1989) Asia’s Next Giant: South Korea and Late Industrialization. New
York: Oxford University Press.
Amsden, Alice (2001) The Rise of the Rest: Challenges to the West from Late-Industrializ
ing Economies. Oxford: Oxford University Press.
Andreoni, Antonio and Roberto Scazzieri (2014) ‘Triggers of Change: Structural Trajecto
ries and Production Dynamics’, Cambridge Journal of Economics 38(6): 1391–408.
Ashman, Sam (2012) ‘Combined and Uneven Development’, in Ben Fine and Alfredo
Saad-Filho (eds) The Elgar Companion to Marxist Economics. Aldershot: Edward Elgar,
pp. 60–5.
Ashman, Sam, Ben Fine, and Susan Newman (2010) ‘The Developmental State and Post-
Liberation South Africa’, in Neeta Misra-Dexter and Judith February (eds) Testing Democ
racy: Which Way Is South Africa Going? Cape Town: ABC Press for Institute for a Democ
ratic South Africa, pp. 23–45.
Behuria, Pritish, Lars Buur, and Hazel Gray (2017) ‘Studying Political Settlements in
Africa’, African Affairs 116(464): 508–25.
Bendell, Jem (2018) ‘Deep Adaptation: A Map for Navigating Climate Tragedy’, IFLAS Oc
casional Paper No. 2. Available at http://www.lifeworth.com/deepadaptation.pdf.
Berik, Günseli, Yana Rodgers, and Stephanie Seguino (2009) ‘Feminist Economics of
Inequality, Development, and Growth’, Feminist Economics 15(3): 1–33.
Best, Michael (2018) How Growth Really Happens: The Making of Economic Miracles
Bhowmik, Sharit K. and Kanchan Sarker (2002) ‘Worker Cooperatives as Alternative Pro
duction Systems: A Study of Kolkata, India’, Work and Occupations 9(4): 460–82.
Page 22 of 29
Blankenburg, Stephanie, Gabriel Palma, and Fiona Tregenna (2008) ‘Structuralism’, in

Lawrence Blume and Steven Durlauf (eds) The New Palgrave: A Dictionary of Economics.
2nd edition. Basingstoke: Palgrave Macmillan, pp. 69–74.
Braunstein, Elissa (2000) ‘Engendering Foreign Direct Investment: Family Structure, La
bor Markets, and International Capital Mobility’, World Development 28(7): 1157–72.
Brenner, Robert (1985) ‘The Social Basis of Economic Growth’, in John Roemer (ed.) Ana
lytical Marxism. Cambridge: Cambridge University Press, pp. 25–53.
Burkett, Paul and Martin Hart-Landsberg (2003) ‘A Critique of “Catch-up” Theories of De
velopment’, Journal of Contemporary Asia 33(3): 147–71.
Çağatay, Nilüfer and Şule Özler (1995) ‘Feminization of the Labor Force: The Effects of
Long-term Development and Structural Adjustment’, World Development 23(11): 1883–
94.
Cárdenas, Andrés (2009) ‘The Cuban Biotechnology Industry: Innovation and Universal
Health Care’. Cambridge, MA: The Academic-Industry Research Network (theAIRnet).
Chang, Dae-oup (2009) Capitalist Development in Korea: Labour, Capital and the Myth of
the Developmental State. Abingdon: Routledge.
Chang, Dae-oup (2013) ‘Labour and the “Developmental State”: A Critique of the Devel
opmental State Theory of Labour’, in Ben Fine, Daniela Tavasci, and Jyoti Saraswati (eds)
Beyond the Developmental State: Industrial Policy into the Twenty-first Century. London:
Pluto, pp. 85–109. (p. 201)
Chang, Ha-Joon (1993) ‘The Political Economy of Industrial Policy in Korea’, Cambridge
Journal of Economics 17: 131–57.
spective. London: Anthem.
Chang, Ha-Joon (2010) 23 Things They Don’t Tell You about Capitalism. London: Allen
Lane/Penguin.
Chenery, Hollis and Tsunehiko Watanabe (1958) ‘International Comparisons of the Struc
ture of Production’, Econometrica 26(4): 487–521.
Chibber, Vivek (2014) ‘The Developmental State in Retrospect and Prospect: Lessons
from India and South Korea’, in Michelle Williams (ed.) The End of the Developmental
State? Pietermaritzburg: UKZN Press, pp. 30–54.
Clamp, Christina (2003) ‘The Evolution of Management in the Mondragon Cooperatives’,

Mapping Cooperative Studies in the New Millennium, University of Victoria.
Clarke, Simon (ed.) (1991) The State Debate. Conference of Socialist Economists. Bas
ingstoke: Palgrave Macmillan.
Page 23 of 29
Cole, Ken (2002) ‘Cuba: The Process of Socialist Development’, Latin American Perspec
tives 29(3): 40–56.
icy, the Shifting Boundaries of the Industrial, and the Implications for Industrial Hubs’, in
Justin Yifu Lin and Arkebe Oqubay (eds) The Oxford Handbook of Industrial Hubs and
Economic Development. Oxford: Oxford University Press.
Di Maio, Michele (2013) ‘Industrial Policies in Developing Countries: History and Perspec
tives’, in Giovanni Dosi, Mario Cimoli, and Joseph E. Stiglitz (eds) The Political Economy
of Capabilities Accumulation. Oxford: Oxford University Press, pp. 107–43.
Dyker, David (1985) The Future of the Soviet Economic Planning System. London: Rout
ledge.
Elson, Diane and Ruth Pearson (1981) ‘“Nimble Fingers Make Cheap Workers”: An Analy
sis of Women’s Employment in Third World Export Manufacturing’, Feminist Review 7:
87–107.
Errasti, Anjel, Ignacio Bretos, and Aitziber Nunez (2017) ‘The Viability of Cooperatives:
The Fall of the Mondragon Cooperative Fagor’, Review of Radical Political Economics
49(2): 181–97.
Esping-Andersen, Gøsta (1990) The Three Worlds of Welfare Capitalism. Princeton, NJ:
Princeton University Press.
Estrin, Saul (1991) ‘Yugoslavia: The Case of Self-Managing Market Socialism’, Journal of
Economic Perspectives 5(4): 187–94.
Evans, Peter (1995) Embedded Autonomy: States and Industrial Transformation. Prince
ton, NJ: Princeton University Press.
Fine, Ben and Zavareh Rustomjee (1996) The Political Economy of South Africa: From
Minerals–Energy Complex to Industrialization. London: Hurst.
Fine, Ben, Costas Lapavitsas, and Dimitris Milonakis (1999) ‘Addressing the World Econo
my: Two Steps Back’, Capital and Class 23(1): 47–90.
Fine, Ben, Jyoti Saraswati, and Daniela Tavasci (eds) (2013) Beyond the Developmental
State: Industrial Policy into the 21st Century. London: Pluto.
Fisher, Mark (2009) Capitalist Realism: Is There No Alternative? Ropley: Zero Books.
Foley, Janice and Michael Polanyi (2006) ‘Workplace Democracy: Why Bother?’, Economic
and Industrial Democracy 27(1): 173–91. (p. 202)
Gerschenkron, Alexander (1962) Economic Backwardness in Historical Perspective.

Page 24 of 29
Giovannoni, Olivier (2010) ‘Functional Distribution of Income, Inequality, and the Inci
dence of Poverty: Stylized Facts and the Role of Macroeconomic Policy’. University of
Texas Inequality Project Working Paper No. 58.
Gunn, Christopher (2006) ‘Cooperatives and Market Failure: Workers’ Cooperatives and
System Mismatch’, Review of Radical Political Economics 38(3): 345–54.
University Press.
Hirschman, Albert (1977) ‘A Generalized Linkage Approach to Development, with Special

Reference to Staples’, Economic Development and Cultural Change 25: 67–98.
İzdeş, Özge and Fiona Tregenna (2020) ‘Gender, Industrialization, and Industrial Hubs’,
in Justin Yifu Lin and Arkebe Oqubay (eds) The Oxford Handbook of Industrial Hubs and
Kaldor, Nicholas (1966) Causes of the Slow Rate of Economic Growth of the United King
dom: An Inaugural Lecture. Cambridge: Cambridge University Press.
Kaldor, Nicolas (1980) Essays on Economic Stability and Growth. 2nd edition. London:
Duckworth.
Khan, Mushtaq (2018) ‘Political Settlements and the Analysis of Institutions’, African Af
fairs 117(469): 636–55.
Klein, Naomi (2014) This Changes Everything: Capitalism vs. the Climate. London: Allen
Lane.
Kucera, David and Tejani, Sheba (2014) ‘Feminization, Defeminization, and Structural
Change in Manufacturing’, World Development 64(C): 569–82.
Lazonick, William and Mary O’Sullivan (2000) ‘Maximizing Shareholder Value: A New Ide
ology for Corporate Governance’, Economy and Society 29(1): 13–35.
Malm, Andreas (2016) Fossil Capitalism: The Rise of Steam Power and the Roots of Global
Warming. London: Verso.
Marx, Karl (1976 [1867]) Capital: A Critique of Political Economy, Vol. 1. London: Pen
guin.
Marx, Karl (1984 [1894]) Capital: A Critique of Political Economy, Vol. 3. London:
Lawrence & Wishart.
Marx, Karl and Friedrich Engels (1938 [1845/6]) The German Ideology. London:
Lawrence & Wishart.
Page 25 of 29
MERG (1993) ‘Making Democracy Work: A Framework for Macroeconomic Policy in

South Africa’, Centre for Development Studies, University of the Western Cape, South
Africa.
Milberg, William and Deborah Winkler (2009) ‘Financialisation and the Dynamics of Off
shoring in the USA’, Cambridge Journal of Economics 34(2): 275–93.
Neale, Jonathan (2019) ‘Social Collapse and Climate Breakdown’, Ecologist: The Journal
for the Post-industrial Age, 8 May. Available at https://theecologist.org/2019/may/08/
social-collapse-and-climate-breakdown.
NEF (2018) Co-operatives Unleashed: Doubling the Size of the UK’s Co-operative Sector.
London: New Economics Foundation.
versity Press.
Pencavel, John (2013) The Economics of Worker Cooperatives. Northampton, MA: Edward
Elgar. (p. 203)
Prebisch, Raúl (1950) ‘The Economic Development of Latin America and its Principal
Problems’, Economic Bulletin for Latin America, No. 7, United Nations Department of
Economic Affairs, New York.
Rainforest Action Network (2019) ‘Banking on Climate Change: Fossil Fuel Finance Re
port Card 2019’. Available at https://www.ran.org/wp-content/uploads/2019/03/
Banking_on_Climate_Change_2019_vFINAL1.pdf.
Ricoy, Carlos (2003) ‘Marx on Division of Labour, Mechanization and Technical Progress’,
European Journal of the History of Economic Thought 10(1): 47–79.
Rodrik, Dani (2014) ‘Green Industrial Policy’, Oxford Review of Economic Policy 30(3):
469–91.
Rosenberg, Nathan (2011) ‘Was Schumpeter a Marxist?’, Industrial and Corporate

Change 20(4): 1215–22.
Seguino, Stephanie and Elissa Braunstein (2019) ‘The Costs of Exclusion: Gender Job Se
gregation, Structural Change and the Labour Share of Income’, Development and Change
50: 976–1008.
Selwyn, Benjamin (2009) ‘An Historical Materialist Appraisal of Friedrich List and his
Modern Day Followers’, New Political Economy 14(2): 157–80.
Selwyn, Benjamin (2011) ‘Trotsky, Gerschenkron and the Political Economy of Late Capi
talist Development’, Economy and Society 40(3): 421–50.
Sender, John and Sheila Smith (1986) The Development of Capitalism in Africa. London:
Methuen.
Page 26 of 29
Soederberg, Susanne (2002) ‘An Historical Materialist Account of the Chilean Capital
Control: Prototype Policy for Whom?’, Review of International Political Economy 9(3):
490–512.
Song, Hae-Yung (2013) ‘Marxist Critiques of the Developmental State and the Fetishism
of National Development’, Antipode 45(5): 1254–76.
Spencer, David (2009) The Political Economy of Work. Abingdon: Routledge.
Spencer, David (2018) ‘Fear and Hope in an Age of Mass Automation: Debating the Fu
ture of Work’, New Technology, Work and Employment 33(1): 1–12.
Stockhammer, Engelbert, Özlem Onaran, and Stefan Ederer (2008) ‘Functional Income
Distribution and Aggregate Demand in the Euro Area’, Cambridge Journal of Economics
33(1): 139–59.
Tabb, William (2008) ‘Four Crises of the Contemporary World Capitalist System’, Monthly
Review, 1 October. Available at https://monthlyreview.org/2008/10/01/four-crises-of-
the-contemporary-world-capitalist-system/.
Tancer, Robert (1995) ‘The Pharmaceutical Industry in Cuba’, Clinical Therapeutics 17(4):
791–8.
Tregenna, Fiona (2009) ‘Characterising Deindustrialisation: An Analysis of Changes in

Manufacturing Employment and Output Internationally’, Cambridge Journal of Economics
33(3): 433–66.
Tregenna, Fiona (2011) ‘What Does the Services Sector Mean in Marxian Terms?’, Review
of Political Economy 23(2): 281–98.
Tregenna, Fiona (2013) ‘The Specificity of Manufacturing in Marx’s Economic Thought’,

European Journal for the History of Economic Thought 20(4): 603–24.
Tregenna, Fiona (2014) ‘A New Theoretical Analysis of Deindustrialisation’, Cambridge

Journal of Economics 3(6): 1373–90. (p. 204)
Tregenna, Fiona (2016) ‘Deindustrialization and Premature Deindustrialization’, in Jayati

Ghosh, Erik S. Reinert, and Rainer Kattel (eds) Handbook of Alternative Theories of Eco
nomic Development. Cheltenham: Edward Elgar, pp. 710–28.
Tregenna, Fiona (2018) ‘Sectoral Structure and Change: Insights from Marx’, Review of
UNCTAD (2019) State of Commodity Dependence 2019. Geneva: United Nations. Avail
able at https://unctad.org/en/PublicationsLibrary/ditccom2019d1_en.pdf.
Verdoorn, Petrus Johannes (1993 [1949]) ‘On the Factors Determining the Growth of
Labour Productivity’, in Luigi Pasinetti (ed.) Italian Economic Papers, Vol. II. Oxford: Ox
Page 27 of 29
Watts, Jonathan (2018) ‘We Have 12 Years to Limit Climate Change Catastrophe, Warns
UN’. Available at https://www.theguardian.com/environment/2018/oct/08/global-
warming-must-not-exceed-15c-warns-landmark-un-report.
Weeks, John (2011) Capital, Exploitation and Economic Crisis. London: Routledge.
38(152): 527–42.
Notes:
(1) See also, for instance, Berik, Rodgers, and Seguino (2009), Braunstein (2000),
Çağatay and Özler (1995), Elson and Pearson (1981), İzdeş and Tregenna (2020), Kucera
and Tejani (2014), and Seguino and Braunstein (2019).
(2) See Ocampo, Chapter 3, this volume for a discussion of the structuralist approach to
industrial policy.
(3) Also of interest here is Babbage’s view of the ‘law of multiples’ in industrial produc
tion, which relates to increasing returns and the division of labour and to the problem of
indivisibility, and to the sector specificity of this. This has implications for the scaling up
of industrial production, and hence for industrial policy (see Andreoni and Scazzieri,
2014).
(4) See also the country and regional experiences of industrial policy discussed in Parts IV
and V of this volume, which draw attention to the diverse experiences of industrial policy
and industrialization internationally and over time.
(5) For reasons of space we are unable to discuss current debates around post-work and
connected policy measures, such as a universal basic income. For relevant suggestions,
see Spencer (2018).
(6) Neale (2019) argues that the imagery of social collapse brought about by climate
change tends to be that of wandering nomads in a post-apocalyptic world where society
has fallen totally apart. Instead, he argues, it will be nothing of the sort: the elites
presently with power will intensify their rule, and climate collapse is more likely to come
‘with tanks on the streets and the military or the fascists taking power. Those generals
will talk in deep green language. They will speak of degrowth, and the boundaries of
planetary ecology…and they will build a new kind of gross green inequality. And in a
world of ecological freefall, it will take cruelty on an unprecedented scale to keep their in
equality in place.’
Sam Ashman
Page 28 of 29
Sam Ashman is associate professor in the School of Economics at the University of

Johannesburg where she is director of the UJ-IDEP MPhil in industrial policy pro
gramme. Her research focuses on the financialization of the world economy and its
various impacts—on firm behaviour, patterns of development, the shaping of econom
ic and social policy, and the worsening of inequality. She also researches industrial
policy and development issues, including the impact of global value chains, the evolv
ing political economy of South Africa, and the history and philosophy of economics
and political economy.
Susan Newman
Susan Newman is professor of economics at the Open University, United Kingdom.

She has published in areas including finance and industrial policy, the political econo
my of industrial development in South Africa, and financialization and the restructur
ing of production. She is particularly interested in the dynamics of wealth and in
come distribution along global value chains. She is a member of Reteaching Econom
ics and the International Initiative for Promoting Political Economy and is committed
to the promotion of political economy in economics education.
Fiona Tregenna
Fiona Tregenna holds the DST/NRF South African Research Chair in Industrial De
velopment and is also a professor of economics at the University of Johannesburg.
She is appointed to policy bodies including the Competition Tribunal and the Presi
dential Economic Advisory Council, sits on a number of boards, and consults for vari
ous international organizations. She has a PhD in economics from the University of
Cambridge, a master’s degree in economics from the University of Massachusetts,
and earlier degrees from the Universities of the Witwatersrand and Natal (South
Africa). Her primary research interest is in issues of structural change, deindustrial
ization, and industrial development.
Page 29 of 29
Global Value Chains and Regionally Coordinated Industrial Policy: The Case
of ASEAN
Global Value Chains and Regionally Coordinated Indus

trial Policy: The Case of ASEAN
Budi A. Djafar and William Milberg
The well-documented failures of the liberal policy regimes of the period 1980–2007 to
generate sustained economic growth and development have led in the past ten years to a
renewed interest in industrial policy. The re-emergence of industrial policy occurs in a
new world of well-developed GVCs in major industries with integrated regional
economies. This chapter assesses the success of industrial policy in this context by con
sidering the case of the Association of Southeast Asian Nations (ASEAN). After laying out
the case for industrial policy in regionalized GVCs, we find that ASEAN, a regional trade
bloc, has helped to coordinate regional production integration and promoting industry
competitiveness for member states in several strategic industries. We conclude that the
future success of ASEAN industrial policy will hinge on continued growth in Chinese de
mand and the ability of ASEAN countries to build some globally competitive, lead firms in
key industries.
Keywords: regionalization, GVC, ASEAN, industrialization, industrial policy, automotive
8.1 Regional Economic Integration, The Core of

ASEAN Industrial Policy
AFTER decades of efforts to transform Southeast Asia into a resilient regional economy,
the Association of Southeast Asian Nations (ASEAN)—a regional intergovernmental orga
nization comprising ten countries in Southeast Asia—has shifted its priorities towards
deepening its regional participation into global value chains (GVCs).1 For ASEAN, greater
participation into the GVC network should help the region improve its trade and invest
ment nexus critical for firm and industry competitiveness amidst twenty-first-century
globalization. Moreover, greater integration into GVCs might expand production linkages
and export destinations of local firms that may not have been accessible prior to regional
and global integration. Above all, at least (p. 208) among ASEAN policymakers, greater
participation into GVCs may play a key role in realizing ASEAN’s grand vision for creat
Page 1 of 31
of ASEAN
ing a single market and production base known as the ASEAN Economic Community
(AEC) by 2025.
In this chapter, we argue that the recent policy shift towards deepening regional and
global production networks may pose both an opportunity and a challenge for the
region’s future economic and development trajectories. We highlight several key points
regarding ASEAN’s industrial policy in the era of GVCs. First, increasing GVC participa
tion may benefit some sectors more than others. We show that the automotive GVCs in
ASEAN have been a success, especially with the expansion of domestic production and
export growth performance in recent times. We note that the success of automotive GVCs
depends on continued institutional support to implement a more integrative production
process. ASEAN’s regionally coordinated policies, such as the ASEAN Brand-to-Brand
Complementation (BBC) scheme, or when parts and components are traded without pay
ing full import duties, have made South East Asia an attractive region for automotive
manufacturing. ASEAN’s BBC scheme, first introduced in 1988, was a breakthrough in
advancing the integrative process of the automotive industry in the Southeast Asian re
gion. Second, the success story of automotive GVCs in ASEAN hinges on regionally coor
dinated industrial policy and the active role of large multinational firms in creating re
gional and local production linkages. Toyota, a Japanese auto manufacturer, has been di
rectly involved in making Southeast Asia its own backyard for production and assembling
since the 1960s. Toyota’s influence in the region’s automotive production and assembly
continues as it maintains a positive relationship with ASEAN’s political and economic es
tablishments. Third, we underline the importance of institutions in improving local firm
capacities and their competitiveness, human capital, technological capabilities, and ‘in
dustrial upgrading’, or moving up the value chain from lower value-added to higher value-
added economic activities in the era of rising participation in GVCs. Thailand’s initiative
to establish universities and research institutions related to the auto industry greatly con
tributed to its advancement in auto manufacturing. Thailand has been dubbed as the ‘De
troit of South Asia’ for rapidly transforming its automotive industry since recovering from
the 1997 Asian financial crisis.
This chapter starts with an overview of ASEAN as a regional institution that collectively
promotes economic growth policies among its member states. The chapter will then trace
the historical evolution of ASEAN’s economic policies since the 1960s. We discuss the
region’s strategic shift from the conventional approach of reducing or eliminating trade
barriers and creating a conducive environment for foreign investors to the deepening of
regional and global production network participation.2 For ASEAN policymakers, regional
integration through GVCs (p. 209) expedites a country’s participation in global markets
and production and eases the process of economic development. Regional geopolitics is
also worth noting here; ASEAN industrial policy in recent years serves to ensure recogni
tion that the region continues to be dynamic and competitive. With a growing number of
free trade agreements (FTAs), including from those regional trade agreements to which
some ASEAN member states are a party, like the Comprehensive and Progressive Agree
ment for Trans-Pacific Partnership (CPTPP), ASEAN has to sustain its competitiveness.
Page 2 of 31
of ASEAN
Second, the spectacular rise of China, and to some extent India, has changed the political
and economic relationships of ASEAN member states and their counterparts.3
Deepening ASEAN’s GVC participation also means that there is more space for local par
ticipation, particularly by small and medium-sized enterprises (SMEs). Improving local
participation and the redistribution of economic gains from regional and global value
chains emerged as a focal point in establishing the ASEAN Economic Community 2020, as
outlined in the ASEAN Blueprint 2020. While this policy may be good for redistributing
gains to local actors, it falls short in guaranteeing the industrial upgrading that often oc
curs in firms with large capital investments. To this extent, ASEAN has yet to produce
lead firms with international presence.
Technological advancements in production and logistics have made it easier to establish

inter- and intra-industry cooperation. While participation in GVCs opens up greater ac
cess to markets and technology, there remains a doubt whether technological spillovers
are immediate or if they occur at all. The gains, such as a transfer of knowledge, from
participation in GVCs have been mixed.
Moreover, the chapter argues that while ASEAN industrial policy in the era of GVCs cen
tred on improving local participation in recent years, more needs to be done to address
the competitiveness of large firms to capture higher economic returns (i.e. profits) and
technological advancements. What we observed from ASEAN’s current policy initiative in
the era of GVCs is the lack of a systematic programme and streamlining policy for indus
trial upgrading. ASEAN’s industrial policy in the era of GVC is skewed towards increasing
participation of the bottom and middle tiers of the value chain. Thus, leaving large local
firms to compete directly with multinational firms.
We conclude that ASEAN industrial policy in the era of GVCs should go beyond the objec
tives of deepening regional and global integration. ASEAN must find new ways to create
value-added, whether that involves a large number of (p. 210) vertically linked activities or
only just a few in the value chain, and establish centres for innovation. Learning from
past experience, ASEAN should reconsider introducing a new form of regionally coordi
nated policies, such as the ASEAN Industrial Joint Venture (AIJV) in 1983 and later the
ASEAN Industrial Cooperation (AICO) in 1996, which was successful in expanding the re
gional division of labour and scale economies of the auto industry. Finally, the chapter
concludes by suggesting that the future success of ASEAN industrial policy hinges on the
continued growth of Chinese demand (particularly in light of the US–China trade war)
and the ability for ASEAN to build some globally competitive, lead firms in key industries.
8.2 The Deepening of GVC Integration: ASEAN,

by Numbers
Southeast Asia has arguably emerged as one of the most dynamic regions in the world in
the last decades. Over the years, ASEAN economies recorded positive real GDP growth
Page 3 of 31
of ASEAN
rates of 5.5 per cent in 2000 to 6.9 per cent in 2010 (IMF, 2019).4 The region’s share of
world GDP almost doubled from 3.3 per cent in 1967 to 6.2 per cent in 2016. ASEAN’s
value-added exports also show signs of improvement. ASEAN’s domestic value added im
proved from 62 per cent in 2013 to 64 per cent in 2018. Domestic value added can be in
terpreted as positive improvement in a country’s domestic capacity to produce goods and
services. Therefore, the more a country creates domestic value added, the more income is
generated. Meanwhile, ASEAN’s foreign value-added exports slowed from 38 per cent in
2013 to 36 per cent in 2018. Contrary to domestic value added, foreign value added is of
ten used to show a country’s exposure to foreign inputs, including raw material re
sources, parts and components, and services. Foreign value added differs by industry and
country. Lower foreign value added may be a positive indication of improvement in do
mestic production.
While ASEAN has had an impressive economic record over the years, the pace of econom
ic growth and export performance varies from country to country. In 2018, for instance,
the domestic value added of exports as a share of GDP for Malaysia and Singapore is
roughly 63 per cent and 43 per cent, respectively; Brunei Darussalam and Thailand are
both 37 per cent; and the rest of the ASEAN economies’ domestic value added of exports
as a share of GDP are below the regional average of 33 (p. 211) per cent.5 Within ASEAN,
the share of the total foreign value-added exports is highest in manufacturing (45 per
cent) compared to primary (7 per cent) and services sectors (24 per cent) in 2015.6
Foreign value-added in manufacturing is driven by motor vehicles and other transport
equipment and electronics sectors with roughly 59 per cent and 50 per cent, respectively,
of the share of the foreign value added in manufacturing.
ASEAN’s GVC integration is deeper and potentially more rapid than other regions. Over
60 per cent of ASEAN member states’ gross exports are linked to GVCs, measured in
terms of the sum of the foreign value added and the indirect domestic value added, or the
domestic value added incorporated in other countries’ exports (Yamaguchi, 2018; ASEAN-
Japan Center, 2019).7 In 2013, about 65 per cent of ASEAN’s gross exports were tied to
GVCs, compared to the North American Free Trade Agreement (NAFTA, 18 per cent),
Trans-Pacific Partnership (TPP, 32 per cent), the South American trade bloc (MERCOSUR,
15 per cent). About 70 per cent of European Union’s gross exports were linked to GVCs in
the same period.
8.3 The Evolution of ASEAN Industrial Policy:

From Import-substitution Industrialization to
Export-oriented Industrialization to Deepening
GVC Integration
Despite impressive regional growth over the years, it is difficult to pin down what is the
ASEAN’s process of industrialization.8 As mentioned, the pace of economic reforms and
the institutional capabilities of each ASEAN member state vary in the early phases of the
Page 4 of 31
of ASEAN
industrialization process. At the outset, ASEAN was born from the collective aspirations
of its leaders to fast-track their economic independence and production (p. 212) diversifi
cation away from primary and commodity-based production to value-added manufactur
ing production as the basis for economic modernization (Bautista, 1983).9 However, varia
tions in each country’s endowments, experiences, and priorities compounded the region’s
ability to implement a broad-based industrial policy (Ariff and Hill, 2012).
From a political economy point of view, the creation of ASEAN’s was driven by the ur
gency of regional governments to be economically self-sufficient or ‘independent’, with
the gradual withdrawal of global powers in the Asian region. The end of the Cold War is a
case in point; ASEAN member states that had previously leaned on US aid scrambled to
establish a new regional order that would favour their national developments. Industrial
policy is therefore motivated by the need to industrialize and to form a new regional dy
namics that would enable growth and sustainable development for all member states
(Chang, 2006). ASEAN economies are small, which makes it even more challenging for
countries to achieve the minimum efficiency required for scale economies.
It is therefore not surprising that the idea to form ASEAN was first conceived by the for
eign ministers of Indonesia, Thailand, Malaysia, the Philippines, and Singapore, or also
known as the ASEAN-5. The ASEAN-5 countries signed the ASEAN Declaration in 1961,
which outlined the aims and purposes as a regional order that promotes economic
growth, social progress, and cultural development. The ASEAN Declaration also man
dates its member states to promote regional peace, engage in regional collaboration (i.e.
training, research, etc.) and mutual assistance on matters of common interest, including
agricultural and industrial cooperation. For ASEAN member states, these collaborations
are necessary to raise the living standards of the people in the region.
The establishment of such regional institutions had a greater purpose than economic and
social goals. Some scholars argued that the common fear of communism from mainland
China was the underlying factor in the creation of ASEAN (Jones, 2011; Narine, 2002;
Broinowski, 1982; Anwar, 1997). The political motivation led to some speculation that the
association should serve as the informal body to curb even if only indirectly, communism
in the region. During this period, we can interpret several events to provide an anti-com
munist rationale for ASEAN: the US involvement in Vietnam escalated in the region; a
joint remark or communique by the Asia and Pacific Council (ASPAC), of which Thailand,
Malaysia, and Philippines are members, explicitly endorsed regional efforts ‘to safeguard
their national independence and integrity against Communist aggression’; and the decla
ration by foreign ministers and prime ministers of member states to defend national inter
ests against communism (Broinowski, 1982: 15).10 However, it is far from certain that the
existence of ASEAN (p. 213) was motivated by the political aim of weakening the potential
spread of ideology even though contemporary developments may suggest otherwise.
There is no official statement of ASEAN explicitly condemning the potential spread of
communism into the region. Member states were strictly espoused to the principles of
consensus building, non-alignment, and self-reliance.11
Page 5 of 31
of ASEAN
The goal of economic independence led to the adoption of import-substitution industrial
ization (ISI) policy in the late 1950–1960s, albeit differing in extent and duration between
countries, before shifting to an export-oriented industrialization (EOI) strategy in the
1970–1980s. Although ISI strategy was short-lived, it marked the region’s first direct ef
fort to transform its regional production to meet global standards. But it was during the
EOI period that Southeast Asian economies began to make their mark as major global ex
porters on the world production stage. Export-oriented growth had a significant impact in
transforming the region into an industrial player. ASEAN’s annual industrial growth rates
of 8 to 10 per cent were recorded for over three decades after introducing EOI policies.
For instance, the structural transformation from ISI to EOI more than doubled the share
of manufacturing for Indonesia and Malaysia (Hill, 1997, 1997a). In the 1970s, the manu
facturing sector constituted less than 10 per cent of all merchandising exports for ASEAN
economies.12 By the mid-1990s, manufacturing had become the primary source of eco
nomic growth.
Scholars have also argued that ASEAN’s rapid economic growth during the EOI phase
was linked to FDI or Transnational Corporations (TNCs) (Hobday, 2011; Pradhan, 2009;
Moudatsou and Kyrkilis 2011). Singapore was the first to adopt such a TNC-led industrial
ization strategy, prompting others to follow suit, however, with varying degrees of suc
cess.13 ASEAN’s outward-looking industrialization strategy of export growth and attract
ing more investment often overlooks the unevenness of economic development and value-
added allocation among individual economies in the region (Oikawa, 2011). While EOI
policy allowed for greater access to international (p. 214) markets and capital inflows, the
region became increasingly vulnerable to global economic fluctuations, like with the 1997
Asian financial crisis.
The 1997 Asian financial crisis upended the region’s decades of EOI strategy and its over
all development trajectories.14 The crisis led to another round of regional industrial policy
‘rethinking’. ASEAN countries that cemented their industrialization strategy based on ex
port performance find it difficult to contain external shocks. As a result, the financial cri
sis led to deregulation and privatization of several key industries, such as the auto indus
try in Indonesia and Thailand. However, countries that continued with strong domestic
protection of ‘infant industry’ were in a better position to contain the financial crisis. For
instance, Malaysia’s unorthodox response to the financial crisis was to impose strict fi
nancial measures or capital controls—thus, restricting the flow of foreign capital in and
out of the domestic economy.15 This partly explains why the Malaysian government, un
like its neighbouring counterparts, was more successful in containing the domestic econ
omy from widespread financial panic.16 Capital controls produced faster economic recov
ery, including declines in unemployment and improvement in real wages (Kaplan and Ro
drik, 2002).17
The severity of the financial crisis left some countries with limited policy tools to take the
domestic economy back into pre-crisis levels. As a response to the financial crisis, ASEAN
economies began implementing a more integrative process of regional production as a
potential solution to reintegrate into global markets. Deepening regional economic inte
Page 6 of 31
of ASEAN
gration was sought to bring back foreign investments and reinstate market confidence.
ASEAN leaders later adopted a more formal language of cooperation in the subsequent
ASEAN Summits.18
In 1992, ASEAN adopted several key policy measures to expedite the process of economic
regionalization and integration into GVCs. The free flow of goods and services was the
cornerstone of ASEAN cooperation as outlined in the ASEAN Free Trade Agreement (AF
TA) in 1992, which was enacted to create a single market and (p. 215) international pro
duction base, attract foreign direct investments, and expand the intra-ASEAN trade and
investment nexus. The primary mechanism to achieving the goals envisioned in AFTA is
the Common Effective Preferential Tariff (CEPT) scheme that underscores a gradual re
duction and/or elimination of intra-regional tariffs and import duties up to 5 per cent for
99 per cent of all product categories by 2010.19 The CEPT includes a phased schedule of
intra-regional tariff reduction that takes into consideration the level of sensitivity of the
products (imports) to the respective ASEAN member states’ domestic industry.
A decade later, at the Bali Summit in 2003, ASEAN leaders announced that the ASEAN
Economic Community (AEC) would be the new aim of regional economic integration by
2015.20 AEC transforms the region into ‘free movement of goods, services, investment,
skilled labour and freer flow of capital’ (ASEAN Secretariat, 2015).21 The AEC replaced
the previously agreed CEPT scheme with the ASEAN Trade in Goods Agreement (ATIGA),
signed in 2009 and which came into force in 2010. The ATIGA renewed the CEPT agree
ment by including comprehensive coverage in trade in goods, full tariff reduction sched
ules and non-tariffs measures, and implementing the ASEAN Single Window (ASW) for
rapid exchange of standardized data among member states’ customs agencies, mutual
recognition agreements (MRAs), and e-ASEAN, among other new initiatives.22 The ATIGA
minimizes the cost of doing business and logistics, deepens economic linkages, creates
greater economies of scale for firms within ASEAN, and further maintains a competitive
investment environment. AEC was a clear signal to reaffirm its intentions to keep abreast
with the rest of the world.
The adoption of AEC and its mechanisms laid down the groundwork for deeper regional
and GVC participation. For ASEAN, increasing regional integration has several advan
tages. First, regional integration improves the overall performance of ASEAN; the more
connected are regional firms with one another, the more likely (p. 216) they are to export
to similar global markets. Cooperation among regional firms would result in more effi
cient production. Second, regional integration would improve the resilience of domestic
markets to global fluctuations; during the 2010 global financial crisis caused by the sub
prime mortgage crisis in the Western world, the ASEAN economies began looking into re
gional and domestic markets, creating an economic buffer from external shocks. Region
alization gave ASEAN the bargaining power it needs to balance geopolitics and anticipate
the economic rise of regional powerhouses, like China and India.
Page 7 of 31
of ASEAN
8.4 A Closer Look at ASEAN Industrial Policy in

the Era of GVCs
For years, traditional economics and leading global economic institutions questioned the
effectiveness and the ambiguous contribution of industrial policy on a country’s economic
growth. Economic miracles that a country may have experienced cannot be directly at
tributed to industrial policy or some kind of government- or state-led interventions, such
as the promotion or targeting of several strategic industries. Therefore, the lack of a stan
dardized definition of industrial policy among scholars and practitioners would often lead
to a dubious, and often inconsistent, method of analysis.
There is no consensus on the definition of industrial policy other than that it is some form
of government intervention that selectively promotes certain industries (Stiglitz, 1996;
Cimoli, Dosi, Nelson, and Stiglitz, 2009; Chang, 2011; Milberg et al., 2014; Naudé, 2010).
Another interpretation of an industrial policy is the deliberate effort to ‘defy’ its compara
tive advantage (Chang, 2003, 2011).23 Some examples of these interventions include in
fant industry protection measures through tariff and non-tariff barriers and imposing lo
cal content requirements as the basis for ‘infant industries’ protection. Note that contrary
to traditional views, countries such as the United States, Great Britain, (p. 217) Japan,
and, more recently, South Korea became the world’s industrialists precisely because of in
dustrial policy that promoted the development of national industries and nurtured innova
tion (Cimoli et al., 2009; Chang, 2011). Amsden’s (1989) seminal work on the political
economy of South Korea’s industrialization shows the important role of the state in indus
trial development. According to Amsden, the South Korean government developed its
economy not by turning to free-market prescription or by getting the prices right; South
Korea succeeded in developing its industries because it got its ‘prices wrong’.
Some scholars have also argued that industrial policy should have a more broad-based
function that promotes the competitiveness of all productive sectors (Sanjaya, 2004;
Naudé, 2010). This view suggests a more neutral approach to industrial policy in that it
has a function to elevate the growth of the entire supply side of the economy and not only
be specific to certain sectors (Rodrik, 2008). In this sense, industrial policy is an interven
tion where state and private sectors exchange information, or dialogue, to reduce or elim
inate constraints in order to allow the creation of efficient market outcomes (Hausmann
and Rodrik, 2003; Rodrik, 2004, 2008). Thereby, the state can identify and overcome bar
riers that may be a hindrance to development. Putting it differently, industrial policy is a
process where the state takes a proactive stance to change the structural characteristics
of the economy to support selective sectors that yield better prospects of economic
growth and development. Others, however, adopt a more pragmatic or practical point of
view in their interpretation of industrial policy, implying that there is no single recipe for
industrial policy (Chang, 2011). Industrial policy should reflect the conditions and specif
ic needs of countries seeking economic growth and development.
Page 8 of 31
of ASEAN
Contrary to the popular view that the Southeast Asia region grew rapidly by relying on
market forces and limited, if not acceptable, interventions on human capital and techno
logical innovations, the role of industrial policy in the development of ASEAN economies
cannot be understated (Jomo, 2001). Over the past four decades, industrial policy has had
a major role in the successful development of the agricultural and agro-processing sec
tors, such as palm oil, in Thailand, Malaysia, and Indonesia. Without the existence of such
direct intervention by the state, Malaysia would not have emerged as a global exporter in
electronics manufacturing. Likewise, Thailand would not have been known as the ‘Detroit
of the East’ had it not used government investments to nurture the development of the
automotive parts and components industry as early as the 1980s.
However, industrial policy in the era of GVCs is different in a sense that it focuses on
firms rather than the role of the state (Milberg et al., 2014). Industrial policy in the era of
GVCs would involve the shifting of intervention from the creation of domestic supply
chains to increasing bargaining power and improving value creation within the value
chains, which includes moving into higher value-added production tasks, or what is com
monly called vertical specialization; facilitating access to competitively priced (p. 218) in
termediate inputs; negotiating and creating linkages with multinational lead firms; and
moving into higher value-added production tasks (Kaplinsky and Morris, 2015; Taglioni
and Winkler, 2014).
ASEAN has had some relative success in promoting industrial competitiveness and trade
facilitation in the region. ASEAN industrial policy began as early as the 1970s. The
ASEAN Industrial Project (AIP), 1971 started ASEAN’s agenda to bring about better in
dustry competitiveness, which encouraged joint projects among ASEAN firms (Bautista,
1983; Ravenhill, 1995; Yoshimatsu, 2002). The original plan was to assign one industrial
project to each of the five (two out of five were implemented) founding members of
ASEAN (Ravenhill, 1995). Under the AIP arrangement, ASEAN firms were encouraged to
come into a joint-venture with regional partners as a way to share fixed costs of produc
tion. However, political squabbles among member states and the institutional limitations
at the national level hindered the full scope of the original project (ASEAN Secretariat,
1991; Suriyamongkol, 1988).
Regional industrial cooperation came to fruition with the subsequent implementation of

the ASEAN Industrial Complementation Arrangement (AICA) in 1981.24 AICA allows for
specialization in a narrower range of products by facilitating manufacturers to produce at
a much lower price and achieve higher optimal output for the regional market. Toyota
played an important role in implementing AICA as a solution to overcome various local
content requirements in automotive production. For instance, in 1991, under AICA, Toy
ota could purchase auto parts and components from other manufacturers in the region at
a low and competitive price, which later became known as the Brand-to-Brand (B2B) com
plementation scheme. Subsequent industrial cooperation was, however, less of a success.
In 1983, the ASEAN Industrial Joint Venture (AIJV) was implemented to grant regional
firms (registered in ASEAN) with 90 per cent preferential margins.25 But despite
ASEAN’s success in improving auto industrial competitiveness through a centralized poli
Page 9 of 31
of ASEAN
cy coordination, like the B2B, other industries, particularly ASEAN’s nineteen priority
sectors, have yet to follow suit.
The future success of ASEAN’s industrial policy hinges on its ability to sustain foreign di
rect investments. The ASEAN Investment Guarantee Agreement (ASEAN IGA) and the
Framework Agreement on the ASEAN Investment Area (AIA), first signed in 1987 and lat
er revised in 1998, provide clear guidelines regarding investment liberalization and pro
tecting foreign investors under a single roof. In 2012, (p. 219) ASEAN upgraded its provi
sions for investors’ protection through implementing ASEAN Comprehensive Investments
Agreements (ACIA). As an economic instrument, the ACIA aims to create a liberal, facili
tative, transparent, and competitive investment environment in ASEAN and further ad
vance regional economic integration.
Under the current investment framework, foreign investors are rewarded with greater
protection and flexibility, including the prohibition of performance requirements (PPRs),
which are simply demands by governments for technology transfers or for some preferen
tial treatment of investment, and the freedom to appoint senior management positions,
which often requires the inclusion of local managers in the upper hierarchy. The current
ACIA framework also sets a clear timeline for member states to implement investment lib
eralization. Additionally, ACIA has included a more comprehensive investor–state dispute
settlement (ISDS) procedure to protect investors from potential damages or losses that
arise from changes in state policy. The controversy surrounding the ISDS mechanism in
its current form is that it is skewed towards the foreign investor who can enter an arbitra
tion tribunal when there is a breach of contract (of rights granted to investors) by the
host state, causing a loss of revenue and profits to private firms. The ISDS is a legal pri
vate arbitration that bypasses national judicial systems, which implies a great disadvan
tage for governments that typically have fewer means to defend their national interests
(including litigation expenses). ISDS is considered an integral part of several bilateral
and multilateral trade treaties such as the North American Free Trade Agreement (NAF
TA) and the Comprehensive and Progressive Trans-Pacific Partnership (CPTPP), which is
the successor of the Trans-Pacific Partnership (TPP) after the US withdrawal (Sturgeon,
Van Biesebroek, and Gereffi, 2007). The renewed terms on investor protection reflect a
bold intention to better accommodate the interests of investors and the maintaining of
foreign capital.
ASEAN has traditionally been a strong proponent of intellectual property rights (IPR) pro
tection, although industrializers of the past—including the United States, the United
Kingdom, and Japan—industrialized without the existence and the enforcement of IPR
protection (Chang, 2011).26 The ASEAN Framework Agreement on Intellectual Property
Rights in 1995 and the ASEAN Intellectual Property Rights Action Plan 2011–2015 are
the two primary documents that outline the general principles of intellectual rights pro
tection in the region.27 The framework on intellectual property rights puts in place effec
tive protection measures for IPR, including a new provision to improve the administrative
body for data collection and human resources (p. 220) development. The existing frame
work on intellectual property rights provides the groundwork to establish a new ASEAN
Page 10 of 31
of ASEAN
patent system, such as the ASEAN Patent Office. Policymakers recognized the urgency of
renewing intellectual property rights protection to promote and encourage more technol
ogy start-ups or ‘unicorns’ in the region.28 Although ASEAN’s new intellectual property
rights regime was put in place to protect the future of ASEAN’s e-commerce industry, it
falls short in addressing the extent of intellectual property rights protection in its current
production structures. In the world of GVCs, government protections of intellectual prop
erty have the effect of ‘locking in’ the monopoly power to the intangible asset creation
(Pagano, 2014; Durand and Milberg, 2019). Intangible assets are increasingly monopo
lized by multinational firms (mostly in the West) who own intellectual property. However,
this ‘intellectual monopoly’ in intangible assets would have great implications on the
presence of economies of scale and network externalities associated with intangible asset
creation. The current interpretation of IPR has yet to address the extent or the deepening
of intellectual monopoly within ASEAN GVCs.
The general evolution of ASEAN’s industrial policy suggests a greater facilitation for for
eign investors, that is, access to regional markets and protection, as opposed to creating
mutually beneficial outcomes for domestic producers and lead firms. While ASEAN ac
commodates member states’ levels of development and considers deeply the sensitivity of
national products to imported goods, recent policy initiatives lacked the creative ap
proaches necessary to improve regional production efficiencies and further encourage in
novation in local economies. For instance, the provisions needed to create economies of
scale and a more specialized division of labour in the auto industry (i.e. the AIP and AICA
provisions) remains a unique case that existed during the pre-GVC era when industrial
policy was centrally coordinated by the state at the national level.29 ASEAN has yet to
replicate similar region-wide policy initiatives in several of its targeted strategic indus
tries.
One must also consider the proliferation of regional trade arrangements that have be
come the preferred alternative to the slow-paced, and often convoluted, multilateral ne
gotiations process. This might restrict the effectiveness of ASEAN industrial policy. For
example, competing for regional trade arrangements such as the CPTPP, formerly known
as TPP, might overshadow ASEAN’s very own CEPT arrangement as the cornerstone to
promote regional trade competitiveness. Four out of ten ASEAN members (Brunei,
Malaysia, Singapore, and Vietnam) are parties to the CPTPP. Indonesia—the largest coun
try in ASEAN in terms of the size of the economy and (p. 221) population—has considered
joining the CPTPP in the past but retracted its intention to join. The proliferation of re
gional trade arrangements pressured ASEAN to respond to the demands for better pro
tection by global investors. This is clear when the number of measures, such as ISDS and
the recent intellectual property rights framework, are considered. We cannot ignore the
spectacular rise of China and its increasing demand for intermediate inputs from ASEAN
suppliers. China’s economic influence in the region poses an economic opportunity to ex
pand ASEAN’s export market, especially given the impasse present in the current state of
international trade negotiations.
Page 11 of 31
of ASEAN
8.4.1 Reshaping ASEAN Industrial Policy in the Era of GVCs: Towards
a more Inclusive Participation
In the effort to expedite the process of regional economic integration in Southeast Asia,
ASEAN has put forward several updates of its strategic action plans to complement the
ASEAN Economic Community (AEC) Blueprint 2025.30 The aim of the AEC Blueprint 2025
was to strengthen and reinforce the characteristics of the ASEAN Economic Community
by 2025, that is: (i) a highly integrated and cohesive economy; (ii) a competitive, innova
tive, and dynamic ASEAN; (iii) enhanced connectivity and sectoral cooperation; (iv) a re
silient, inclusive, and people-oriented, people-centred ASEAN; and (v) a global ASEAN. A
key feature in its consolidated strategic action plan (CSAP) is to allow for a more struc
tured monitoring and reporting of the implantation progress of the AEC 2025. The newly
implemented CSAP makes it more transparent and open for stakeholders to provide feed
back towards achieving ASEAN economic integration priorities. The newly adopted CSAP
reinforces existing integration areas and a new focus, such as cross-sector complementar
ities and coordination among ASEAN.
The new CSAP made clear efforts to strengthen the characteristics of regional economic
integration and ‘ASEAN centrality’. While the language of regional cooperation and indus
trial policy within ASEAN has been consistent in endorsing the free movement of goods
and services, people, and capital, the ASEAN 2025 CSAP shifted towards the promotion of
increased participation in regional and global value chains. For instance, in efforts to cre
ate a ‘highly integrated and cohesive economy’, ASEAN policymakers have focused on
five key elements that include trade in goods; trade in services; investments; financial in
tegration, financial inclusion, and financial stability; (p. 222) facilitating movements of
skilled labour; and finally enhancing participation in GVCs. In each of these elements,
new provisions have been included to better facilitate the participation of SMEs—espe
cially, in the least-developed ASEAN member states—within the GVC structures.31
Some distinct key action plans outlined in AEC 2025 include increasing the engagement
of the private sector with a particular emphasis on the development of SMEs. This policy
area is in response to the need to improve the economic distribution and participation of
local suppliers in regional supply chains. ASEAN has included new provisions to allow for
better opportunities and resources for SMEs to increase their share of participation and
reduce constraints such as high costs for exports (Bernard et al., 2007). SMEs face more
resource constraints than domestic large firms, including access to competitive interme
diate inputs. Moreover, they capture a low share of the foreign value added of exports
and, in turn, contribute indirectly by supplying to domestic firms (Lopez-Gonzales, 2017).
The structural contribution of SMEs to ASEAN economies is significant. SMEs represent
a majority (above 90 per cent) of commercial enterprises in South East Asia and employ
58 to 91 per cent of the domestic workforce. SMEs contribute over 50 per cent of
ASEAN’s GDP (OECD/ERIA, 2018).
SMEs’ employment in the production of intermediate goods traded within the value
chains has grown at a much faster rate than overall employment growth (Lopez-Gonzales,
Page 12 of 31
of ASEAN
2017: 9). This illustrates the contribution that SMEs make in supplying intermediate in
puts to regional supply chains (Kowalski et al., 2015). However, SMEs’ contribution to to
tal exports is relatively insignificant, estimated at around 10 to 30 per cent. The recent
change in policy shows that ASEAN policymakers have put a great deal of effort into re
ducing costs and improving access to physical and non-physical infrastructures for SMEs,
including practical and/or vocational training. Although SME policies have traditionally
been associated with creating social cohesion, it has evolved into stimulating innovation
and entrepreneurship in ASEAN.
Over the last thirty years, ASEAN’s industrial strategy focused on creating greater mar
ket access, through reducing or eliminating barriers to trade and investments, and main
taining high foreign investments. While such policies may be necessary to motivate the
flow of goods and services between different production nodes in the value chain, they
fall short of guaranteeing firms and industries from ‘industrial upgrading’, or moving up
the value chains from relatively lower to higher value-added economic activities (Pietro
belli and Rabellotti, 2006). Industrial upgrading captures more of the value-added in the
production process and increases profits for local firms. The recent policy priorities to in
crease (p. 223) intra-ASEAN trade and investment flows consequently created a vacuum
for structural change and the ability to foster innovation.
8.4.2 ASEAN Industrial Policy in GVCs: Success or Failure?
Given the recent adoption of ASEAN’s GVC policies, it is too early to tell whether such a
shift was a success. The AEC Blueprint outlined the agenda to strengthen and reinforce
regional economic integration through trade facilitation and increasing the participation
of SMEs within GVCs. In this section, we assess the shortcomings of ASEAN industrial
policy in the era of GVCs and put forward potential solutions to ensure that the current
policy focus is consistent to the objectives of ASEAN economic integration.
First, ASEAN’s regional policy coordination has helped to deepen regional integration,
however, it remains incoherent. The pace and quality of policy implementation at the do
mestic level vary across member states (Bautista, 1983; Yoshimatsu, 2002). For instance,
while reducing and/or eliminating tariffs across the region has been relatively successful
(i.e. zero tariffs for most products for ASEAN countries), other technical aspects, such as
improving customs procedures, dispute settlement mechanisms, the rule of origin, and
the removal of non-tariff barriers, have weak institutional enforcements (Kadarusman,
2016). The capacity for national ministries and agencies to coordinate the regionally
agreed principles also matters a great deal (Chang, 2011; Kadarusman, 2016). The lack of
policy enforcement is partly because ASEAN policy relies on a consensus-building
process, which allows flexibility, such as additional time and assistance, to implement the
agreed regional commitments.
Second, while increasing SME participation is a positive attempt to redistribute gains (i.e.
employment and income) from taking part in the global economy, more needs to be done
in expanding regional cooperation on SME development. Firms that are more engaged in
Page 13 of 31
of ASEAN
export markets tend to be more efficient and pay more in wages (Bernard and Jensen,
2004). However, SME limitations are well documented and policy dimensions to over
come them have been proposed (Lévy, 1988; Jinjarak et al., 2014; OECD, 2014; Taglioni
and Winkler, 2014; Kowalski et al., 2015; Lopez-Gonzales, 2017). At the current stage, re
gional cooperation on SME development takes the form of information and experience
sharing. More direct cooperation would be needed to better streamline the lessons
learned from national to regional level. ASEAN SMEs could benefit from adopting a more
standardized system of exchange, including the financial information of SMEs among
ASEAN member states.
Third, ASEAN’s trade and investment outlook has persistently fostered export perfor
mance but lacked the infrastructure and framework for industrial upgrading, (p. 224) pri
marily for the nineteen priority sectors. This partly explains why the regional production
network in ASEAN continues to be concentrated in ‘buyer-led’ GVCs as opposed to ‘pro
ducer-led’ GVCs (Gereffi, 1994). The former refers to production chains whose lead firms
outsource the entire production process and final product to suppliers.32 Conversely, in a
producer-led production chain, the lead firm controls the means of production and tech
nologies in the host country.33 This field have argued that GVC participation in a buyer-
led GVC is increasingly difficult and costly for local firms in developing economies.
ASEAN’s industrial policies lack the reward (or penalty) system that East Asian industrial
ists received during their process of industrialization (Amsden, 1989; Chang, 2006;
Naudé, 2010). A reward system, such as an export subsidy, encouraged South Korean
chaebols, or large industrial conglomerates, to enter export markets. ASEAN may consid
er incorporating a reward system (such as an export subsidy) for local firms that success
fully achieved some form of upgrading (i.e. functional) for certain product categories.
Without such incentives, policies to eliminate tariffs and the removal of non-tariff barriers
would, in the long run, only benefit large and foreign multinational firms.
Finally, there is a great deal of uncertainty regarding ASEAN’s intellectual property

rights regime. The new initiatives to create a new ASEAN patent system, which includes
the ASEAN Patent Office to promote regional patent protection, respond directly to the
growth of the non-productive and technology sectors. While we cannot undermine the im
portance of an ASEAN patent system, a more general approach that would allow for
knowledge transfers would be more appropriate in GVCs.
8.5 ASEAN’s Regional Policy Coordination: The

Case of the Auto Industry in Indonesia and
Thailand
This section reviews the progress of ASEAN industrial policy in the automotive industry
that began as early as the 1970s. There are historical and political economy (p. 225) ele
ments that contributed to the rapid growth of the automotive parts and components and
assembly sectors in Southeast Asia. Japanese automakers, particularly Toyota, and, to
Page 14 of 31
of ASEAN
some extent, Kia of South Korea, played a key role in transforming the region as the hub
in auto manufacturing. These conglomerates developed new business ‘relationships’ with
regional authorities. The growth of the automotive industry in ASEAN is congruent with
the internal political interests of the current regimes. The development of the automotive
industry is arguably one of the recent success stories that emerge directly from ASEAN
regionally coordinated industrial policy. Today, ASEAN is emerging as the hub of regional
automotive manufacturing with the production of auto parts and components integrated
across member states.
The automotive industry is undoubtedly one of the most complex global industries, with
production centres (or clusters) scattered across several continents. The auto industry re
lies on large capital investment and research and development, and institutional support
for capacity building. Auto firms are increasingly becoming more dependent on each oth
er for inputs and for delivering quality products to the end market. The auto industry is
persistently transforming from geographical, organizational, and technological stand
points (Sturgeon and Van Biesebroeck, 2011; Sturgeon et al., 2016). Auto production is no
longer concentrated in advanced countries, such as the United States, Germany, and
Japan. There has been a gradual shift away from the global North and into the global
South (Traub-Merz, 2017). In 2015, for instance, China (26 per cent of global production
share, 24 million units) became the largest automotive manufacturer, followed by the
United States (13 per cent, 12 million units), Japan (10 per cent, 9 million units), Ger
many (6 per cent, 6 million units), and South Korea (5 per cent, 4.5 million units). At the
regional level, India, Mexico, Indonesia, and Thailand are also becoming prominent auto
manufacturers. The trend in automotive production is likely to divide into automotive
manufacturers in the global North producing a ‘core competence’ of skills-based produc
tion, such as engineering and design, while manufacturers in the global South take up
most of the labour-intensive and assembling process (Sturgeon and Van Biesebroeck,
2011).
As a whole, global auto production has been on the uptrend and Asia (including China)
has been the major contributor. Global sales in automotive products (excluding motorcy
cles) rose from 65 million units in 2005 to 90 million units in 2015; Asia is the largest au
tomotive consumer market, with growing motor vehicle sales as high as 43 million units,
followed by North America (25.6 million units) and Europe (19 million units). However,
demand for motor vehicles in Latin America (4.5 million units) and Africa (1.5 million
units) has not been as dynamic as in other regions in recent years. The rise of Asia’s mid
dle-income group partly explains the demand surge for vehicles. From a regional perspec
tive, Asia captures more than half of the global production. In 2015, Asia produced about
48 million units of vehicles, up from 18 million in 2010 (International Organization of Mo
tor Vehicle Manufacturers. Global auto (p. 226) production absorbs roughly 60 million
workers worldwide or about 5 per cent of the world’s total manufacturing employment
(OICA, 2015).34
Page 15 of 31
of ASEAN
At the regional level, there is a clear shift in investment into countries with relatively low
er production costs, that is, Thailand, Indonesia, and Vietnam in Southeast Asia; China in
East Asia; and Mexico and the American South in North America (Sturgeon et al.,
2016).35 Conversely, auto industry manufacturing in the less-industrialized economies re
mains in labour-intensive production, such as in parts and components and assembling
(Milberg and Winkler, 2010).
The development of the auto industry is consistent with the industry’s global transforma
tion in production structures, distribution, and innovation. Indonesia and Thailand are the
largest automotive manufacturers in Southeast Asia, with a combined production of over
3 billion units of vehicles, and export values of roughly US$30 billion in 2015 alone. For
eign direct investment (FDI) reached US$4 billion for Indonesia in 2014 and net FDI in
vestments of US$15 billion for Thailand from 2000 to 2014 (Phungtua, 2017).
Indonesia and Thailand have made new commitments to place the auto industry at the
centre of their national economic and development programmes. The government of In
donesia, for instance, has set a production target of 2 million vehicle units (from 1 mil
lion) and similarly the government of Thailand has increased its production target to 4
million vehicles (from 2 million) by 2020.36 Indonesia issued a presidential decree to re
structure the country’s automotive industry away from traditional parts and components
production to producing environmentally friendly and technology-based vehicles. The de
cree also outlined new initiatives for more R&D and investments in technology-based uni
versities.
The auto industry represents a small but growing sector. Recent figures show that the
sector’s contribution to employment grew impressively from roughly 58,000 to 125,000 in
Indonesia and similarly 90,000 to 193,000 in Thailand from 1998 to 2006 (UNIDO, 2019).
The industry’s contribution to value-added per worker also showed a growth trend from
US$20,000 per worker to US$34,000 per worker in Thailand and US$19,000 per worker
to US$53,000 per worker in Indonesia.37 (p. 227)
8.5.1 The GVC in the Automotive Industry
Mainstream economics predict that increased openness to international trade, typically

through various policies such as export promotions, deregulation, and, more recently,
greater participation in the GVC network, leads to faster economic growth. The argument
is that openness to the international market opens up new economic opportunities for
host countries, including foreign investment inflows from capital-rich countries to emerg
ing countries, providing better access to technology for production. The neo-liberal agen
da38 (i.e. the Washington Consensus policy) of the 1970s also contributed significantly to
global production fragmentation, which enabled developing economies to contribute to
the global supply chains by producing the labour-intensive aspects of the production
(Feenstra, 1998).39 The new pattern of global trade led to production offshoring part of
the production processes to factories in developing economies, which meant that multina
tional firms could lower their production unit costs and keep their core competence (i.e.
Page 16 of 31
of ASEAN
design, engineering, etc.) and developing countries could specialize in certain production
tasks without having to build an entire industry from the ground up (Gereffi and Kaplin
sky, 2001 Humphrey and Memedovic, 2003; Humphrey, 2004; Pietrobelli and Rabellotti,
2006; Cattaneo et al., 2010).
Production fragmentation gave way to more sophisticated trade flows known as GVCs.40
The GVC is an intricate web of modes of production; GVC production is not just a linear
stream of a value-added creation that involves a multiple cross-border of production
mode (i.e. export and re-export). Intermediate input goods cross and re-enter the border
more than once via other countries’ exports. What the GVC concept also proposes is the
opportunity for local producers to learn from the global leaders of the chains, which may
be buyers or producers, and capture the dynamics of the trade from trade in final goods
to intermediate inputs trade, also known as ‘trade-in-tasks’. The GVC analysis includes
not only production aspects but also the flow of information, clarifying that firm linkages
within and between industries involve skills and knowledge-sharing, critical for increas
ing value added.
Greater participation in GVCs is also a way for developing economies’ suppliers to

(p. 228)
improve their prospects of earning economic profits (Gereffi, 1999; Humphrey, 2004).
However, governance in GVCs and competitive pressure make it difficult for developing
economies’ suppliers to earn higher profit margins (Humphrey and Schmitz, 2001). Sup
pliers are under pressure to provide quality products at a low unit price. In most cases,
firms may earn profits outside production. It is necessary for suppliers to have the ability
to upgrade and capture more value within the value chain. We can categorize automotive
GVCs into six main stages: vehicle design and development; parts and components; mod
ules and subsystems; system integration and final assembly; marketing and sales; and fi
nally, replacement parts and recycling. Of these categories, the lead firms specialize in
vehicle design and development and marketing and sales. It outsources other product
categories to local firms in the supply chain.
In 2014, 46 per cent (or about US$637 billion) of the value of global trade was of parts
and components and 5 per cent (or US$74 billion) was of sub-assemblies (Sturgeon et al.,
2016). The value of parts and components rose from previous years—41 per cent in 2007.
From this, Sturgeon et al. estimated that final products for passenger vehicles con
tributed to about 49 per cent of total exports in the automotive sector in 2014 (a fall from
53 per cent in 2007). Parts and components also represent a large chunk of Indonesia’s
and Thailand’s automotive exports. The source of auto demand mostly comes from emerg
ing economies with a rising middle-income class. China, Indonesia, India, and Brazil are
among the world’s largest economies, with strong market growth rates of 15.1, 8.5, 8.2,
and 7.4 per cent, respectively, in 2014. However, advanced economies such as the United
States, Japan, and Germany are experiencing a fall in demand with market growth rates
of –0.4, –0.5, and –0.7, respectively, in the same period. This trend in automotive demand
is likely to continue.
Page 17 of 31
of ASEAN
8.5.2 A Path to Development and Industrial Upgrading
Participation in GVCs presents emerging countries with an alternative channel or path

way for economic development. For local firms in developing economies, GVC participa
tion is an opportunity to have greater access to technological advancements and poten
tially have a positive spillover effect for the domestic economy (Humphrey and Meme
dovic, 2003).41 GVC participation is closely associated with growing productivity and less
concentration in a country’s export basket, including exports of more (p. 229) sophisticat
ed products (Kowalski et al., 2015). GVCs can also be the driving force in creating local
network clusters, or a ‘collective efficiency’ or ‘clusters’, to the overall knowledge of the
economy. Silicon Valley in California, United States, and Bangalore, India, are prime ex
amples of clusters in the tech industry. These clusters create positive network externali
ties. But the important question is whether GVC participation leads to a ‘low-
road’ (increasing profits via squeezing wages) or ‘high-road’ (sustainable income growth)
process of development. So far, the channels for development in GVCs have not been di
rect (Milberg et al., 2014).
Recent scholarly work has tried to connect (and measure) both the economic and social
gains from increased GVC participation (Rossi, 2013; Barrientos et al., 2010; Bernhardt
and Milberg, 2011; Milberg and Winkler, 2001, 2013). We have expected to see economic
upgrading from export expansion and social upgrading from improvements in the work
ers’ working conditions such as employment and real wages increases. However, evi
dence suggests that economic upgrading does not lead to social upgrading, and vice ver
sa, and there are significant variations between countries and sectors (Taglioni and Win
kler, 2014; Kaplinsky and Morris, 2015).
From a development standpoint, there are potential shortcomings with GVC analysis.
First, emerging countries face barriers (i.e. technology, access to capital, etc.) and asym
metrical power structures skewed towards lead firms (Gereffi and Kaplinsky, 2001). Thus,
knowledge transmission for upgrading is not automatic and may depend on the relation
ships between local firms and lead firms (Humphrey and Schmitz, 2002). The literature
on FDI mostly shows weak evidence linking FDI inflow and its spillover effects, although
there is evidence showing a significant association with improvements in human capital
(Slaughter, 2002). Second, industrial upgrading functions at the level of the individual
firm in a particular value chain. The success stories of GVC upgrading have been spo
radic for selected industries and countries, which makes it difficult for policymakers to
replicate successes in other industries (Bair, 2005; Brewer, 2011). Thus, industrial up
grading in one particular sector does not result in overall development at the national lev
el (Bair and Gereffi, 2004). Third, there is no such thing as a ‘one-size-fits-all’ policy pre
scription for countries to capture the gains from international trade and greater partici
pation in a GVC network (Pietrobelli and Rabellotti, 2006). This would require countries
to identify specific transmission channels, such as market entry (backward or forward
linkages), market structure, and the labour market conditions within the value chain to
capture the gains.
Page 18 of 31
of ASEAN
8.5.3 ASEAN’s Regional Policy Coordination in the Automotive Indus
try
ASEAN has had a significant role in ensuring the region takes full advantage of global
production in industries such as textiles and garments, electronics, and automotive.
The ASEAN Complementation schemes ‘allow parts manufacturers to supply final
(p. 230)
assemblers in all member states from a single ASEAN location at favourable terms of
trade as long as inter-ASEAN trade among taking part companies remains
balanced’ (Sturgeon and Florida, 2000). ASEAN’s complementation schemes include pro
grammes such as the ASEAN Industrial Joint Venture (AIJV) in 1983, Brand-to-Brand
Complementation (BBC) in 1988, and ASEAN Industrial Cooperation (AICO) in 1996.
These programmes rely on resource-pooling and market-sharing among member states to
generate economies of scale. One application of the complementation scheme is to allow
countries to trade different aspects of the production process, including as parts and
components, and engine transmissions, from different origins at lower costs than would
have been possible if the scheme were not in place. More than that, complementation
schemes (like the ASEAN AICO) allows for inter-firm cooperation, encouraging increased
intra-ASEAN trade, FDI flows, and technological transfers. The ASEAN complementation
scheme plays a unique role in ensuring that member states benefit from the global pro
duction network. As Traub-Merz (2017) asserts in his study on the rise of automotive in
emerging markets, ‘without membership of large economic zones in which duty-free in
ternal market trade exists, no country has yet entered export markets for cars on a signif
icant level’.
8.6 The Future Milestones in ASEAN Industrial

Policy
This chapter looked more closely into ASEAN industrial policy in the era of the GVC. After
periods of industrial policy evolution, from import-substitution industrialization in the
1960s to export-oriented growth in the 1970s and 1980s to focusing on deepening GVC
participation, ASEAN has emerged as one of the most dynamic economic regions in the
world. ASEAN has made significant inroads into deepening its participation in regional
and global value chains by eliminating tariffs, removing non-tariff barriers across mem
ber states, and improving general and conformance standards and providing greater pro
tection and rights for foreign investors. However, ASEAN industrial policy in the era of
the GVC has yet to bring significant outcomes in the automotive sector. ASEAN’s industri
al policy in the automotive industry, like AIJV and AICO, provides a great example of how
a regionally coordinated industrial policy creates scale economies and better distributes
the gains to member states. ASEAN should look back to its past policies and rethink how
it can introduce new forms of policy initiatives suited to address the intricacies of the
global production network.
Page 19 of 31
of ASEAN
We noticed that ASEAN’s industrial policy in the last few years shifted to addressing the
participation and inclusivity of SMEs in regional value chains. ASEAN has been particu
larly active in SME policy in recent years, which includes facilitating SMEs with (p. 231)
greater access to financial resources and vocational training, to name just a few. While
the redistribution of gains from increased GVC participation may be good for SMEs in the
sense that it improves employment and income, it does not necessarily address the asym
metries of market power within the GVC network. While we cannot understate the value
of their contribution in supplying intermediate input goods for large domestic firms and
multinational firms, it leaves a large vacuum in policy for large domestic firms that are in
a better position to compete with multinational lead firms within the existing GVC struc
ture.
The future success of ASEAN would hinge on its ability to build globally competitive lead
firms. Our assessment of ASEAN industrial policy in the era of GVCs, as outlined in the
ASEAN Economic Blueprint 2025, showed that limited policies were available to ensure
industrial upgrading, or moving up from relatively lower value-added to higher value-
added economic activities. ASEAN’s priority to improve local participation and the inclu
sivity of SMEs should not outweigh policy initiatives for innovation for large domestic
firms that are directly competing with foreign multinational firms.
Stakeholders need to rethink regional industrial policies within the existing intricate web
of global production. Policymakers should reconsider the aim of regional integration
based only on increasing the number actively participating in supply chains and look be
yond and consider how local actors can take full advantage of the GVC network. ASEAN
industrial policy in the era of GVCs could push countries into a ‘low value-added trap’, or
increased specialization in low value-added labour-intensive manufacturing. Policies that
would increase SME participation fall short in encouraging innovation and the possibility
of capturing profits for local firms. ASEAN policymakers should also consider new ways
to build a regional hub for research and development that would be useful for identifying
the possibility and the scope of functional upgrading in priority sectors.
References
ford: Oxford University Press.
Anwar, Dewi Fortuna (1997) ‘ASEAN and Indonesia: Some Reflections’, Asian Journal of
Political Science 5(1): 20–34.
Arfani, Riza Noer (2017) ‘Political Economy of Regionalism in ASEAN and Its +3 Part
ners: Contemporary Changes in the Automotive and Electronics Production Networks’,
Ritsumeikan International Affairs 15: 67–119.
Ariff, Mohammed and Hal Hill (2012) Export-oriented Industrialisation: The ASEAN Expe
rience. London: Routledge.
Page 20 of 31
of ASEAN
ASEAN-Japan Center (2019) ‘Global Value Chains in ASEAN: A Regional Perspective’,
ASEAN Promotion Center on Trade, Investment, and Tourism. January. (p. 232)
ASEAN Secretariat (1991) ‘ASEAN Economic Cooperation for the 1990s’. A Report Pre
pared for the ASEAN Standing Committee. Philippines: Philippine Institute for Develop
ment and ASEAN Secretariat.
ASEAN Secretariat (2015) ‘ASEAN Economic Community Blueprint’. Association of South-

East Asian Nations.
Bair, Jennifer, (2005) ‘Global Capitalism and Commodity Chains: Looking Back, Going For
ward’, Competition & Change 9(2): 153–80.
Bair, Jennifer and Gary Gereffi (2004) ‘Upgrading, Uneven Development, and Jobs in the
North American Apparel Industry’, in William Milberg (ed.) Labor and the Globalization of
Production. London: Palgrave Macmillan, pp. 58–87.
Barrientos, Stephanie, Gary Gereffi, and Arianna Rossi (2010) ‘Economic and Social Up
grading in Global Production Networks: Developing a Framework for Analysis’, Interna
tional Labor Review 150(3–4): 319–40.
Bautista, Romeo M. (1983) Industrial Policy and Development in the ASEAN Countries.
Manila: PIDS.
Bernard, Andrew B., J. Bradford Jensen, Stephen J. Redding, and Peter K. Schott (2007)
‘Firms in International Trade’, Journal of Economic Perspectives 21(3): 105–30.
Bernhardt, Thomas and William Milberg (2011) ‘Economic and Social Upgrading in Glob
al Value Chains: Analysis of Horticulture, Apparel, Tourism and Mobile Telephones’. Glob
al Development Institute Working Paper Series, University of Manchester.
Brewer, Benjamin D. (2011) ‘Global Commodity Chains and World Income Inequalities:
The Missing Link of Inequality and the Upgrading Paradox’, Journal of World-Systems Re
search 17(2): 308–27.
Broinowski, Alison (ed.) (1982) Understanding ASEAN. Basingstoke: Macmillan Interna

tional Higher Education.
Cattaneo, Oliver, Gary Gereffi,and Cornelia Staritz (eds) (2010) ‘Global Value Chains in a
Post-crisis World: A Development Perspective’. Washington, DC: World Bank.
Chang, Ha-Joon (2003) ‘Trade and Industrial Policy Issues’, in Ha-Joon Chang (ed.) Re
thinking Development Economics. London: Anthem Press, pp. 257–76.
Chang, Ha-Joon (2006) The East Asian Development Experience: The Miracle, the Crisis
and the Future. London: Zed Books.
Chang, Ha-Joon (2011) ‘Industrial Policy: Can We Go beyond an Unproductive Confronta

tion?’, Annual World Bank Conference on Development Economics, pp. 83–109.
Page 21 of 31
of ASEAN
Cimoli, Mario, Giovanni Dosi, Richard R. Nelson, and Joseph E. Stiglitz (2009) ‘Institu
tions and Policies Shaping Industrial Development: An Introductory Note’, in Mario
Cimoli, Giovanni Dosi, and Joseph E. Stiglitz (eds) Industrial Policy and Development: The
Political Economy of Capabilities Accumulation. Oxford: Oxford University Press, pp. 1–
16.
Durand, Cédric and William Milberg (2019) ‘Intellectual Monopoly in Global Value
Chains’, Review of International Political Economy 27(2): 1–26.
Feenstra, Robert C. (1998) ‘Integration of Trade and Disintegration of Production in the

Global Economy’, Journal of Economic Perspectives 12(4): 31–50.
Gereffi, Gary (1994) ‘The Organization of Buyer-driven Global Commodity Chains: How
US Retailers Shape Overseas Production Networks’, in Gary Gereffi and Miguel Korze
niewicz (eds) Commodity Chains and Global Capitalism. Greenwood, CA: Praeger Publish
ers, pp. 95–122.
Gereffi, Gary (1999) ‘A Commodity Chains Framework for Analyzing Global Industries’,
Institute of Development Studies 8(12): 1–9. (p. 233)
Gereffi, Gary and Raphael Kaplinsky (2001) ‘Introduction: Globalisation, Value Chains and
Development’, IDS Bulletin 32(3): 1–8.
Goldstein, Morris (1998) The Asian Financial Crisis: Causes, Cures, and Systemic Implica
tions. Washington, DC: Peterson Institute for International Economics.
Harvey, David (2007) A Brief History of Neoliberalism. New York: Oxford University
Press.

Journal of Development Economics 72(2): 603–33.
Hill, Hal (1997) ‘Rapid Industrialization in ASEAN: Some Analytical and Policy Lessons’,
Agenda: A Journal of Policy Analysis and Reform 4(4): 419–32.
Hill, Hal (1997a) ‘Towards a Political Economy Explanation of Rapid Growth in ASEAN: A
Survey and Analysis’, ASEAN Economic Bulletin 14(2): 131–49.
Hobday, Mike (2011) ‘Learning from Asia’s Success Beyond Simplistic “Lesson-Making”’.
UNU-WIDER Working Paper No. 2011-42. Helsinki: UNU-WIDER.
Humphrey, John (2004) ‘Upgrading in Global Value Chains’. Working Paper No. 28. Gene
va: ILO Policy Integration Department.
Humphrey, John and Olga Memedovic (2003) ‘The Global Automotive Industry Value
Chain: What Prospects for Upgrading by Developing Countries?’. UNIDO Sectorial Stud
ies Series Working Paper. Vienna: UNIDO.
Page 22 of 31
of ASEAN
Humphrey, John and Hubert Schmitz (2002) ‘How Does Insertion in Global Value Chains
Affect Upgrading in Industrial Clusters?’, Regional Studies 36(9): 1017–27.
IMF (International Monetary Fund) (2019) World Economic Outlook. Washington, DC:
IMF.
Jinjarak, Yothin, Paulo J. Mutuc, and Ganeshan Wignaraja (2014) ‘Does Finance Really
Matter for the Participation of SMEs in International Trade? Evidence from 8,080 East
Asian Firms’. ADBI Working Paper No. 470. Asian Development Bank Institute.
Jomo, Kwame S. (2001) ‘Rethinking the Role of Government Policy in South East Asia’, in
Joseph E. Stiglitz and Shahid Yusuf (eds) Rethinking the East Asian Miracle. Oxford: Ox
Jones, Lee (2011) ASEAN, Sovereignty and Intervention in Southeast Asia. Basingstoke:
Palgrave Macmillan.
Kadarusman, Yohanes B. (2016) ‘AEC 2015 and Beyond: Realizing ASEAN Value Chain’,
Thailand and the World Economy 34(2): 1–21.
Kaplan, Ethan and Dani Rodrik (2002) ‘Did the Malaysian Capital Controls Work?’, in Se
bastian Edwards and Jeffrey A. Frankel (eds) Preventing Currency Crises in Emerging
Markets. Chicago, IL: University of Chicago Press, pp. 393–440.
Kaplinsky, Raphael and Mike Morris (2015) ‘Thinning and Thickening: Productive Sector
Policies in the Era of Global Value Chains’, European Journal of Development Research,
8(4): 625–45.
Kaplinsky, Raphael and Jeff Readman (2005) ‘Globalization and Upgrading: What Can
(and Cannot) Be Learnt from International Trade Statistics in the Wood Furniture Sec
tor?’, Industrial and Corporate Change 14(4): 679–703.
Kowalski, Przemyslaw, Javier Lopez Gonzalez, Alexandros Ragoussis, and Cristian Ugarte
(2015) ‘Participation of Developing Countries in Global Value Chains: Implications for
Trade and Trade-related Policies’. OECD Trade Policy Paper No. 179. Paris: OECD Pub
lishing.
Lévy, Benny (1988) ‘Korean and Taiwanese Firms as International Competitors: The Chal
lenges Ahead’. Center for Development Economics 110, Department of Economics,
Williams College. (p. 234)
Lin, Justin and Ha-Joon Chang (2009) ‘Should Industrial Policy in Developing Countries
Conform to Comparative Advantage or Defy It? A Debate between Justin Lin and Ha-Joon
Chang’, Development Policy Review 27(5): 483–502.
Lopez-González, Javier (2017) ‘Mapping the Participation of ASEAN Small- and Medium-
sized Enterprises in Global Value Chains’. OECD Trade Policy Papers No. 23. Paris:
OECD.
Page 23 of 31
of ASEAN
Milberg, William and Deborah Winkler (2001) ‘Economic and Social Upgrading in Global
Production Networks: Problems of Theory and Measurement’, International Labour
Review 150(3–4): 341–65.
Milberg, William and Deborak Winkler (2010) ‘Trade Crisis and Recovery: Restructuring
of Global Value Chains’. Washington, DC: World Bank.
Milberg, William and Deborah Winkler (2013) Outsourcing Economics: Global Value
Chains in Capitalist Development. Cambridge: Cambridge University Press.
Milberg, William, Xiao Jiang, and Gary Gereffi (2014) ‘Industrial Policy in the Era of Spe
cialized Industrialization’, in José Manuel Salazar-Xirinachs, Irmgard Nübler, and Richard
Kozul-Wright (eds) Transforming Economies: Making Industrial Policy Work for Growth,
Jobs and Development. Geneva: International Labour Organization, pp. 151–78.
Moudatsou, Argiro and Dimitrios Kyrkilis (2011) ‘FDI and Economic Growth: Causality for
the EU and ASEAN’, Journal of Economic Integration 26(3): 554–77.
Narine, Shaun (2002) Explaining ASEAN: Regionalism in Southeast Asia. Boulder, CO:
Lynne Rienner Publishers.
Naudé, Wim (2010) ‘Industrial Policy: Old and New Issues’. Working Paper No. 2010–106,
World Institute for Development Economics Research.
OECD (2014) ‘Small and Medium-sized Enterprises in Global Markets: A Differential Ap
proach for Services?’. OECD Trade Policy Paper No. 165. Paris: OECD Publishing.
OECD/ERIA (2018) SME Policy Index: ASEAN 2018: Boosting Competitiveness and Inclu
sive Growth. Paris, Jakarta: OECD Publishing, Economic Research Institute for ASEAN
and East Asia.
OICA (The International Organization of Motor Vehicle Manufacturers) (2015) ‘Passenger

Car Production Statistics’, available at http://www.oica.net/category/production-sta
tistics/.
Oikawa, Hiroshi (2011) ‘To Be or Not to Be a Supplier to TNCs? An Entrepreneurial Ap

proach to Linkage Formation in the Malaysian Electronics Industry’, in Momoko Kawaka
mi and Timothy J. Sturgeon (eds) The Dynamics of Local Learning in Global Value Chains.
London: Palgrave Macmillan, pp. 136–66.
Padilla, Miguel A. E., Dyah W. Sari, and Rossanto D. Handoyo (2017) ‘Formation of Pro
duction Networks in ASEAN: Measuring the Real Value-added and Identifying the Role of
ASEAN Countries in the World Supply Chains’, Business and Economic Horizons 13(2):
237–55.
Pagano, Ugo (2014) ‘The Crisis of Intellectual Monopoly Capitalism’, Cambridge Journal
of Economics 38(6): 1409–29.
Page 24 of 31
of ASEAN
Pettman, Simon (2013) ‘Standards Harmonisation in ASEAN: Progress, Challenges and
Moving beyond 2015’. ERIA Discussion Paper Series 30. Jakarta: ERIA.
Phungtua, Wilawan (2017) ‘The Impacts of Japanese MNCs and Foreign Direct Invest
ment on Thailand’s Automotive Industry’, EAU Heritage Journal Social Science and Hu
manities 7(2): 14–21. (p. 235)
Pietrobelli, Carlo and Roberta Rabellotti (eds) (2006) Upgrading to Compete: Global Val
ue Chains, Clusters, and SMEs in Latin America. Washington, DC: Inter-American Devel
opment Bank.
Pradhan, Rudra Prakash (2009) ‘The FDI-led-growth Hypothesis in ASEAN-5 Countries:

Evidence from Cointegrated Panel Analysis’, International Journal of Business and Man
agement 4(12): 153–64.
Radelet, Steve, Jeffrey D. Sachs, Richard N. Cooper, and Barry P. Bosworth (1998) ‘The
East Asian Financial Crisis: Diagnosis, Remedies, Prospects’. Brookings Papers on Eco
nomic Activity 1998(1): 1–90.
Ravenhill, John (1995) ‘Economic Cooperation in Southeast Asia: Changing Incentives’,

Asian Survey 35(9): 850–66.
Rodrik, Dani (2004) ‘Industrial Policy for the Twenty-first Century’, CEPR Discussion Pa
pers No. 4764. London: CEPR.
Rodrik, Dani (2008) ‘Normalizing Industrial Policy’. Working Paper No. 3. Washington,
DC: Commission on Growth and Development.
Rossi, Arianna (2013) ‘Does Economic Upgrading Lead to Social Upgrading in Global Pro
duction Networks? Evidence from Morocco’, World Development 46: 223–33.
Sanjaya, Lall (2004) ‘Reinventing Industrial Strategy: The Role of Government Policy in
Building Industrial Competitiveness (No. 28)’. Geneva: United Nations Conference on
Trade and Development.
Slaughter, Matthew J. (2002) ‘Does Inward Foreign Direct Investment Contribute to Skill
Upgrading in Developing Countries?’. Working Paper 2002-08. New York: Schwartz Cen
ter for Economic Policy Analysis.
Stiglitz, Joseph E. (1996) ‘Some Lessons from the East Asian Miracle’, World Bank Re
search Observer 11: 151–77.
Sturgeon, Timothy and Richard Florida (2000) ‘Globalization and Jobs in the Automotive
Industry’, Final Report to the Alfred P. Sloan Foundation. International Motor Vehicle Pro
gram, Center for Technology, Policy, and Industrial Development, Massachusetts Institute
of Technology.
Page 25 of 31
of ASEAN
Sturgeon, Timothy, and Johannes Van Biesebroeck (2011) ‘Global Value Chains in the Au
tomotive Industry: An Enhanced Role for Developing Countries’, International Journal of
Technological Learning, Innovation and Development 4(1): 181–205.
Sturgeon, Timothy J., Jack Daly, Stacey Frederick, Penny Bamber, and Gary Gereffi (2016)
‘The Philippines in the Automotive Global Value Chains’. Durham, NC: Duke University
Center on Globalization, Governance and Competitiveness.
Sturgeon, Timothy J., Johannes Van Biesebroeck, and Gary Gereffi (2007) ‘Prospects for
Canada in the NAFTA Automotive Industry: A Global Value Chain Analysis’. Report for In
dustry Canada, Ottawa.
Suriyamongkol, Marjorie L. (1988) Politics of ASEAN Economic Co-operation: The Case of

ASEAN Industrial Projects. New York: Oxford University Press.
Taglioni, Daria and Deborah Winkler (2014) ‘Making Global Value Chains Work for Devel
opment’. Washington, DC: World Bank Group.
Traub-Merz, Rudolf (2017) The Automotive Sector in Emerging Economies: Industrial

Policies, Market Dynamics and Trade Unions: Trends and & Perspectives in Brazil, China,
India, Mexico, and Russia. Berlin: Friedrich-Ebert-Stiftung, Global Policy Development.
UNIDO (2019) INDSTAT 4 Industrial Statistics Database at 3- and 4-digit level of ISIC Re
vision 3 and 4. Vienna: UNIDO. (p. 236)
Wade, Robert H. (1998) The Asian Debt-and-development Crisis of 1997–? Causes and
Consequences’, World Development 26(8): 1535–53.
Yamaguchi, A. (2018) ‘Global Value Chains in ASEAN’. Institute for International Mone
tary Affairs, Newsletter 1.
Yoshimatsu, Hidetaka (2002) ‘Preferences, Interests, and Regional Integration: The De
velopment of the ASEAN Industrial Cooperation Arrangement’, Review of International
Notes:
(1) The ten countries of ASEAN are Singapore, Brunei Darussalam, Indonesia, Philip
pines, Malaysia, Thailand, Vietnam, Laos, Myanmar, and Cambodia. ASEAN member
states have been in an ongoing dialogue regarding the accession of Timor-Leste, or East
Timor, as the eleventh member. It has been more than a decade (March 2011) since Tim
or-Leste applied for formal membership to ASEAN.
(2) For example, the ASEAN Free Trade Agreement (AFTA) and its primary mechanism
ASEAN Trade in Goods Agreement (ATIGA) have ensured the realization of the free flow
of goods within ASEAN, including tariff liberalization, removal of non-tariff barriers, trade
facilitation, customs, standards and conformance, and other measures such as sanitary
and phytosanitary. The inception of AFTA provides a clear timeline for the tariff reduction
Page 26 of 31
of ASEAN
schedule for each product by 2015 and has since promoted the export performances of
ASEAN member states.
(3) ASEAN member states are divided in their support for China, especially with the ongo
ing dispute over the South China Seas between China and ASEAN claimant states (includ
ing Brunei Darussalam, Malaysia, the Philippines, Vietnam, and more recently Indonesia).
(4) Real GDP growth rates have slowed to 4.6 per cent in 2016.
(5) On average, ASEAN’s domestic value added as share of export was 35 per cent in
2013.
(6) The share of foreign value added of exports showed little to no change from previous
years.
(7) The domestic value-added is composed of three parts: the domestic value-added em
bodied in the final demand or intermediate goods consumed directly by the importing
economy; domestic value-added sent to third parties (forward GVC participation); and do
mestic value-added that are sent back to country of origin used to produce exports.
(8) ASEAN inter governmental interventions exist to promote trade and changes to re
gional and global production structures (Arfani, 2017: 79).
(9) The leaders of ASEAN are referred to as the original founding members or the
ASEAN-5.
(10) A number of leaders made direct remarks to defend their national interests against
the growing threat of communism. This includes statements made by Thanat Khoman of
Thailand in 1967 about the need to counter ‘revived germs of an old disease—imperialism
—which are still being cultured in [a] large area of mainland Asia and are threatening the
spread to neighbouring lands’ and the statement by Narciso Ramos of the Philippines,
which declared that ‘the time has come for a truly concerted struggle against the forces
which arrayed against our very survival in these uncertain and critical times’ (Broinowski,
1982).
(11) Indonesia, and to some extent Malaysia, was particularly vital in voicing the principle
of non-alignment within ASEAN. Indonesia’s colonial past played a crucial role in shaping
its views and approach to international affairs, which is an ‘independent and active’ for
eign policy or posture that is not subservient to the interest of major powers (including
the west and China).
(12) With the exception of Singapore, which had relatively higher manufacturing exports
(less than 10 per cent) than its ASEAN neighbours. Indonesia did not record significant
manufacturing exports prior to the 1970s.
(13) China and Taiwan are relatively successful in implementing this model of industrial
ization, particularly in creating special economic zones for electronic exports. Later,
Page 27 of 31
of ASEAN
Southeast Asian economies followed this model of industrialization with Singapore taking
the lead.
(14) For some countries in the region, the financial crisis turned into a political crisis.
President Suharto of Indonesia was forced to step down after thirty-two years of authori
tarian rule due to domestic and political turmoil in many parts of the country.
(15) Capital controls are established to regulate financial flows from capital markets into
and out of a country’s capital account. These controls can be economy-wide or specific to
a sector or industry. Government monetary policy can impose capital control. They may
restrict the ability of domestic citizens to acquire foreign assets, referred to as capital
outflow controls, or foreigners’ ability to buy domestic assets, known as capital inflow
controls. Tight controls are often found in developing economies where the capital re
serves are lower and more susceptible to volatility.
(16) A great deal of scholarly work has been written about the 1997 Asian financial crisis.
Refer to Wade (1998), Radelet et al. (1998), and Goldstein (1998) for a more detailed dis
cussion.
(17) Financial restrictions, like capital controls, are becoming a more accepted policy tool
among policymakers and think tanks. For example, the IMF has been flexible in advocat
ing capital controls as a policy alternative to contain excessive financial speculation.
(18) The ASEAN Summit is held annually by the ten member states to discuss the
progress and agreed commitments on political, security, and economic cooperation.
(19) Tariff reduction of up to 5 per cent by 2010 was a commitment made by the original
five ASEAN member states—Indonesia, Thailand, Malaysia, Singapore, and the Philip
pines. The remaining CMLV countries, which includes Cambodia, Myanmar, Laos, and
Vietnam, were given additional time to implement the tariff reduction rates.
(20) The ASEAN Economic Community (AEC) goal was initially targeted for 2020 but was
accelerated to 2015 by ASEAN leaders during the 12th ASEAN Summit in January 2007.
(21) The ASEAN Economic Community (AEC) blueprint outlines the principles for greater
connectivity between member states and ASEAN’s participation in regional and global
supply chains (Pettman, 2013; Padilla, Sari, and Handoyo, 2017). Moreover, the Initiative
on ASEAN Integration (IAI) provides guidelines to fast-track the building of infrastructure
as well as the legal framework in areas such as broadband connections, e-commerce,
technical skills, and many others.
(22) The ASEAN Trade in Goods Agreement includes improvements in a number of techni
cal areas such as disciplines on Technical Barriers to Trade (TBT), Sanitary and Phytosan
itary (SPS) Measures, and the Temporary Modification and Suspension of Concessions,
which outlines some technical guidelines to compensate losses that arise from any modifi
cation to changes from existing commitments. Refer to the ASEAN Guidelines on Stan
Page 28 of 31
of ASEAN
dards, Technical Regulations, and Conformity Assessment Procedures (STRACAP) for a
detailed assessment of technical cooperation.
(23) Comparative advantage is a theoretical tool for determining a country’s production

and trade specialization. However, a strict interpretation of comparative advantage may
understates the analysis for industrialization and economic development, say from agri
culture to manufacturing and consequently from manufacturing to services. The debate
surrounding the importance for countries to rely on or defy its comparative advantage
has shed light on the importance of industrial policy and the role of institutions in coordi
nating resource maintenance and production output (Lin and Chang, 2009).
(24) AICA was formulated with the close cooperation of member state governments and
the private sector. The proposal was put forward by the ASEAN Automotive Federation,
which was originally made by the Ford Motor Company in 1971.
(25) The preferential margins are the absolute difference in the preferential rate of duty
between the most-favoured nation and the duty for like products. The preferential mar
gins are extendable up to four years with a maximum extension of eight years extension.
(26) China has been a member of the World Intellectual Property Organization since 1980
and acceded to the Paris Convention for the Protection of Industrial Property in 1984,
however IPR infringements and copyright violations are commonly reported.
(27) The two provisions on intellectual property rights are formulated directly by the
ASEAN Working Group on Intellectual Property Cooperation (AWGIPC) and coordinated
by ASEAN Secretariat.
(28) Regional firms with ‘unicorn’ status include Grab (Singapore), Gojek (Indonesia),
Lazada (Singapore), Traveloka (Indonesia), VNG (Vietnam), and Revolution Precrafted
(Philippines).
(29) ASEAN’s undemocratic regimes and their close association with Japanese multina
tional firms, like Toyota, have been fostered since the 1960s and 1970s. This made it pos
sible for the state to single-handedly manage the process of industrialization.
(30) The newly updated action plans include Strategic Action Plan 2016–2025 for ASEAN
Taxation Cooperation, the ASEAN Work Programme on Electronic Commerce 2017–2025,
and the AEC 2025 Trade Facilitation Strategic Action Plan. These updated action plans
serve as a single reference and key action lines towards achieving the ASEAN economic
integration agenda 2016–2025.
(31) The ASEAN Roadmap for Integration codifies ASEAN policies on industrial competi
tiveness by way of providing mutual technical assistance, especially for the least-devel
oped economies such as the CMLV (Cambodia, Myanmar, Laos, and Vietnam).
Page 29 of 31
of ASEAN
(32) Buyer-led is when large retailers and global brands contract the entire process of pro
duction, including designs and production, to the supplier on a short-term contractual ba
sis.
(33) Industries that value highly intellectual property, like the automotive industry, are
good examples of producer-led GVCs. ASEAN’s buyer-led GVCs may extend the patterns
of production towards a low value-added manufacturing export rather than shifting to
wards producer-led GVCs, which would require significant investments and access to fi
nancial resources. There is a big gap in the distribution of these resources among ASEAN
member states (Chang, 2011; Kadarusman, 2016). With the current policy initiatives,
ASEAN suppliers will probably face power asymmetries within the GVC governance.
(34) This figure includes employment for the production of parts and components and as
sembling. The lack of disaggregated employment data makes it difficult to assess employ
ment compositions. This raises an important question regarding local job improvements
from increased participation in global production networks or global value chains.
(35) In North America, vehicle design, development, and specific parts and production are
concentrated in the United States and Canada (as well as the southern part of the US bor
der) in the business of assembling (Sturgeon et al., 2007). The North American automo
tive industry is shifting towards sub-assemblies of parts of components, supplying not on
ly in its region but also to the rest of the world. Thailand and Indonesia assemble most of
the vehicles for East and South East Asia.
(36) The automotive industry in Thailand has strategic importance of 12 per cent to the
country’s economic development and accounts for 12 per cent of the GDP in 2016 and is
part of the country’s economic developmental agenda.
(37) The value-added per worker is computed by taking the total value added from the au
to industry divided by the number of employees.
(38) Harvey (2007) argues that the spread of neo-liberalism in the 1980s played a central
role in modern economic processes towards the creation of a free market economy. Har
vey underscores how the role of the state was to facilitate that freedom and protect capi
tal and private property.
(39) The Heckscher-Ohlin model of international trade stipulates that a country’s exports
hinge on their comparative advantage and abundant factor endowments. For instance, de
veloped countries are abundant in capital- and/or skilled-intensive production and devel
oping countries are abundant in labour- and/or low-skilled production. Following the
Heckscher-Ohlin model, the direction of trade would be that developed countries export
capital-intensive goods/services and developing countries export labour-intensive goods/
services.
(40) The GVC is a series of activities needed to turn raw materials into finished products
and sell on the value-added at each node of the production processes (Gereffi and Kaplin
sky, 2001; Kaplinsky and Readman, 2005). GVCs can be argued to be the practical appli
Page 30 of 31
of ASEAN
cation of World System Theory that predicted a global production dependency among
core or advanced countries on the one hand and periphery countries on the other.
(41) This includes improvements in human resources and skills-upgrading, and increased
investment in infrastructure and R&D.
Budi A. Djafar
Budi A. Djafar is a PhD candidate in economics at the New School for Social Re
search. His dissertation investigates the implications of international trade and the
rise of global production networks for national income and gender-based employ
ment in the manufacturing sector in Indonesia from 1970 to 2010. Budi previously
served in the Ministry of Foreign Affairs of the Republic of Indonesia and the Presi
dential Advisor’s Council of the Republic of Indonesia, 2010‒14.
William Milberg
William Milberg is dean and professor of economics at The New School for Social Re
search and director of the Heilbroner Center for Capitalism Studies at The New
School. His research focuses on the relation between globalization, income distribu
tion, and economic growth, and the history and philosophy of economics. His most
recent book is Outsourcing Economics: Global Value Chains in Capitalist Develop
ment, co-authored with Deborah Winkler. He has served as a consultant to UNCTAD,
ILO, WTO, and the World Bank. He is a member of the Advisory Board for the Center
for Business and Human Rights at New York University.
Page 31 of 31
Managing Trade through Productive Integration: Industrial Policy in an In
terdependent World
Managing Trade through Productive Integration: In

dustrial Policy in an Interdependent World
Richard Kozul-Wright and Piergiuseppe Fortunato
This chapter reviews the debates around trade and industrial policy and discusses how
the composition of trade and investment flows, as well as the spread and form of partici
pation in GVCs, affects structural transformation. It focuses on three characteristics that
have been identified in the literature as critical to assessing the export structure of an
economy and its potential to accelerate industrialization: the diversification of produc
tion, the level of sophistication of the exported products and upgrading of productive ca
pacities/capabilities required to sustain the production and export of increasingly sophis
ticated goods, and the establishment of linkages within and across sectors. The chapter
also discusses the critical components of a national export strategy which could support
the insertion of national firms in international markets, adopt a strategic approach to at
tracting FDI and enable constant upgrading along global (and regional) value chains. Be
cause, success comes not simply from shifting resources from primary activities to
labour-intensive manufactures but also anticipating future challenges in these industries
(as costs rise and new competitors emerge) and nurturing new linkages and more sophis
ticated products. An effective national export strategy depends on active industrial poli
cies, targeted support for upgrading, and regional economic arrangements.
Keywords: diversification, sophistication, linkages, productive integration, global value chains
9.1 Introduction
THE links between trade and development have been a perennial feature of clashes over
economic policy since the Industrial Revolution, if not before. A strict interpretation of
the comparative advantage story tends to contrast a desirable policy of ‘free trade’ and
unbridled competition with distortionary interventions in support of favoured industrial
sectors (Becker, 1985), prone to ‘government failures’ (Krueger, 1990); looser interpreta
tions of that story open up possibilities for a more engaged discussion (Krugman, 1986;
Lin, 2010), albeit with much hand-wringing over ‘picking winners’ (Harrison and Ro
Page 1 of 34
terdependent World
driguez-Clare, 2009). However, there is more to trade and development dynamics than
can be extracted from comparing factor endowments.
Taking a macroeconomic focus changes the terms of the debate; exports can provide a
‘vent’ for domestic production surpluses and a source of external demand while imports
can help correct supply-side shortfalls even as they leak domestic demand; the resulting
examination of economic imbalances necessarily links trade to the balance-of-payments
constraint and related financing issues, as well as focusing attention on the structure of
global markets—through both the negotiation of trade rules and the power of large firms
that dominate international trade—and raising the possibility of ‘unequal’ or at least ‘un
equalizing’ exchange (Singer, 1950; Emmanuel, 1972). Moreover, the dependence on key
imports at different stages of the development process—particularly technology, capital,
and intermediate goods—highlights possible structural (p. 238) obstacles to integrating
successfully into the global economy, and in the process linking the discussion of trade to
industrial policy.
Introducing an historical and comparative dimension to the debates on trade and industry
adds institutional and behavioural details missing from much of the conventional trade
discussion. Exporting, for example, is seen by some as facilitating entrepreneurial and
other firm-level capabilities through buyer–seller links and competitive pressures while at
the same time giving rise to ‘first-mover advantages’ that can stymie competitive pres
sures (Gomory and Baumol, 2000). More generally, international trade can heighten dis
tributional conflicts, both within and across countries (Harrison et al., 2016), and intro
duce geo political pressures into the workings of the global economy (Findlay and
O’Rourke, 2008). These matters necessarily shift the attention on the critical role of the
state in managing trade relations (Caporaso, 1981; Storm, 2017) and that role has re
ceived particularly close attention in the success stories of East Asia, reinforcing, accord
ing to some, the advantages of export-oriented industrialization over import-substitution
industrialization in devising a national development strategy but, on other counts, broad
ening the policy challenges arising from efforts to create ‘dynamic advantage’, including
through the rapid mobilization of resources for accelerated capital formation and the es
tablishment of a strong nexus between profits, investment, and exports (UNCTAD, 1994;
Krugman, 1994).
Comparative studies have also revealed how differences at the firm level and in the orga
nization of production can influence trade relations. With international trade traditionally
dominated by large firms (Bernard et al., 2007), how these emerge, their links with small
er enterprises, and the wider impact of their economic power and influence provide a
critical part of the trade and industrialization narrative. Conventional trade models, con
structed around a bias for perfect competition, have been amended by introducing
economies of scale and scope. However, other features of a trading system dominated by
large firms, including, for example, the role of financial markets (themselves prone to
scale effects) and the presence of rent-seeking behaviour, have been less adequately re
searched (Buckley, 1996). A more descriptive literature has examined some of these is
sues through the lens of global value chains (GVCs) which have become a more promi
Page 2 of 34
terdependent World
nent feature of international trade over recent decades. Establishing links in these chains
has, on some accounts, made it easier to begin an industrialization drive in poorer coun
tries (Baldwin, 2016). However, on other accounts, stronger intellectual property rights
and weakened labour laws have further shifted the balance of power towards the multina
tional firms leading these chains, not only further strengthening their dominant market
positions and ability to generate high profits but also stalling the industrialization process
in developing countries (Gereffi et al., 2001; Phillips and Henderson, 2009; Kozul-Wright
and Fortunato, 2015).
The chapter is organized as follows. Section 9.2 critically discusses the role played by
trade integration in the process of economic development, insisting that trade is a means
not an end and that the potential gains from the effective management of trade involves
more than specialization. Section 9.3 focuses on productive integration, (p. 239) analysing
the channels through which the slicing of the value chains across different economies af
fects the structure and composition of exports. In this context, the section discusses the
emergence of global value chains (GVCs) and presents a simple mapping exercise de
signed to locate different developing economies in the ‘GVCs space’ along two critical di
mensions, the extent and the typology of participation in production networks. Section 9.4
analyses some factors which are of critical importance when it comes to policy design in
an interdependent world economy: the difference between active and passive policy
stance, the challenges of product upgrading along GVCs, and the relevance of regional
value chains (RVCs). Section 9.5 offers some concluding remarks.
9.2 The Multifaceted Connections Between

Trade Integration and Development
9.2.1 Gains from Trade?
Most successful economies, starting from a relatively specialized primary structure, have
moved resources into higher productivity manufacturing activities which, along with ac
cess to low-cost inputs, has facilitated access to foreign markets; subsequently more sec
tors are added, along with more economic linkages within and across sectors, including
with more sophisticated service activities which are initially located inside manufacturing
firms but subsequently become the basis for more independent sectors. In the process,
successful economies come to rely increasingly on intangible assets, scale economies, and
learning to generate further increases in productivity and further gains from internation
al trade. At higher levels of income, these economies begin to specialize once again, but
now in service activities. Strong empirical support associates this pattern with technolog
ical progress, the rise of a large middle class, and the progressive development of finan
cial markets.
These components of successful development have persisted over the last fifty years even
as the world economic landscape has changed considerably. During this period, growth in
world trade has consistently exceeded growth in global output and today’s global econo
Page 3 of 34
terdependent World
my is a much more open and interconnected space than it used to be (UNCTAD, 2016).
Furthermore, the relative importance of developing countries in world trade has been ris
ing—steadily from the early 1970s, more sharply after the early 1990s—reaching between
30 and 40 per cent of the total. The composition of world trade has also changed signifi
cantly, shifting away from primary products (particularly metals, minerals, and fuel) with
a corresponding rise in the share of manufacturing and services, a shift that has also
been pronounced in developing countries.
This evolving landscape is on some accounts the inevitable outcome of a rules-

(p. 240)
based liberal world economy in which countries have adhered to their comparative advan
tage and where economic gaps are destined to become smaller and smaller (Baldwin,
2014). For standard economic theory, specializing according to comparative advantage al
lows countries to reap efficiency gains from moving to a production and exporting profile
that concentrates on exploiting their relatively abundant resources and importing goods
that embody relatively scarce resources. Even countries lagging behind in all sectors
would benefit by following this path (Clarida and Findlay, 1992) and, assuming markets
remain competitive, more mobile capital, particularly in the form of direct investment,
should ensure strong gains for developing countries.
On this basis, the policy conclusion is, invariably, rapid trade liberalization along with
complementary measures to deregulate domestic markets (particularly the labour mar
ket) and remove other government measures that ‘distort’ prices and stymie international
competition. A vast academic industry has sprung up employing sophisticated models and
statistical techniques which purport to prove a direct link from trade liberalization
through increased trade flows to rapid economic growth, poverty reduction, and im
proved societal welfare. Trade simulation exercises, which seem to multiply whenever
multilateral or regional trade negotiations are getting underway (or getting stuck), offer
headline-grabbing multi-billion dollar gains from trade liberalization, often accompanied
by finger wagging at those who are reportedly stalling or opposing the process.
While the argument is much admired for its mathematical elegance, it rests on a set of se
verely restrictive assumptions whose distance from reality has troubled generations of
economists beginning with Adam Smith, no less, who insisted that a universal system of
free international trade was more a utopian ideal than a coherent blueprint for policy, and
that the costs of adjusting to it required that it be done ‘only by slow gradations, and with
a good deal of reserve and circumspection’ (cited in Panic, 1988: 124). Smith’s conclusion
followed from his observation of a world of underutilized and misused resources in which
lagging regions faced a set of unfavourable circumstances.1 Smith also recognized, that
in this world, and whatever the causes of the initial gaps among countries, the free move
ment of goods, technical know-how, and factors of production could give an advantage to
those who opened up an early lead, allowing them to maintain or extend that lead at the
expense of later developers.2
While Smith’s observations have been obscured by conventional trade models, de
(p. 241)
velopment economists and economic historians have long recognized that policymakers
Page 4 of 34
terdependent World
do not operate in a world populated by small firms, with perfect information about con
sumer tastes and available production technologies, where learning or scale economies
are absent, and immobile factors of production fully employed (Toner, 1999). This has
cautioned against rapid trade liberalization, with economic divergence just as plausible
an outcome from such moves as narrowing income and technology gaps, with even more
conventional economists acknowledging that, under some circumstances, ‘the roulette
wheel of evolving comparative advantage’ can lead to ‘genuine harm’ (Samuelson, 2004:
142).
In reality, trade takes place among countries at very different levels of development and
is shaped by a variety of forces other than factor endowments, including economic struc
ture, firm size, technological learning, and power imbalances. This makes for a much
more uncertain trading environment than allowed for in standard trade models and im
plies that trade policy will be poorly designed if it is focused on achieving some single
best outcome rather than on making tangible gains from an evolving set of trading rela
tions among partners of varying degrees of economic power and sophistication.
In this world, trading possibilities are created rather than given and the adjustments to
opening up can take time and have significant distributional effects, historical accidents
have long-run economic consequences, and ‘market forces do not select a single, prede
termined outcome, instead they tend to preserve the established pattern, whatever that
pattern may be’ (Gomory and Baumol, 2000). Indeed, as Gomory and Baumol insist, given
that the modern trading system is so different from the implicit eighteenth-century histor
ical setting of the free-trade model, the analysis of how trade works needs to start from a
very different set of stylized facts. Consequently, a ‘win–win’ outcome is just one among a
range of possibilities in a more open trading system and highlight that international mar
ket forces, in conjunction with varying national capabilities, can produce results that are
beneficial for some but detrimental to others.
Not surprisingly, the idea of a simple positive association between trade and economic
growth is still contested (Kim, 2011; Rodrik, 2017). The vague definition of openness, and
the failure to separate episodes of export promotion from those of import liberalization,
can easily lead to the misrepresentation of trade regimes, making it difficult to draw
meaningful cross-country comparisons and correctly interpret the findings. Cross-section
al averages hide country-specific differences and breaks in the series. Furthermore, sev
eral studies have found that results reporting a strong link between openness and growth
may be sensitive to the specific trade measures employed (Pritchett, 1996), the introduc
tion of additional control variables (Levine and Renelt, 1992), cyclical factors, and peri
odization (Wacziarg and Welch, 2003). It also needs to be stressed that even when a close
statistical association between trade and growth is found, this still leaves open the ques
tion of the direction of causation, with plenty of reasons to suppose that it runs from do
mestic success in raising productivity to increased trade, and not the reverse (Clerides et
al., 1998).
Page 5 of 34
terdependent World
From a policy perspective, rather than assuming a positive association between
(p. 242)
trade and growth, what is needed is a closer examination of the channels through which
the two variables might be related. From an industrial policy perspective, it is of particu
lar interest to explore how trade (both exporting and importing) can affect productivity
growth via its impact on aggregate patterns of structural change and diversification.
9.2.2 Trade Composition, Structural Transformation, and Diversifica

tion
While there is considerable disagreement over how to weigh up the potential losses from
trade and even more on what to do about it, there is general agreement that what a coun
try produces and exports matters, that is, whether it exports apples or iPhones is impor
tant for its development process and prospects. This was the lesson drawn by the first
catch-up economy, the United States, under the guidance of Alexander Hamilton who, as
its First Secretary of the Treasury, ruptured its agrarian destiny and actively pursued a
pro-industrialization agenda through high tariff barriers, big infrastructure projects and
support for technological learning (Cohen and de Long, 2016)—a strategy reproduced,
with local variation, in the European late industrializers of the late nineteenth and early
twentieth centuries.
As a new post-war international division of labour took shape, reinforcing the advantage
of industrial as opposed to primary producers, development economists embraced the
idea that trade could generate polarized outcomes between ‘centre and
periphery’ (Prebisch, 1950) but also, by privileging certain activities, locations, and firms,
could further fracture the international division of labour as large international firms with
tight control over finance and knowledge were able to reinforce their dominant market
positions at the expense of weaker players (Hymer, 1972; UNCTAD, 2018). Whether
through polarization or fragmentation pressures, the resulting room for slippages be
tween trade and development implied that even if countries were able to increase ex
ports, and even achieve higher shares in global trade, they might still fail to diversify
their economies and achieve broader development goals like poverty reduction, increased
employment, improved welfare, etc. As a result, policymakers in many developing coun
tries looked for new ways to reconnect trade to diversification and wider development
challenges through more active measures to transform the underlying structure of their
economies.
Structural transformation refers to the movement of labour, capital, and other productive
resources from low-productivity to higher-productivity economic activities within and
across sectors. By expanding destination markets and realizing scale economies, interna
tional trade can favour the development of higher productivity sectors and the shift of
labour and other resources in their direction. While the opportunities are not exclusive to
manufacturing, the presence in that sector of various productivity-enhancing forces has
given it a privileged place in the discussion on trade (p. 243) and development (Prebisch,
1950; Toner, 1999; Cherif and Hasanov, 2019). Moreover, exporting manufactures can not
only foster a productivity spurt within that sector, it can also raise an economy’s aggre
Page 6 of 34
terdependent World
gate productivity by redistributing existing resources across a broad range of economic
sectors, bringing dynamic gains as the access to better technologies fosters skills-upgrad
ing and other positive externalities.
However, positive outcomes are not predetermined; when there is surplus labour, strong
import competition, or the exit of less productive firms, trade liberalization can result in
declines in aggregate (economy-wide) productivity even as it raises productivity in the in
dustrial sector or among trading firms (McMillan and Rodrik, 2011). The net impact ulti
mately depends on wider employment dynamics and on whether the productivity growth
in industry is outweighed by a larger shift of labour and resources into low productivity
work outside the sector. Evidence of such shifts underlies concerns about weak industrial
ization (including premature de-industrialization) in the developing world in recent
decades (UNCTAD, 2003, 2016; Felipe et al., 2014; Rodrik, 2016; Tregenna, 2009).
From this perspective, structural transformation is less a one-off adjustment and more a
continuous process and the attendant policy challenges vary, inter alia, with a country’s
level of income, the structure and sophistication of its productive base, the size of its
firms and their technological capacities, and the fiscal space to manage the employment
challenge (UNCTAD, 1996, 2003, 2006, 2016). Each level of economic development is a
point along the continuum from a low-income agrarian economy, where most of the out
put and labour are concentrated in agriculture, to a high-income economy, where the
lion’s share of production and labour accrues to manufacturing and services. The struc
ture of the economy continuously changes as technological change leads it to upgrade to
more and more sophisticated goods and production methods. This involves both a pro
gressive diversification of the production base and an upgrade of the goods produced
within each industry. Diversification is therefore the key economic development challenge
to which industrial and trade policy must adapt accordingly.
The critical importance of diversification, or horizontal evolution of production, has been

underscored by the findings of Imbs and Wacziarg (2003). Examining sectoral concentra
tion in a large cross-section of countries, they document an important empirical regulari
ty: as poor countries get richer, sectoral production and employment become less concen
trated, that is, more diversified. Such a diversification process goes on until relatively late
in the process of development. This highlights another potential channel of connection
between trade liberalization and development; by favouring the access to international
markets, openness can help to overcome domestic demand constraints and may therefore
facilitate the horizontal evolution of production. On the other hand, an excessive or pre
mature liberalization, taking place when the economy still lags very far behind the fron
tier in terms of productive capacities, might instead foster an overspecialization on natur
al resources or low-skilled intense manufacturing products according to the logic of the
comparative advantage.
Structural transformation can also proceed through the gradual process of mov
(p. 244)
ing towards higher value-added and more productive activities and the increasing sophis
tication of the goods produced. Empirical evidence has demonstrated that countries that
Page 7 of 34
terdependent World
have managed to upgrade their productive structures and export more sophisticated
goods have grown faster. Hausmann et al. (2007) develop a quantitative index of coun
tries’ export sophistication generally denoted as EXPY. Unsurprisingly, the authors show
that this measure of export sophistication is highly correlated with per capita income. But
what is important from our perspective is that they also show the existence of a positive
correlation between the initial level of EXPY and the subsequent rate of economic growth.
That is to say, if a country has a sophisticated export basket relative to its level of income,
subsequent growth is much higher (see also Fortunato and Razo, 2014). It is telling that
China and India, among the most successful economies in the recent past, had in 1992
more sophisticated export profiles than their income levels might have suggested.
Along with diversification and upgrading, the third component in the transformation nar
rative is linkages. The immense appeal of manufacturing lies in its potential to generate
productivity and income growth (Kaldor, 1966), and because such gains can spread across
the economy through production, investment, knowledge, and income linkages. Several
linkages deserve mention here. To begin with, expanding production can help build ‘back
ward’ linkages (to source inputs for production), and ‘forward’ linkages in so far as the
produced goods are used in other economic activities (Hirschman, 1958). Intersectoral
linkages emerge as knowledge and efficiency gains spread beyond manufacturing to oth
er sectors of the economy, including primary and service activities (Cornwall, 1977; Tre
genna, 2010). Investment linkages are created when investments in productive capacity,
new entrepreneurial ventures, and the related extensions of manufacturing activities in
one enterprise or subsector trigger additional investments in other firms or sectors,
which otherwise would not occur because the profitability of a specific investment project
in a certain area of manufacturing activity often depends on prior or simultaneous invest
ments in a related activity (Rodrik, 2004). Income linkages emerge from rising wage in
comes generated from industrial expansion; these add to the virtuous cycle through ‘con
sumption linkages’. Income linkages also operate through supplementary government
revenues (i.e. ‘fiscal linkages’), which may therefore expand public expenditure. The cre
ation of such income linkages can strengthen the self-reinforcing aspect of industrializa
tion through increasing domestic demand and therefore GDP growth.
While diversification, upgrading, and linkages frame the structural transformation in gen
eral, what determines whether and in which direction a country transforms its production
structure is country specific and often difficult to identify even ex post. But among the
many variables that influence the outcome of this process, industrial and trade policy
have received particular attention in academic and policy debates. Since countries cannot
produce a good for which they have no knowledge or expertise, deploying policies to fos
ter learning, accumulation of productive capabilities, and technological change becomes
of crucial importance. Trade policy also matters in (p. 245) this context, not only because
access to markets abroad can provide new growth opportunities but also because the
commercial partners and the type of integration strategy pursued may significantly affect
the characteristics of the exported products and the opportunities to further diversify the
economy.
Page 8 of 34
terdependent World
9.3 Productive Integration: The Emergence of

Global Value Chains
For a good deal of the post-war period, policy debates on how best to establish a robust
connection between trade and industry was pursued through the largely misleading di
chotomy between import-substitution industrialization and export-oriented industrializa
tion and a false comparison between free and managed trade. This, by draining the dis
cussion of a specific structural, institutional, and historical context was an unfortunate
turn, because the case for international trade and its implications for growth, employ
ment, and distribution is a subtle one that has always depended heavily on context (Ro
drik, 2011).
Those debates have become, if anything, even less nuanced with the rise of hyper-global
ization marked by large flows of footloose capital and the reorganization of production
around global value chains (GVCs). Advances in information and communication technolo
gies (ICTs) along with extensive market liberalization have made it easier and cheaper for
large international firms to move capital around, including by illicit means, as well as to
manage far-flung production networks, which nowadays account for a rising share of in
ternational trade, global output, and employment (UNCTAD, 2002 and 2014). Participa
tion in these chains by developing countries is expected to attract more foreign direct in
vestment (FDI), provide easier access to export markets, advanced technology and know-
how, and generate rapid efficiency gains from specializing in specific tasks, appropriately
guided by the ‘lead firm’ in the chain. Such participation is seen as particularly important
for developing countries with small domestic markets whose firms confront a range of
technological and organizational constraints stemming from the fact that the minimum ef
fective scale of production often far exceeds that required to meet their prevailing level of
domestic demand.
This has encouraged policymakers to focus on providing an attractive business climate

(including adequate infrastructure and a sufficiently trained labour force) for the lead
firms that manage these chains, avoiding any restrictions on the free flow of goods and fi
nance that connect suppliers along the chain. With the spread of these chains, the foreign
content of exports, or backward participation, which measures the value added generated
outside the country that completes and exports a product, has increased significantly in a
number of developing economies but also in Europe and North America. According to
Timmer et al. (2013), foreign content shares increased (p. 246) for 85 per cent of the prod
uct chains between 1995 and 2008. As a counterpart, forward participation of many
economies, that is, how much domestic value added is embodied as intermediate inputs in
third countries’ gross exports, has also been on the rise; in 2015 it accounted for over 30
per cent in the United States, China, and the European Union.
Such data indicate the pervasiveness of international fragmentation. No longer are prod
ucts simply made in one country and shipped to another for sale. Indeed, products often
go through many stages, traversing several geographic and organizational borders and
Page 9 of 34
terdependent World
adding components and value before they reach their final markets. This is also reflected
in the significant increase in trade in intermediates, whose share of global trade in
creased from 57 per cent in 1995 to a peak of over 63 per cent in 2011 (explaining two-
thirds of the total growth in trade over the period), and in the continuous rise of the glob
al FDI stock, from just over 12 per cent of global GDP in 1995 to over one-third in 2015.3
In recent years, outsourcing and producer-driven value chains have tended to concen
trate especially in capital- and technology-intensive industries such as automobiles, elec
tronics, and machinery. The underlying rationale for this reorientation from the firms’
perspective is straightforward; first, their control of intangible assets (derived from in
vestments in R&D, design, marketing, and branding) are less prone to competition, as
they are based on unique resources and capabilities that other firms find difficult to ac
quire, and are therefore sources of superior returns or rents (Kaplinsky, 2005) and sec
ond, their dominant position as lead firm also gives them a monopsonistic position with
respect to suppliers, squeezing their margins and adding to their own super profits. On
the other hand, fragmentation has made integration into these chains attractive for many
developing countries, who, by becoming niche suppliers of parts of the chain, present an
attainable first step towards building industrial capacity, creating employment, and inte
grating into global trade. Accordingly, policymakers are increasingly turning to integra
tion and upgrading in GVCs as a means of driving economic development but, more often
than not, without the policy tools required to make this happen.
9.3.1 Mapping GVCs
Production is nowadays fragmented across national boundaries much more than it used
to be in the past, and it entails the sourcing of inputs and components from multiple sup
pliers based in several countries. Understanding how this is affecting trade patterns and
wider development prospects relies on an accurate measurement of the intensity of par
ticipation in the GVCs of each individual economy but also on the identification of the ex
tent to which countries are specializing in different stages of the global production
processes.
Since many developing countries have faced difficulties in achieving their development
policy objectives, their place in GVCs has tended to be located on the lower (p. 247) por
tions of what is sometimes referred to as the ‘smile curve’ (Figure 9.1). The smile curve
conceptualizes international production as a series of linked tasks and sees international
trade organized within GVCs as involving trade in those tasks rather than trade in goods.
The resulting fragmentation of production carries significant consequences for the spatial
division of labour and the distribution of economic power and privilege. Most of the pre
production and post-production segments of the manufacturing process, with their higher
return activities, are usually located in advanced economies, with developing countries
often left with the lower value added activities of the production segment of the manufac
turing process.
Page 10 of 34
terdependent World
Figure 9.1 Stylized manufacturing value chain smile

curve
Source: Authors.
To get a more granular measure of the nature of integration of any given country in inter
national production networks, we employ a participation index calculated as the sum of
forward linkages, that is, domestic value added embodied in foreign exports (as a share of
total exports), and backward linkages that is, foreign value added (as a share of total ex
ports). Our measure of upstreaming is taken from Fally (2012) and Antràs et al. (2012)
and is meant to gauge the distance of each specific production sector in a country i from
final demand. The index collects information on the extent to which the industry produces
goods that are sold directly to final consumers or to sectors that themselves sell to final
consumers. A country is specializing in upstream activities if it imports a low share of in
termediates and exports a big share of intermediates to third countries. Upstream activi
ties are, for instance, the production of raw materials, but also intangibles such as re
search and development or the design of industrial products. Formally, we calculate the
total share of products sold for final use over the gross output of each sector and then ag
gregate across sectors to evaluate the upstreamness of country i. We take from Fally
(2012) also our index of downstreaming. The index is meant to capture the distance of a
given sector in country i from the economy’s (p. 248) primary factors of production. Ac
cording to this measure an industry in country i is more downstream when its production
process uses intensively intermediate inputs (rather than primary production factors) and
thus when the value added associated with its production is relatively low. Typical down
stream activities include the assembly of processed products and post-sales customer ser
vices. Formally, the measure is calculated as the complement to unity of the ratio be
tween the value added generated by each sector and its gross output. Following Marel
(2015), the data are subsequently aggregated across sectors to obtain the value of the in
dex for a country i.
We focus specifically on low- and middle-income countries within the set of forty coun
tries covered by the World Input Output dataset (Timmer et al., 2015)4 and look at the
years 2005 and 2011, and to the changes that occurred during the period. We study first
Page 11 of 34
terdependent World
how changes in participation rates are associated to specialization in upstreaming activi
ties (Figure 9.2). On the horizontal axis we measure the intensity of participating in GVCs
(participation index) while the vertical axis measures distance from final demand (up
streamness).
Figure 9.2 Evolution of GVC positioning in selected

developing economies (participation and upstream
ness), 2005–11
Source: WIOD (2013) and OECD TiVA.
All the economies under scrutiny, with the notable exception of China, experi
(p. 249)
enced an increasing participation rate coupled with a shift of the production away from
the final demand, that is, upstreaming in 2011 is higher than in 2005. This reflects both
increased outsourcing in advanced economies and the relative distance from the market
of those productive activities sourced by developing economies. China, on the contrary,
increased its upstreaming but reduced its overall participation in GVCs. This reflects its
increased specialization in intermediate inputs trade, which in turn increases the dis
tance from final demand, and the reduction of foreign value addition in domestic produc
tion.
Figure 9.3 introduces the distance from an economy’s primary factors of production
(downstreamness). We measure downstreamness on the horizontal axis while, once again,
the vertical axis measures distance from final demand (upstreamness).
Page 12 of 34
terdependent World
Figure 9.3 Evolution of GVC positioning in selected

developing economies (downstreamness and up
streamness), 2005–11
Source: WIOD (2013).
Figure 9.3 shows the existence of a strong and positive correlation between the two indi
cators. Countries characterized by more upstream production according to the produc
tion-staging distance from final demand are at the same time closer to primary factors of
production. In other words, economies that sell higher shares of their output directly to fi
nal consumers tend also to display relatively high value added over gross output ratios,
reflecting a limited amount of intermediate inputs embodied (p. 250) in their production.5
This is the case since modern economies tend to specialize alternatively in ‘short’ or
‘long’ value chains. Services are provided through ‘short’ chains with both a high ratio of
sales to final consumers and little use of intermediate inputs in production. Indeed, pay
ments to labour comprise a larger share of the production costs in services industries.
Conversely, manufacturing processing can be more easily fragmented into stages involv
ing separate parts and components. Consequently, manufacturing is characterized by rel
atively ‘longer’ chains with a lower share of output going directly to final use and a more
intense use of intermediate inputs compared to services.
Figure 9.4 highlights the significance of these sectoral differences displaying the relative
position of manufacturing, services, and the primary sector in the BRIC’S economies be
tween 2005 and 2011. In all cases, the services sector falls in the lower left (p. 251) quad
rant, indicating both proximity to the final demand and to the production factors. The
manufacturing sector displays higher variation across the four countries. In line with the
idea of ‘long’ production chains, China’s manufacturing industries are more distant from
the final demand, reflecting their increasing specialization in intermediate inputs. Manu
facturing in India and Brazil, on the other hand, appears closer to final demand, thanks to
their greater dependence on service activities and primary products respectively.
Page 13 of 34
terdependent World
Figure 9.4 Sectoral positioning of BRIC economies,

2005–11
Source: WIOD (2013) and OECD TiVA.
9.3.2 Assessing the Impact of GVCs
The connection between participation in GVCs and development is not a straightforward

one. When the low unit cost that attracts TNCs is largely dependent on low wages rather
than rising productivity, the danger of a race to the bottom is ever present. Moreover,
while the evidence presented in UNCTAD (2016), for example, shows that when increases
in the foreign value added of exports occurs in a larger context of greater production and
exports of manufactures (as in much of the Asian region, for instance), GVC participation
can complement industrialization and structural when increasing participation in GVCs
reflects a reduction of domestic sourcing in a context of weak export performance of man
ufactures, GVC participation may actually delay structural transformation, as in the case
of many developing economies in Africa and Latin America.
Kowalski et al. (2015) analyse the impact of GVC participation on economic upgrading as
measured by sophistication of the export structure. They find that while growing back
ward participation to GVCs, measured using the share of foreign value added in exports,
is associated with product upgrading in high- and middle-income countries, wider frag
mentation does not have the same effects in low-income economies. This heterogeneity
across income groups suggests that the level of economic development must be taken se
riously into consideration when evaluating integration strategies based on GVC participa
tion. Kowalski et al. (2015) also study the determinants of functional upgrading, defined
as the capacity of acquiring new functions within a given value chain. Key determinants
of functional upgrading turn out to be the import of sophisticated non-primary intermedi
ates and the distance from economic poles of activity.
Page 14 of 34
terdependent World
Claims for how GVCs can strengthen productivity (and contribute to economic growth)
are largely based on conventional trade models (OECD, 2013; WTO, 2017). But from a de
velopment perspective, the structure of most GVCs and the distribution of power along
them require a more nuanced analysis. In particular, the conditions that ease access to
GVCs can also act as barriers to upgrading since more accessible parts of the chain are
associated with few forward and backward linkages, limited institutional development,
and little possibility for knowledge externalities in the wider economy. A broad body of ev
idence indicates that only a small number of developing countries— (p. 252) mostly in East
Asia—have been able to build the needed linkages between domestic and foreign firms
and achieve upgrading within GVCs (UNCTAD 2016, 2018) and even with respect to job
creation the impact, in most cases, has been limited (Ingram and Oosterkamp, 2014).
Divergence between expectations and outcomes from participation in GVCs is, in part, a
reflection of the fact that the private interests of international firms do not necessarily co
incide with the developmental interests of the host countries. This disconnect is, of
course, familiar to many developing countries from their participation in commodity-
based value chains, reflecting, in part, the asymmetric structure of markets and the pric
ing power of firms from the North and South. It also highlights the importance of strate
gic policies, as countries look to shift towards a greater reliance on manufacturing (and
service) activities and exports and is an important reminder that reductions in policy
space can hamper industrialization and catching up in late developers (UNCTAD, 2014).
As Stephen Hymer (1972: 101) recognized over forty years ago, as international produc
tion fragments along task lines, ‘output is produced cooperatively to a greater degree
than ever before, but control remains uneven’; in particular, the lead firm tends to con
centrate its own tasks at the two ends of the smile curve where ‘information and money’
provide the main sources of control and where profit margins tend to be higher. These
‘headquarter’ economies are still located predominantly in the North (now including parts
of East Asia) while ‘factory’ economies are, largely, in parts of the South. Indeed, as these
chains have spread across more countries and sectors over the past three decades they
have been accompanied by a more and more uneven distribution of those benefits. In de
veloped countries, the concern is that low and medium-skilled production jobs in tradi
tional manufacturing communities have been ‘outsourced’, first to lower-wage regions of
the developed world and then ‘offshored’ to developing countries, and wages have stag
nated while new jobs created at the ends of the chain have not only been insufficient in
number to replace those being lost, but are often out of reach to those ‘left behind’, both
geographically and in terms of the skills required. The result is socio-economic polariza
tion and a vanishing middle class in advanced economies as well as some emerging
economies.
Developing economies with limited productive capacities can therefore remain trapped
in, and competing for, the lowest value-adding activities at the bottom of value chains,
which can ultimately result in ‘thin industrialization’, weak linkages and slow economic
growth (Gereffi, 2014; UNCTAD, 2006, 2014). Participation in GVCs also carries the addi
tional risk of leading to specialization in only a very narrow strand of production with a
Page 15 of 34
terdependent World
concomitantly narrow technological base and overdependence on multinational enterpris
es (MNEs) for GVC access. Such shallow integration manifests itself in asymmetric power
relations between lead firms and suppliers and in weak bargaining positions for develop
ing countries. For example, the experiences of Mexico and Central American countries as
assembly manufacturers have been likened to the creation of an enclave economy, with
few domestic linkages (Gallagher and Zarsky, 2007). The same can be said about the elec
tronics and automotive industries in Eastern and Central Europe (Plank and Staritz, 2013;
Pavlínek, 2015; Pavlínek and Zenka, 2016). (p. 253) In all these cases, there has been sig
nificant ‘internal upgrading’ within MNE affiliates, but this has involved very few
spillovers to the local economies in the form of productivity improvements and imitation
by domestic firms, partly due to limited linkages of MNEs with local firms and labour
markets (Fons-Rosen et al., 2013; Paus, 2014). Moving up the chain into more capital-in
tensive or higher value-added production is particularly challenging in such an environ
ment, because it necessitates relationships with lead firms at the top that are ultimately
focused on maintaining their profitability and flexibility. Indeed, these firms sometimes in
tentionally use GVCs to induce and intensify competition among suppliers and countries
for their own benefit (Levy, 2005; Phillips and Henderson, 2009).
In a recent paper, Rigo (2017) presents some interesting stylized facts on the extent to
which firms operating in developing countries benefit from GVCs. The author compares
groups of firms along several measures of technology adoption and knowledge creation
(running training programmes, using foreign-licensed technology, possessing quality cer
tifications, and communicating with customers and suppliers via the Internet). He finds
that two-way traders, which are firms typically involved in GVCs, display on average a
higher propensity to adopt new technologies than other groups of firms and that local
suppliers to these firms, that is, non-trading domestic firms involved in upstream opera
tions with them, strongly benefit from technological spillovers. However, he also shows
that foreign-owned firms involved in GVCs are more likely to be dependent on the global
sourcing policies of their parent companies and have generally a low propensity to en
gage in local sourcing. The opportunities for generating local spillovers in developing
countries are therefore to a large extent still unexploited.
South East Asia represents an interesting exception to this trend. Local enterprises in
fact seem to be much more integrated with the MNEs operating in the region. A recent
OECD-UNIDO (2019) report shows that foreign manufacturers in ASEAN source consid
erably from local producers. In Thailand, Lao PDR, Indonesia, Malaysia, and the Philip
pines, foreign MNEs source over 60 per cent of intermediate inputs from firms that pro
duce locally. The average share of local sourcing by foreign MNEs in Vietnam, though
somewhat lower, is still significant.
An emerging literature is trying to assess empirically the impact of the rise of GVCs on
structural transformation by using input–output matrices recently made available by a
number of new databases (e.g. the World Input Output Database and the Trade in Value
Added Database). These studies show that despite the global label, production in value
chains is concentrated in a small number of industries and countries. Lead firms are gen
Page 16 of 34
terdependent World
erally from advanced economies and production tends to be most fragmented in clothing
and textiles, electronics, and automotive industries (De Backer and Miroudot, 2013; Tim
mer et al., 2013; UNCTAD, 2014). Another common finding in this literature is that while
participation of developing countries in GVCs has increased tremendously over recent
decades, firms headquartered in developed economies have been the big winners from
the spread of GVCs (Milberg et al., 2014; UNCTAD, 2002, 2018).
Dedrick et al. (2010) use the examples of the Apple iPod and notebook personal comput
ers to illustrate how profits are distributed between the participants of these two (p. 254)
GVCs. The intuition behind this exercise is relatively straightforward: an iPod and a com
puter are made up of lots of components produced by different firms in different coun
tries. Each of these firms charges a price for its component or activity and in turn pays
other firms for the intermediate goods needed to complete its stage of production. Table
9.1 presents different indicators of profit margins of the main participants in the iPod
global value chain. The table clearly depicts the gap between the profits enjoyed by firms
that specialize in product design (or the production of critical components, such as the
controller chip or the video chip) and firms that specialize in assembly or production of
low-tech standardized components like memory chips.
Page 17 of 34
Managing Trade through Productive Integration: Industrial Policy in an Interdependent World
Table 9.1 Uneven returns in the iPod value chain
Function Supplier Gross margin Operating margin Return on assets
Controller chip Portal Player 44.8 20.4 19.1
Les firm Apple 29.0 11.8 16.6
Video chip Broadcom 52.5 10.9 9.8
Primary memory Samsung 31.5 9.4 10.3
Battery TDK 26.3 7.6 4.8
Retailer Best Buy 25.0 5.3 9.6
Display Toshiba-Matsushita 28.2 3.9 1.8
Hard drive Toshiba 26.5 3.8 1.7
Assembly Inventec Appl. 8.5 3.1 6.1
Distribution Ingram Micro 5.5 1.3 3.1
Minor memory Elpida 17.6 0.1 –1.0
Minor memory Spansion 9.6 –14.2 –9.2
Page 18 of 34
Managing Trade through Productive Integration: Industrial Policy in an Interdependent World
Note: Grey areas evidence the gaps in profit margins between different participants in the iPod global value chain.
Source: Dedrick et al. (2010: 92).
Page 19 of 34
terdependent World
In light of the multifaceted and highly contingent flow from trade to productive integra
tion to economic development, we need to understand under which particular conditions
integration can actually deliver development in a given country and what are the critical
policy challenges to be faced in this respect. This could serve as a guide to better target
industrial and trade policy. This is the objective of section 9.4.
9.4 Managing Integration, Facing the Policy

Challenges Ahead
This section discusses several issues which are facing policymakers on linking productive
development and trade integration: the long-standing debate on targeting, the challenges
of product upgrading along the value chains, and the potential of regional value chains
(RVCs).
(p. 255) 9.4.1 Active versus Passive Policies
The internationalization of production and accompanying changes in the direction and

composition of international trade has not made industrialization easier or lessened the
need for public policy to help build a virtuous circle between productivity growth, in
creased investment, and strategic integration into the global economy. Indeed, in many
economies, there is an even greater need for state involvement, precisely because private
actors on their own face greater uncertainty and higher uninsurable risks in today’s in
creasingly interconnected but unstable global economy.
For many developing countries these concerns are particularly pronounced because, in
addition to having to reverse widening technology and productivity gaps, efforts to catch
up also have to deal with issues relating to global production chains that are under the
tighter control of large international firms, and as they have been subject to increasing
restrictions on national policy autonomy arising from the proliferation of free trade agree
ments and bilateral investment treaties. In such a context, governments in developing
countries must be ambitious without being unrealistic, striving for a high development
road by creating new sources of growth and dynamism, rather than simply trying to do
the best with what they currently have by relying on existing advantages.
Small and incremental steps can be useful (Lin and Treichel, 2014); but more radical
‘comparative-advantage-defying’ measures will be needed to shift towards higher value-
added and employment-generating activities with high-income elasticities and greater
scope for boosting productivity through knowledge creation (Wade, 2015; UNCTAD,
2016). The flip side of aiming high is that failure must also be accepted, but managed,
with mechanisms for monitoring performance and underperformance, leading to a rectifi
cation of the latter or to a removal of state assistance. Accordingly, the focus should be
not on whether to design and implement industrial policy at all, but on how to do it prop
Page 20 of 34
terdependent World
erly (Naudé, 2010), or, as Cohen and DeLong (2016: 23) have put it, ‘getting the political
economy right’.
In the debates around industrial policy, a good deal of attention has been paid to distin
guishing between vertical policies targeted at particular firms, sectors, or activities, and
horizontal industrial policies aiming at general improvements to the wider economic envi
ronment, such as providing transport infrastructure, reliable supplies of energy, and a
sufficiently educated workforce. However, this distinction is somewhat artificial, because,
in practice, even supposedly neutral horizontal policies may have vertical effects by bene
fiting some activities or sectors more than others, depending on the particular character
istics of those activities. Exports of cut flowers, for instance, are facilitated more by infra
structure projects related to air travel, whereas trade in cars and commodities benefits
from the upgrading of sea ports. A policy decision to ease credit restrictions may have an
impact on interest rates in general, but affects particular industries differently, depend
ing on their reliance on such factors as bank credit and degrees of profitability.
(p. 256)No matter how much governments may seek to avoid explicit targeting, even
seemingly universal and undifferentiated policies will have varying effects on different ac
tivities. Since policymakers are ‘doomed to target’, it is better to accept this fact and try
to get the targeting right. In the recent case of China, for example, the state has played a
prominent role in establishing a dynamic ‘profit–investment–export’ nexus through a mix
ture of more general measures, as well as selective and targeted interventions at differ
ent levels, with the mixture changing over time (Knight, 2012). As China shifted towards
a more export-oriented growth strategy in the early 1990s, targeting sectors such as au
tomobiles, semiconductors, and high-speed trains, with public finance pouring into mas
sive investments in infrastructure development. Meanwhile, both state-owned enterprises
and MNEs (often through joint ventures) were encouraged and cajoled into undertaking
industrial upgrading (Lo and Wu, 2014). While China has its own unique features, this is a
familiar policy approach from previous success stories in East Asia (Poon and Kozul-
Wright, 2019).
The mixture of more general and selective measures in less developed countries, such as
in sub-Saharan Africa, will need to be substantially different from more standard industri
al policy packages, since these countries are still predominantly rural, with less devel
oped markets, a smaller industrial base, and weak public institutions. Moreover, the bulk
of non-farm employment is generated in small firms or microenterprises, inter-firm spe
cialization and collaboration are often absent, and economic transactions are strongly in
fluenced by informal institutions that are not necessarily well aligned with the prevailing
norms of market economies. To overcome these constraints and nurture larger and more
competitive enterprises in both industry and agriculture, the state will need to assume a
particularly active role. This will involve raising productivity in the rural economy in par
allel with developing manufacturing activities in urban agglomerates, strengthening inte
gration, and creating linkages among those activities.
Page 21 of 34
terdependent World
The process will likely involve significant investment in boosting the institutional capaci
ties of both the government and the private sector. In this context, it is useful to distin
guish between ‘passive’ and ‘active’ industrial policies. A ‘passive’ industrial policy essen
tially accepts the existing endowments and institutional structures, and aims to reduce
the costs of doing business, including coordination and transaction costs. By contrast, an
‘active’ industrial policy targets deeper changes in corporate structure and strategy, such
as the links between investment, exporting, and upgrading. The institutional prerequi
sites for active and passive policies are likely to be different. In particular, effective tar
geting of active measures requires substantial state capacity and a degree of discipline,
which is an area often neglected in discussions of industrial policy. In practice, while an
active policy is almost always accompanied by a passive policy, the reverse is not the
case.
Clearly, it is not enough simply for governments and businesses to develop a vision and
design targets together; governments must also have some means of ensuring that busi
nesses make the subsequent investments and changes in performance as agreed. Various
ly described as ‘reciprocal control’ (Amsden, 2001) or the ‘support/performance’ (p. 257)
bargain (Evans, 1998), this disciplinary function is essential for industrial policy to suc
ceed, but it has received insufficient attention in much of the renewed discussion on in
dustrial policy (Schneider, 2016; Peiffer, 2012; Kozul-Wright and Poon, 2019).
In the East Asian examples, governments were able to link the application of their policy
tools (such as the provision of lower-cost capital, dealing with the threat of foreign com
petition, or privileged access to scarce foreign exchange) to measurable improvements in
business performance in terms of production efficiency or exports. All certainly saw one
of their principal tasks as that of increasing the supply of investible resources and assum
ing part of the long-term investments. State-sponsored accumulation involved variously
the transfer of land and other assets, efforts to decrease competition in some areas while
increasing it in others, strong regulation and control, and in some cases ownership, of the
financial system and a pro-investment macroeconomic policy, including direct public in
vestment in some lines of activity. Critically, these developmental states did not simply
measure success in terms of increasing investment to fuel economic growth, but also in
terms of guiding the investment into activities that could sustain a high-wage future for
their citizens. This implied a coordinated effort to shift resources from traditional sectors
by raising agricultural productivity and channelling the resulting surplus to emerging in
dustrial activities (Grabowski, 2003; Studwell, 2013). It also meant deliberately reducing
risks and augmenting profits in industries deemed important for future growth (Wade,
1995; Amsden, 2001). Like their late nineteenth- and twentieth-century precursors, this
meant making full use of the creative impulses of global markets, even while protecting
some domestic producers from excessive competition, through strategically guided inte
gration into the international economy.
Building similar relations has proved more difficult in other contexts. In Latin America, a
form of ‘hierarchical capitalism’ (Schneider, 2013) has been associated with undermining
government’s abilities to persuade businesses to transform. From the 1950s onwards, the
Page 22 of 34
terdependent World
big national firms were encouraged to invest heavily in import-substituting industries be
hind protective tariffs and trade restrictions, but policymakers did not impose adequate
performance standards in return for the higher profits earned as a result of these mea
sures (Schneider, 2016; Agosin, 2013). Similarly, during the market reforms of the 1990s,
explicit performance standards were rarely imposed, even where governments structured
privatization programmes to favour particular business groups. Utilities were subject to
the usual sectoral regulations (i.e. for essential services or monopolies) but, according to
Rodrik (2008), overall policymakers in Latin America used too much carrot and too little
stick.
9.4.2 Upgrading and Sophistication
As discussed in section 9.2.2, in order to sustain development a country must be able to

progressively upgrade its production structure and raise internal value added by export
ing (p. 258) more sophisticated and complex manufactured products. But if ‘what you ex
port matters’ for economic growth, the natural question to address becomes what mat
ters for your exports? What type of integration (and value chains) favour the export of
those sophisticated and complex products which foster economic development?
Upgrading in GVCs is crucially affected by the governance structure of value chains (Ger
effi, 2014; Gereffi et al., 2001, 2005). Governance structures depend on firm characteris
tics such as size, crucial for achieving economies of scale and establishing linkages with
global lead firms, and the existing level of capabilities, which determines the potential for
productivity growth and upgrading towards higher-value-added activities and more so
phisticated products. Governance structures influence the impact that GVCs can have on
firms in developing countries by determining the power relations within the chain. When
some players gain too much power in the chain, they might adopt strategies to capture
higher shares of value added. For example, by creating trade-related constraints in the
form of tariffs and other taxes, lead firms in downstream activities can reduce the profit
margins of upstream firms. Alternatively, they might hamper technological upgrading and
entry into downstream activities, for example by limiting knowledge and technology
transfers or by imposing standards through trade and investment agreements (Milberg
and Winkler, 2013; UNCTAD, 2014). These strategies are likely to cement the asymme
tries in power and skills between developed and developing country firms. Governments
in developing countries can help local firms negotiate contracts with foreign firms, for ex
ample, by encouraging long-term contracts between them, supporting collective bargain
ing through producer associations, or providing training in bargaining and model con
tracts (Milberg et al., 2014; UNCTAD, 2013).
Finally, the potential for upgrading also depends on local suppliers and on their capacity
to source intermediate inputs and to acquire, assimilate, and successively exploit the val
ue of information and knowledge coming through the interaction with other firms (ab
sorptive capacity). Participation in GVCs exposes domestic suppliers to multiple interac
tions with foreign firms, thereby offering them unique opportunities of absorption. How
ever, MNEs do not always activate linkages with local firms by preferring international
Page 23 of 34
terdependent World
sourcing strategies. Government efforts to strengthen learning capabilities is likely to
help companies, both for start-ups and older firms, better adapt to the challenges arising
from participation in GVCs.
Recent research shows that technology transfers are more effective when firms possess
previously accumulated knowledge and innovative capabilities. OECD and UNIDO (2019)
document how the gap between foreign firms and local SMEs remains high in some
ASEAN member states and how precisely in these countries the spillovers from local
sourcing become less intense. From a policy perspective it is therefore of paramount im
portance to couple ‘push’ measures which might incentivize local sourcing of parts by
MNEs (e.g. local content requirements) with ‘pull’ policies designed to improve produc
tive capacity and the ability of local suppliers to match the quality of imported compo
nents and intermediate products (e.g. promotion of entrepreneurship through incubators,
training, or support with venture capital and scaling up of domestic capabilities through
technical vocational education and training programmes).
(p. 259) 9.4.3 Regional Value Chains
Another pillar for the design of integration strategies able to foster productive capacity
building and structural transformation is represented by productive regional integration.
As mentioned above, East Asian countries, particularly South Korea, Taiwan, Singapore,
but also China, have long recognized the importance of regional production networks
with a vast literature spawned around the ‘flying geese’ pattern of development based on
the work of Akamatsu (1962; UNCTAD, 1996).
In response to the collapse in trade after the financial crisis of 2007–8, suppliers in other
developing regions shifted their end markets from the North to the South in an effort to
regionalize their supply chains. South African clothing manufacturers, for example,
moved into other countries in sub-Saharan Africa such as Lesotho and Swaziland, leading
to an expansion of the regional value chain led by South African retailers (Gereffi, 2014);
and the same path is being followed by Mauritian textile producers who are expanding
their operations in Madagascar.
Such regional value chains (RVCs) are characterized by the end-product being exported
by a country within the region, more often to a regional partner, and with many high val
ue-adding activities also undertaken within the region. They can therefore significantly
contribute to the creation of value at the local level and offer more opportunities to par
ticipate, gain experience, and build those local capacities needed to compete globally,
thus potentially serving as a stepping stone into GVCs (UNCTAD, 2015). Regional markets
might also exhibit better upgrading potentials, particularly in terms of functional upgrad
ing, including design, marketing, branding, and distribution. Furthermore, given the size
and the capacity constraints faced by many developing countries, a local industrial strate
gy might quickly reach its limits. This can be overcome through a regional perspective as
different complementary advantages in the region could be leveraged and economies of
scale, vertical integration, and horizontal specialization could be promoted.
Page 24 of 34
terdependent World
RVCs can be instrumental in increasing value added in the developing regions (Kozul-
Wright and Fortunato, 2019). Recent evidence, however, suggests that despite their ad
vantages they remain far less developed than GVCs (Los et al., 2015). Indeed, over the
last couple of decades, outside of Europe and East Asia, the gap between the two has
been widening rather than shrinking and the extra regional component of foreign value
addition remains significantly higher than the intra-regional component.
Expanding regional productive integration in developing regions beyond the current lim
its critically relies on the capacity of developing countries to provide an environment con
ducive to GVC participation and that would make domestic firms competitive along these
chains. This, in turn, requires the adoption of a vast set of strategic policy measures such
as, for example, enhanced cooperation among governments to identify and prioritize en
try points into value chains and exploit regional complementarities, facilitating the con
nections between firms operating in different countries at different parts of the chain.
Setting-up Rapid Development Zones or Free Industrial Areas in those regions where nat
ural resources are concentrated to foster resource (p. 260) exploitation and upstream par
ticipation to value chains could also help, but the record of these mechanisms should
warn against treating them as panacea to a complex set of interrelated policy challenges.
9.5 Conclusions
This chapter has reviewed the debates around trade and industrial policy and discussed
how the composition of trade and investment flows, as well as the spread and form of par
ticipation in GVCs, affect structural transformation. It focused on three characteristics
that have been identified in the literature as critical to assessing the export structure of
an economy and its potential to accelerate industrialization: the diversification of produc
tion, the level of sophistication of the exported products, and upgrading of productive ca
pacities/capabilities required to sustain the production and export of increasingly sophis
ticated goods and the establishment of linkages within and across sectors. The relative
importance of each one of these features changes through time, and with them the struc
ture of trade and the policies needed for linking trade and production in ways that main
tain a virtuous economic circle of rising productivity, expanding exports, increased invest
ment, rising wages, and deeper domestic markets, fuelling further productivity rises.
The chapter also discussed the critical components of a national export strategy which
could support the insertion of national firms in international markets, favour the strategic
attraction of FDI, and enable constant upgrading along global (and regional) value
chains. What seems to be the case is that to expand production capabilities and foster
structural change, a focus on exporting manufactures is not enough. Moreover, success
comes not simply from shifting resources from primary activities to labour-intensive man
ufactures but also from anticipating future challenges in these industries (as costs rise
and new competitors emerge) and nurturing new linkages and more sophisticated prod
ucts. Accordingly, an effective national export strategy must still involve active industrial
policies, targeted support for upgrading, and regional economic arrangements.
Page 25 of 34
terdependent World
References
Agosin, Manuel (2013) ‘Productive Development Policies in Latin America: Past and
Present’, Serie de Documentos de Trabajo No. 382, University of Chile, Santiago.
Akamatsu Kaname (1962) ‘A Historical Pattern of Economic Growth in Developing Coun

tries’, Journal of Developing Economies 1(1): 3–25.
Amsden Alice H. (2001) The Rise of the ‘Rest’: Challenges to the West from Late-Industri
alizing Economies. Oxford: Oxford University Press. (p. 261)
Antràs, Pol, Davin Chor, Thibault Fally, and Russell Hillberry (2012) ‘Measuring the Up
streamness of Production and Trade Flows’, American Economic Review 102(3): 412–16.
Baldwin, Richard (2014) ‘Trade and Industrialization after Globalization’s Second Un
bundling: How Building and Joining a Supply Chain Are Different and Why It Matters’, in
Robert C. Feenstra and Alan M. Taylor (eds) Globalization in an Age of Crisis: Multilateral
Economic Cooperation in the Twenty-first Century. Chicago, IL: Chicago University Press,
pp. 165–214.
Globalization. Cambridge, MA: Harvard University Press.
Becker, Gary (1985) ‘The Best industrial Policy Is None At All’, Business Week, 26 August.
Bernard, Andrew B., J. Bradford Jensen, Stephen J. Redding, and Peter K. Schott (2007)
‘Firms in International Trade’, Journal of Economic Perspectives 21(3): 105–30.
Buckley, Peter J. (1996) ‘Government Policy Responses to Strategic Rent-seeking Transna

tional Corporations’, Transnational Corporations 5(2): 1–17.
Caporaso, James A. (1981) ‘Industrialization in the Periphery: The Evolving Global Divi
sion of Labor’, International Studies Quarterly 25(3): 347–84.
Named: Principles of Industrial Policy’. IMF Working Paper No. 19/74. Washington, DC:
IMF.
Clarida, Richard H. and Ronald Findlay (1992) ‘Government, Trade, and Comparative Ad
vantage’, The American Economic Review 82(2): 122–7.
Clerides, Sofronis K., Saul Lach, and James R. Tybout (1998) ‘Is Learning by Exporting
Important? Micro-Dynamic Evidence from Colombia, Mexico, and Morocco’, The Quarter
ly Journal of Economics 113(3): 903–47.
Cohen Stephen and Bradford de Long (2016) Concrete Economics: The Hamilton Ap
proach to Economic Growth and Policy. Cambridge, MA: Harvard Business Review Press.
Page 26 of 34
terdependent World
Cornwall, John (1977) ‘The Relevance of Dual Models for Analyzing Developed Capitalist
Economies’, Kyklos 30(1): 51–73.
De Backer, Koen and Sébastien Miroudot (2013) ‘Mapping Global Value Chains’. OECD
Trade Policy Paper No. 159. Paris: OECD Publishing.
Dedrick, Jason, Kenneth Kraemer, and Greg Linden (2010) ‘Who Profits from Innovation
in Global Value Chains? A Study of the iPod and Notebook PCs’, Industrial and Corporate
Change 19(1): 81–116.
Emmanuel, Arghiri (1972) Unequal Exchange: A Study of the Imperialism of Trade. New
York: Monthly Review Press.
Evans, Peter (1998) ‘Transferable Lessons? Re-examining the Institutional Prerequisites

of East Asian Economic Policies’, Journal of Development Studies 34(6): 66–86.
Fally, Thibault (2012) ‘Production Staging: Measurement and Facts’, University of Col
orado, unpublished manuscript.
Felipe Jesus, Aashish Mehta, and Changyong Rhee (2014) ‘Manufacturing Matters…But
It’s the Jobs That Count’. Asian Development Bank Economics Working Paper No. 420.
Findlay, Ronald and Kevin H. O’Rourke (2008) Power and Plenty: Trade, War, and the
World Economy in the Second Millennium. Princeton, NJ: Princeton University Press.
Fons-Rosen, Christian, Sebnem Kalemli-Ozcan, Bent Sorensen, Carolina Villegas-Sanchez,

and Vadym Volosovych (2013) ‘Quantifying Productivity Gains from Foreign Investment’.
Working Paper No. 18920. Cambridge, MA: National Bureau of Economic Research.
Fortunato Piergiuseppe and Carlos Razo (2014) ‘Export Sophistication, Growth and the
Middle-income Trap’, in José M. Salazar-Xirinachs, Irmgard Nübler, and Richard (p. 262)
Kozul-Wright (eds) Transforming Economies: Making Industrial Policy Work for Growth,
Jobs and Development. Geneva: International Labour Organization, pp. 267–87.
Gallagher Kevin and Victor Zarsky (2007) The Enclave Economy: Foreign Investment and
Sustainable Development in Mexico’s Silicon Valley. Cambridge, MA: MIT Press.
Gereffi, Gary, John Humphrey, Raphael Kaplinsky, and Timothy Sturgeon (2001) ‘Globali
sation, Value Chains and Development’, IDS Bulletin 32(3): 1–8.
Value Chains’, Review of International Political Economy 12(1): 78–104.
Gomory, Ralph and William Baumol (2000) Global Trade and Conflicting National Inter
ests. Cambridge, MA: MIT Press.
Page 27 of 34
terdependent World
Grabowski Richard (2003) ‘Promoting Industrialization: The Role of the Traditional Sec
tor and the State in East Asia’, Journal of International Development 15(5): 587–605.
Harrison, Ann E. and Andres Rodriguez-Clare (2009) ‘Trade, Foreign Investment, and In
dustrial Policy for Developing Countries’. NBER Working Paper No.15261.
Harrison, Ann E., John Mclaren, and Margareth McMillan (2016) ‘Recent Perspectives on
Trade and Inequality’. World Bank Policy Research Working Paper No. 5754. Washington,
DC: World Bank.
Hausmann, Ricardo, Jason Hwang, and Dani Rodrik (2007) ‘What You Export Matters’,
Journal of Economic Growth 12(1): 1–25.
Hymer, Stephen H. (1972) ‘The Internationalization of Capital’, Journal of Economic Is

sues 6(1): 91–111.
Imbs Jean and Romain Wacziarg (2003) ‘Stages of Diversification’, The American Econom
ic Review 93(1): 63–86.
Ingram, Verina and Elsje Oosterkamp (2014) ‘Literature Review on the Labour Market
Impacts of Value Chain Development Interventions’. LEI Report. LEI Wageningen UR
(University and Research Centre).
Kaldor, Nicholas (1966) ‘Marginal Productivity and the Macro-Economic Theories of Dis
tribution: Comment on Samuelson and Modigliani’, Review of Economic Studies 33(4):
309–19.
Kaplinsky, Raphael (2005) Globalization, Poverty and Inequality: Between a Rock and a
Hard Place. Cambridge: Polity Press.
Kim, Dong-Hyeon (2011) ‘Trade Growth and Income’, Journal of International Trade and
Economic Development 20(5): 677–709.
Knight John (2012) ‘China as a Developmental State’. CSAE Working Paper No. 2012–13,
University of Oxford, Oxford.
Kowalski, Przemyslaw, Javier Lopez-Gonzalez, Alexandos Ragoussis, and Cristian Ugarte

(2015) ‘Participation of Developing Countries in Global Value Chains: Implications for
Trade and Trade-related Policies’. OECD Trade Policy Papers No. 179. Paris: OECD.
Kozul-Wright, Richard and Piergiuseppe Fortunato (2015) ‘Sustaining Industrial Develop

ment in the South’, Development 58(4): 463–72.
Kozul-Wright, Richard and Piergiuseppe Fortunato (2019) ‘Development Unchained:

Trade and Industrialization in the Era of International Production’, Global Policy 10(1):
29–40. (p. 263)
Page 28 of 34
terdependent World
Kozul-Wright Richard and Daniel Poon (2019) ‘Economic Openness and Development’, in
Deepak Nayyar (ed.) Asian Transformations: An Inquiry into the Development of Nations.
Oxford: Oxford University Press, pp. 124–45.
Krueger, Anne O. (1990) ‘Government Failures in Development’, Journal of Economic Per

spectives 4(3): 9–23.
Krugman, Paul (ed.) (1986) Strategic Trade Policy and International Economics. Cam
bridge, MA: MIT Press.
Krugman, Paul (1994) ‘The Myth of Asia’s Miracle’, Foreign Affairs, November/December.
Levine, Ross and David Renelt (1992) ‘A Sensitivity Analysis of Cross-country Growth Re
gressions’, American Economic Review 82(4): 942–63.
Levy, Daniel L. (2005) ‘Offshoring in the New Global Political Economy’, Journal of Man
agement Studies 42(3): 685–93.
ment’. Policy Research Working Papers No. 5197. Washington, DC: World Bank.
Lin, Justin Yifu and Volker Treichel (2014) ‘Making Industrial Policy Work for Develop
ment’, in José M. Salazar-Xirinachs, Irmgard Nübler, and Richard Kozul-Wright (eds)
Transforming Economies: Making Industrial Policy Work for Growth, Jobs and Develop
ment. Geneva: International Labour Office, pp. 65–78.
Lo Dic and Mei Wu (2014) ‘The State and Industrial Policy in Chinese Economic Develop
ment. Geneva: International Labour Office, pp. 307–26.
Los, Bart, Marcel Timmer, and Gaaitzen de Vries (2015) ‘How Global are Global Value
Chains? A New Approach to Measure International Fragmentation’, Journal of Regional
Science 55: 66–92.
Marel, Erik (2015) ‘Positioning on the Global Value Chain Map: Where Do You Want to
Be?’, Journal of World Trade 49(6): 915–49.
McMillan, Margaret S. and Dani Rodrik (2011) ‘Globalization, Structural Change and Pro
ductivity Growth’, in Marc Bacchetta and Marion Jansen (eds) Making Globalization So
cially Sustainable. Geneva: International Labour Organization, pp. 49–84.
Milberg, William and Deborah Winkler (2013) Outsourcing Economics: Global Value
Chains in Capitalist Development. New York: Cambridge University Press.
Milberg, William, Xiao Jang, and Gary Gereffi (2014) ‘Industrial Policy in the Era of Verti
cally Specialized Industrialization’, in José M. Salazar-Xirinachs, Irmgard Nübler, and
Page 29 of 34
terdependent World
Richard Kozul-Wright (eds) Transforming Economies: Making Industrial Policy Work for
Growth, Jobs and Development. Geneva: International Labour Office, pp. 151–80.
Naudé William (2010) ‘New Challenges for Industrial Policy’. Working Paper No.
2010/107. Helsinki: UNU-WIDER.
OECD (2013) ‘Interconnected Economies: Benefiting from Global Value Chains’. Synthe
sis Report. Paris: OECD.
OECD-UNIDO (2019) ‘Integrating Southeast Asian SMEs in Global Value Chains: En
abling Linkages with Foreign Investors’. Paris: OECD.
Panic, Milivoje (1988) National Management of the International Economy. London:

Macmillan.
Paus, Eva (2014) ‘Industrial Development Strategies in Costa Rica: When Structural
Change and Domestic Capability Accumulation Diverge’, in José M. Salazar-Xirinachs, Ir
mgard Nübler, and Richard Kozul-Wright (eds) Transforming Economies: Making Industri
al (p. 264) Policy Work for Growth, Jobs and Development. Geneva: International Labour
Office and UNCTAD, pp. 181–211.
Pavlínek, Petr (2015) ‘Whose Success? The State–Foreign Capital Nexus and the Develop
ment of the Automotive Industry in Slovakia’, European Urban and Regional Studies
23(4): 571–93.
Pavlínek, Petr and Jan Zenka (2016) ‘Value Creation and Value Capture in the Automotive
Industry: Empirical Evidence from Czechia’, Environment and Planning 48(5): 937–59.
Peiffer, Caryn (2012) ‘Reform Coalitions, Patterns and Hypotheses from a Survey of the
Literature’. Concept Paper No. 03, Developmental Leadership Program. Available at
http://publications.dlprog.org/Reform%20Coalitions.pdf.
Phillips, Richard and Jeffrey Henderson (2009) ‘Global Production Networks and Industri
al Upgrading: Negative Lessons from Malaysian Electronics’, Austrian Journal of Develop
ment Studies 25(2): 38–61.
Plank, Leonhard and Cornelia Staritz (2013) ‘Precarious Upgrading in Electronics Global
Production Networks in Central and Eastern Europe: The Cases of Hungary and Roma
nia’. Working Paper No. 31. GDI, University of Manchester.
Poon, Daniel and Richard Kozul-Wright (2019) ‘Learning from East Asia: Catch-up and the
Making of China’s Developmental State’, in Arkebe Oqubay and Kenichi Ohno (eds) How
Nations Learn. Oxford: Oxford University Press, pp. 38–62.
Problems’. Santiago: Economic Commission for Latin America and the Caribbean
(ECLAC).
Page 30 of 34
terdependent World
Pritchett, Lant (1996) ‘Measuring Outward Orientation in LDCs: Can it be Done?’, Journal
of Development Economics 49(2): 307–35.
Rigo, Davide (2017) ‘A Portrait of Firms Participating in Global Value Chains’, Technical
Report. Geneva: IHEID.
Rodrik, Dani (2004) ‘Industrial Policy for the Twenty-First Century’. CEPR Discussion Pa
pers No. 4767.
Rodrik, Dani (2008) ‘Normalizing Industrial Policy’. Commission on Growth and Devel-
opment Working Paper No. 3, Washington, DC.
Rodrik, Dani (2011) The Globalization Paradox: Democracy and the Future of the World
Economy. New York: W. W. Norton.
Rodrik, Dani (2016) ‘Premature Deindustrialization’, Journal of Economic Growth 21: 1–

33.
Rodrik, Dani (2017) Straight Talk on Trade: Ideas for a Sane World Economy. Princeton,
NJ: Princeton University Press.
Samuelson, Paul (2004) ‘Where Ricardo and Mill Rebut and Confirm Arguments of Main
stream Economists Supporting Globalization’, Journal of Economic Perspectives 18(3):
135–46.
Schneider, Ben Ross (2013) Hierarchical Capitalism in Latin America: Business, Labour
and the Challenges of Equitable Development. Cambridge: Cambridge University Press.
Schneider, Ben Ross (2016) ‘Upgrading in the 21st Century: New Challenges for Industri
al Policy and Institution Building’. Background paper prepared for UNCTAD’s Trade and
Development Report 2016. Geneva: UNCTAD. Unpublished.
Singer, Hans (1950) ‘The Distribution of Gains between Investing and Borrowing Coun
tries’, The American Economic Review 40(2): 473–85.
Storm, Servaas (2017) ‘The Political Economy of Industrialization’, Development and

Change. Available at https://doi.org/10.1111/dech.12281. (p. 265)
Studwell, Joe (2013) How Asia Works: Success and Failure in the World’s Most Dynamic
Region. London: Profile Books.
Timmer, Marcel, Erik Dietzenbacher, Bart Los, Robert Stehrer, and Gaaitzen de Vries
(2015) ‘An Illustrated User Guide to the World Input–Output Database: The Case of Global
Automotive Production’, Review of International Economics 23: 575–605.
Timmer, Marcel, Bart Los, Stephen Rand, and Gaaitzen de Vries (2013) ‘Fragmentation,
Incomes and Jobs: An Analysis of European Competitiveness’, Economic Policy 28(76):
613–61.
Page 31 of 34
terdependent World
Toner, Phillip (1999) Main Currents in Cumulative Causation: The Dynamics of Growth
and Development. London: Macmillan.
Tregenna, Fiona (2009) ‘Characterising Deindustrialisation: An Analysis of Changes in

Manufacturing Employment and Output Internationally’, Cambridge Journal of Economics
33: 433–66.
Tregenna, Fiona (2010) ‘How Significant is Intersectoral Outsourcing in South Africa’, In
dustrial and Corporate Change 19(5): 1427–57.
UNCTAD (1994) Trade and Development Report 1994. New York: United Nations.
UNCTAD (2013) World Investment Report 2013—Global Value Chains: Investment and
Trade for Development. New York: United Nations.
UNCTAD (2015) Global Value Chains and South-South Trade. New York: United Nations.
UNCTAD (2016) Trade and Development Report, 2016: Structural Transformation for In
clusive and Sustained Growth. New York: United Nations.
UNCTAD (2018) Structural Transformation and Export Diversification in Southern Africa.

New York: United Nations.
Vaggi, Gianni and Peter Groenewegen (2003) A Concise History of Economic Thought:
From Mercantilism to Monetarism. Basingstoke: Palgrave.
Wacziarg, Romain and Karen Welch (2003) ‘Trade Liberalization and Growth: New Evi
dence’. NBER Working Paper No. 10152. Cambridge, MA: National Bureau of Economic
Research.
Wade, Robert H. (1995) ‘Resolving the State–Market Dilemma in East Asia’, in Ha-Joon
Chang and Robert Rowthorn (eds) The Role of the State in Economic Change. Oxford:
Clarendon Press, pp. 114–36.
Wade, Robert H. (2015) ‘The Role of Industrial Policy in Developing Countries’, in Re
thinking Development Strategies after the Financial Crisis, Vol. I: Making the Case for
Policy Space. New York: United Nations, pp. 67–79.
World Trade Organization (2017) ‘Measuring and Analyzing the Impact of GVCs on Eco
nomic Development: Global Value Chains Development Report 2017’. Geneva: WTO.
Page 32 of 34
terdependent World
Notes:
(1) Smith offered a number of reasons why some countries lagged behind, including un
favourable cost conditions (which led some countries to forsake manufacturing); a hostile
policy environment (which included insecure property rights and misguided trade policy);
weak infrastructure (which augmented geographical obstacles to market expansion); and
small or scattered populations (which limited the division of labour).
(2) In his Lectures on Jurisprudence published before The Wealth of Nations, Smith noted
that ‘it is easier for a nation, in the same manner as for an individual, to raise itself from a
moderate degree of wealth to the highest opulence, than to acquire this moderate degree
of wealth’ (cited in Vaggi and Groenewegen, 2003: 113).
(3) Of course, a significant proportion of the increased stock of FDI is linked to non-trad
able tertiary sectors of the economy.
(4) Bulgaria, Brazil, China, India, Indonesia, Mexico, Romania, Russia, and Turkey.
(5) Among the BRIC economies, Brazil and India have experienced a marginal decline in
both measures while Russia and China experienced a rise, albeit for different reasons.
Richard Kozul-Wright
Richard Kozul-Wright is director of the Globalisation and Development Strategies Di

vision in UNCTAD. He has worked at the United Nations in both New York and Gene
va. He holds a PhD in economics from the University of Cambridge. He has published
widely on economic issues including, inter alia, in The Economic Journal, Cambridge
Journal of Economics, The Journal of Development Studies, and Oxford Review of
Economic Policy. He is the author of many books, including The Rise and Fall of Glob
al Microcredit: Development, Debt and Disillusion (2018, with S. Blankenburg and M.
Bateman), Securing Peace: State-Building and Economic Development in Post-Con
flict Countries (2011, with P. Fortunato), Climate Protection and Development (2012,
with Frank Ackerman) and The Resistible Rise of Market Fundamentalism (2008,
with Paul Rayment). His edited volumes include Transnational Corporations and the
Global Economy, Economic Insecurity and Development, Securing Peace, Climate
Protection and Development, and Industrial Policy. He is a frequent contributor to
newspapers worldwide on economic issues, including the Financial Times, The
Guardian, and Project Syndicate.
Piergiuseppe Fortunato
Piergiuseppe Fortunato is an economist at the United Nations Conference for Trade

and Development. He previously worked in the Department of Economic and Social
Affairs of the United Nations in New York and held positions at the University of
Bologna and Université Paris I. His research interests include political economics,
economic development, and industrial policy. In these fields he has published in jour
nals including Journal of Economic Growth, The Economic Journal, and European
Page 33 of 34
terdependent World
Economic Review. He is also heavily involved in policy-oriented research, has edited
a volume on post-conflict recovery, published in policy journals such as Global Policy
and contributes to several blogs.
Page 34 of 34
Greening Industrial Policy

John A. Mathews
The scope of industrial policy is broadened in this chapter to encompass energy, materi
als, and finance. Given the unprecedented scale of the industrialization of India, China,
and the late latecomers in Africa, Asia, and Latin America, the traditional options of fossil
fuels, linear economy, and generic finance have been found wanting, for reasons of imme
diate environmental spoliation but also to do with countries running up against geopoliti
cal limits. The greening of industrial policy presents itself as a solution, involving the shift
to renewables, to a circular economy, and to the greening of finance (green bonds and
green loans from development banks), encompassed under the rubric of green growth.
The details of these new, green options are explored, and contrasted with traditional
green industrial policy. The case is made that, far from being a special case, the greening
of industrial policy promises to become the core development strategy, and the general
case in industrialization, as the twenty-first century unfolds.
Keywords: renewable energy, circular economy, green finance, fossil fuels, green growth, learning curve, urban
mining
10.1 Why Green Industrial Policy?

IT is widely agreed that the West grew wealthy through its mastery of the novel processes
of industrialization, when fossil fuels were harnessed to provide abundant energy in new
factory settings that liberated production from age-old constraints. Industrial capitalism
underpinned the prosperity of Europe and North America, before diffusing East to Japan
in the twentieth century, and then in the second half of that century diffusing to the East
Asian Tigers (Korea, Taiwan, Singapore, and Hong Kong), utilizing developmental states
as means of catch-up. Now in the twenty-first century the process encompasses China,
followed by India and many other industrializing giants looking to enjoy their time in the
sun. But no sooner do these emergent industrializing giants begin the same process of
burning fossil fuels and wasting resources as practised by their industrialized predeces
Page 1 of 20
sors than they are confronted with an inconvenient truth: their business model for indus
trialization will not scale.
These industrializing giants have to confront the reality that they cannot rely on the con
ventional fossil-fuelled pathway, nor on the traditional linear pathway of resource usage
(extracting resources from nature, and then dumping the residues in nature at the end of
the process), if they wish to see their industrialization through to completion. Consider
these traditional or conventional pathways from the perspective of energy and resource
security. The burning of fossil fuels (whether coal in power generation and industry or oil
in transport) at the scale involved in China or India (with their total current population of
2.8 billion) creates so much urban particulate pollution that the air becomes unbreath
able. China has already paid a terrible price in terms of this urban pollution, and India
will do so as well as the scale of its fossil-fuel burning rises. Similarly, the linear exploita
tion of resources, extracting them from nature at one end of the industrial process and
dumping them back in nature at the (p. 267) other, creates unmanageable waste issues
and shortages of key resources like water. Even more significantly, the huge appetite for
fossil fuels and virgin resources brings these industrializing giants up against political
and economic limits (and ultimately military limits) of what can be allowed within a
densely populated planet. The Western powers evaded these geopolitical limits through
colonization and imperialism—but such a strategy is not open to China, India, and other
industrializing giants. How then are they to secure the energy and resource inputs need
ed by their quest for industrialization at a scale never before attempted? What are the in
dustrial policies that would bring them the fruits of modernization?
The answer to this conundrum is provided by green growth, or the greening of the indus
trial growth process—balancing growth against sustainability. China stumbled on this so
lution in the early years of the twenty-first century, as it was engaging in tearaway
growth. While continuing to burn a lot of coal, oil, and gas, it has found that its energy se
curity is enhanced by relying more and more on renewables—to the point that it now has
targets for renewables usage that dwarf those of other countries and have already turned
China into a renewables superpower (Mathews and Tan, 2014a, 2014b, 2016; Mathews et
al., 2019). And this for the very good reason that renewables are products of manufactur
ing, and as such are subject to cost reductions achieved through the learning curve.
While fossil-fuel extraction is subject to arbitrary cost increases or decreases, the costs of
manufacturing renewables devices (wind turbines, solar PV cells, batteries) are diminish
ing relentlessly, in accordance with the learning curve.
As costs fall, markets expand—and so the process opens up and expands markets for re
newables devices, in a process that can be expected to lead to fossil fuels being super
seded as energy sources by the middle of the century, if not before. In most parts of the
world electric power generated from renewables is already cheaper than thermal power
generated from burning fossil fuels—and the costs of renewable power will only continue
to fall. Figure 10.1 shows that costs of solar PV have been falling by 28.5 per cent for
every doubling of production, which has occurred every two to three years. The chart
takes the story up to 2020 when solar PV power generation will approach 1 trillion watts,
Page 2 of 20
opening up the terawatt era in solar power. Why would any country wish to continue
burning fossil fuels, given their rising and fluctuating costs, their heavy burden on the
balance of payments, and their geopolitical unreliability?
Figure 10.1 Learning curve associated with solar PV

power
Source: BNEF.
China has discovered a radical solution to the problem of resource security by introduc
ing circular flows of materials in place of the linear flows of conventional industrializa
tion. The recirculation of resources is based on manufacturing (or rather, ‘demanufactur
ing’ or disassembling) and is subject likewise to diminishing costs, so that the costs of re
circulated materials can be expected to continue to fall, eventually falling below the cost
of virgin materials (indeed they are already lower in cost in some activities such as ‘urban
mining’ of electrical and electronic waste in Chinese cities). The recirculation of re
sources solves not only the problem of resource accessibility, but also the problem of
waste accumulation. It provides a sustainable solution to the problem of dealing with
geopolitical limits to resource extraction. As China expands its adoption of a circular
economy to enhance its own resource security, so it creates (p. 268) markets and technolo
gies that can be adopted by other industrializing countries, starting with India, and en
compassing late latecomers in Africa, Asia, and Latin America.
This perspective on greening views the process as fundamentally driven by the quest for
energy security and resource security necessitated by the unprecedented scale at which
China and India are industrializing. It is a very different perspective from the one that in
forms almost all commentary on green industrial policy, which begins with a concern over
global climate change and deduces from this the need for a low-carbon economy (and in
more extreme versions, for a zero-growth economy as well). Such a perspective can only
result in industrializing giants facing energy and resource choices that compel them to
confront the ethical and moral challenges of decarbonization; little wonder that this per
spective is resisted by late industrializers like India, given that it would condemn them to
abandoning their search for industrialization before it has started, and with it the search
for increased wealth and income to bring them closer to the levels enjoyed by Western in
Page 3 of 20
dustrialized countries. And they were not responsible for climate change in the first
place.
Consider the impact of switching to renewables to drive power generation, industry, and
transport for reasons of energy security. Security is enhanced by such a move because re
newables are products of manufacturing, and as such come under the control of the coun
try implementing the policy. A green choice entails a process of decarbonization—the only
known solution to the problem of rising carbon levels. As the industrializing giants like
China and India switch over to renewables, leaving behind (p. 269) the fossil fuels of an
earlier era, so they are led to achieve the very results anticipated by a ‘climate change’
perspective on greening. Likewise, choosing recirculating resources for reasons of re
source security decouples industrial activity from natural processes, reduces the materi
als footprint of industrial activity, and allows the Earth to begin to reclaim its natural
processes and cycles. The aspirational goal of zero-growth advocates is thus achieved
through green growth rather than zero growth.
It is considerations like these that have led international agencies like UNEP to argue
that the greening of economies ‘is not generally a drag on growth but rather a new en
gine of growth’.1 The discourse on developing countries and green industrial policy is
thus moving away from ‘burden sharing’, where the costs of renewables or circular flows
are viewed as higher than in conventional fossil-fuelled or linear flows. Countries are now
discovering profitable opportunities associated with a green shift—notably the falling
costs of solar PV (as shown in Figure 10.1) and the complementary cost reductions found
in wind turbines, batteries, electric vehicles, and other instances of the global green shift.
Green growth is not so much a response to market failure (as in the standard account of
neoclassical economics) as a response to market opportunities opened up by the Schum
peterian creative destruction of the incumbent fossil-fuelled energy systems and linear re
source flows by renewables and recirculated resources, with the new industries driven by
green finance. These are the real-world industrial dynamics of the greening of industrial
policy.2
This chapter develops the argument that in a wider technoeconomic setting, as compared
to the narrower economic setting of traditional industrial policy, it is the energy and re
source choices made by industrializing countries that will come to be central. Indeed,
they will determine the success or failure of the industrialization aspirations of the emer
gent giants like China and India, and following them, the late latecomers in Africa, Asia,
and Latin America. The material and energy foundations of these industrialization strate
gies constitute the core of green growth industrial policies.3
Page 4 of 20
10.2 General Features of the Greening of In

dustrial Policy
The core of industrial policy concerns the strategies deployed to shift an economy from
lower- to higher-productivity sectors, and within sectors, from lower value-added to
(p. 270) higher value-added activities. These strategies inevitably focus on manufacturing,
where such productivity improvements can be concentrated, while manufacturing itself

can be concentrated in clusters, or industrial hubs, where an export orientation can be
pursued. These strategic choices by successful industrializers (in particular those in East
Asia) now characterize the choices made by smart late latecomers around the world.4
The greening of industrial policy broadens the canvas to encompass the energy and mate
rials/resources aspects of manufacturing activities—widening the scope from economic to
technoeconomic considerations. As soon as this wider perspective is taken, the material
underpinnings of industrialization begin to exert powerful effects. Let us consider the fun
damental aspects of these technoeconomic policy choices, spanning green energy, green
resources, and the green finance that drives the choices being made in energy and re
sources.5
10.2.1 Energy
Energy choices are a principal aspect of the greening of industrial policy. A worldwide
green transition is underway, shaped initially by the choices made by China and India,
and now diffusing to encompass the choices made by countries in Africa, Asia, and Latin
America.6
In the energy domain, the move towards renewables like solar and wind has the great ad
vantage that the resource is free, and marginal costs of generating power from these free
resources are correspondingly low—lower than burning expensive fuel. The renewable re
sources are diffuse, meaning that they are available almost everywhere, and to everyone,
countering the trends towards centralization and gigantism in the traditional fossil-fu
elled industry. Renewables-based industries tend to be labour-intensive (think of in
stalling rooftop solar modules) and generate jobs in rural and regional areas. And of
course renewable energy is clean, in the sense that it carries no pollution risks (in con
trast to the filth associated with coal mining and burning, or the radiation risks associat
ed with nuclear power), and poses no security dilemma, as in the ever-present risk that
civil nuclear power industries could be converted to military use virtually overnight. The
renewables industries favour small and medium-sized firms as protagonists, basing their
competitive dynamics on innovation as much as on (p. 271) imitation. By contrast, how
much innovation has there been in the automotive industry over the past century (until
the appearance of electric vehicles)?
Page 5 of 20
10.2.2 Materials/Resources
The circular economy emerges as an alternative to the traditional linear economy in the
resources or materials domain. The fundamental attribute of circular resource flows is
that they enhance resource security as more and more resources are extracted from cir
cular flows rather than as virgin resources (think of water recycling and treatment). Re
source extraction thus becomes a branch of manufacturing (or ‘demanufacturing’)—as in
‘urban mining’—and its pursuit becomes a goal of industrial policy. Recirculated re
sources have declining costs, as demanufacturing generates its own learning curve and
the market for recirculated resources enlarges—to the point where costs of recovered re
sources dip below the costs of extracting virgin resources.7 Capturing circular flows by
closing industrial loops generates rich linkages between industrial sectors and multiplies
opportunities for capturing increasing returns. For an industrializing economy, the shift to
circular flows saves on resources, saves on waste, generates abundant business opportu
nities, and creates jobs in rural and regional areas. What is there not to like?
10.2.3 Finance
What drives these shifts in energy and resources/materials flows is finance, which, as
Schumpeter correctly observed, is the ‘engine room’ of capitalism. Here we introduce an
other departure from the usual treatment of greening of industrial policy. In the way that
the issues are typically framed by United Nations agencies, finance means creating funds
from the resources of governments, meaning taxpayers’ funds.8 But this approach to fi
nance ignores the central feature of capitalism, namely that it runs on credit—and credit
is created in capital markets. The issue becomes: how to create instruments of credit that
draw from the vast capital markets created by capitalism?
One solution to this issue is to target the bonds markets, which globally are double the
size of equity (stocks) markets. Banks and financial institutions have found that they can
attract the interest of professional investors—managers of wealth funds, hedge funds, in
surance funds, and pension funds—with bonds targeted at green investments, or green
bonds. Ever since the Korean Export-Import Bank first floated a green bond (p. 272) suc
cessfully on global bond markets in March 2013, raising US$500 million in funds to be in
vested in green projects by Korean firms around the world, the idea has caught on in a
big way. China in particular has taken to green bonds, viewing them as a way of tapping
global capital markets to fund its green operations both at home and abroad (under the
Belt and Road Initiative). Green bonds are diffusing to late latecomers around the world.
The Bank Windhoek in Namibia, for example, issued a green bond in 2018 targeted at re
newable energy projects as well as at reducing carbon emissions from fossil-fuel activities
(a controversial aspect).9 The scale of green bond issuances continues to grow, reaching
US$167 billion in 2018, and US$117.8 billion in the first half of 2019.10
The other way that industrializing countries can channel finance towards green growth
initiatives—both energy and resources/materials aspects—is through development banks.
China is at the forefront, with its two principal development banks, the China Develop
Page 6 of 20
ment Bank (CDB) and the China Exim Bank, both providing Chinese green shift compa
nies with long credit lines in the billions of dollars to sustain them in international compe
tition against less well-endowed competitors. Brazil, too, has been able to finance green
projects throughout the country through the operations of the Brazilian Development
Bank (BNDeS) which has grown to be larger than the World Bank.
How then are the leading industrializing countries today putting these green initiatives to
work in driving green growth, or greening industrial policy?
10.3 Effective Industrial Policy/Green Growth

Initiatives in India and China
10.3.1 Green Industrial Policy in India and its Impact on Trade
Let us start with India, which is not only about to become the world’s most populous
country (expected to overtake China by the mid-2020s) but is actively pursuing green
projects across the board.11 The hallmarks of India’s newfound green growth zeal (p. 273)
(emulating China’s strategies with a lag of perhaps a decade) can be found in the Nation
al Solar Mission, designed to achieve solar PV capacity in India of 100 GW by 2022, and
in the corresponding target for wind power of 60 GW, again to be achieved by 2022. As
shown in Figure 10.2, the National Solar Mission was steadily driving the uptake of solar
power—until it met the unexpected obstacle of trade barriers created by the advanced
countries.
The Indian case reveals the hypocrisy of the West in advancing the concept of decar
bonization as a means of mitigating climate change, and yet opposing it as soon as a
country like India starts to take practical steps to build its own green industries. Emulat
ing China’s great success with building a wind-power industry in the early 2000s, utiliz
ing the instrument of local content requirements (LCRs) (see section 10.3.2), as part of
the National Solar Mission India stipulated that there should be steadily rising propor
tions of domestic manufactures in the national solar PV output and in the solar PV instal
lation sector.12 These requirements, while universally recognized as necessary to build a
new industry in an industrializing country, are technically in breach of WTO rules, and In
dia was duly taken to the WTO for disciplinary action by the United States in a case that
started in 2010.13 India defended its policies on the grounds that they were needed to en
able India to meet its clean-tech targets under its (p. 274) Paris commitments, but these
arguments were rejected by the panel put in place to hear the case and upheld by the
highest WTO authority, an Appellate Court, in 2018. By 2019 the expansion of India’s
world-beating solar PV industry was moderating. This is the perverse result of a global
system where the WTO and the UNFCCC are dangerously out of alignment.
Page 7 of 20
Figure 10.2 India’s solar PV installed capacity,

2010‒19
Source: Author.
The data on India’s solar PV generation in Figure 10.2 tell the story: rapidly growing PV
generation up to the year 2019, when it started to decline under the impact of India hav
ing to dismantle the LCR aspects of the National Solar Mission, on pain of expulsion from
the WTO.
10.3.2 Green Industrial Policy in China and its Impact on Trade
While there is ample documentation on China’s vast consumption of coal and its growing
involvement in oil and gas production and consumption (its black economy), much less at
tention is paid to the serious greening that has been under way in China over the course
of the past decade.14 The green shift in the electric power sector, where the trends are
concentrated, is clearly shown in Figure 10.3.
Page 8 of 20
Figure 10.3 China: Rising proportion of electric pow

er sourced from water, wind, and sun, 1990‒2018
Source: Author.
Figure 10.3 reveals in striking detail how China effected a switch to dependence on re
newable sources (water, wind, and sun) around 2007, and has since been pursuing this
green shift as fast as is practicable (given the vast scale of the electric power system). In
terms of capacity (upper line) the proportion of electric power capacity sourced from wa
ter, wind, and sun (WWS) reached 37 per cent by 2018, rising from 22 per cent in 2008—
a 15 per cent green shift in just one decade. If this green shift is maintained through con
sistent green industrial policy for the next decade, China’s WWS capacity could reach
more than 50 per cent before 2030, meaning China’s electric power system (the largest in
the world) would have reached the tipping point where it is more green than black, with a
corresponding impact across wider industrial and transport sectors. The data in terms of
actual electricity generated show the same trend, if not the same absolute values (be
cause of varying capacity factors across different generating sources).
China has judiciously utilized all the tools of green industrial policy to effect this green
transition in its electric power sector. The new industries of wind power (encompassing a
manufacturing value chain for wind turbines, and the operation of huge wind farms like
the Gansu 10 GW farm) and solar PV power have been carefully (p. 275) nurtured. China
utilized LCRs and foreign direct investment (FDI) to build its wind-turbine industry in the
first decade of the twenty-first century, artfully deploying graded increases in LCRs and
(once they had achieved their objective) dismantling them before a protest could be
lodged at the WTO in Geneva. The emerging wind-turbine giants like Goldwind were
equipped with long credit lines by CDB, and leveraged latecomer technology to good ef
fect to acquire new technologies like permanent magnet direct drive (PMDD), which facil
itated the expansion of the wind-power industry offshore. The solar PV industry also ex
panded, with new entrants like Suntech Power, founded in Wuxi by young entrepreneur
Dr Shi Zhengrong in 2001, scaling up production of PV modules to a degree never before
attempted, and driving down costs as they did so.
Page 9 of 20
Of course it was not all plain sailing. Because China opened up the solar PV industry as
an exemplary case of unrestricted growth, there was a rapid build-up of over-capacity, re
sulting in many bankruptcies. Rather than being seen as a failure of industrial policy, this
should be viewed instead as a success, in that the PV industry was exhibiting ‘normal’ in
dustrial dynamics of free entry and free exit via bankruptcy or corporate acquisition. Sun
tech Power itself was one of the casualties. China’s power grid proved incapable of ac
cepting all the renewable power generated from the new (p. 276) wind and solar PV
sources, and much of the power was wasted in a process called ‘curtailment’. But this in
itself stimulated rapid innovation, with the grid upgraded as a vehicle of transmission and
distribution (T&D) and the State Grid Corporation of China leapfrogging world electrical
engineering leaders with the introduction of new T&D technologies such as ultra-high
voltage (UHV) transmission. The new UHV transmission lines built over the second
decade of the twenty-first century have enabled China to generate vast amounts of power
from renewable sources in inland provinces and then transport the electrical energy to
the eastern seaboard with minimal losses and enhanced reliability. This was a major
achievement, revealing China’s rapid passage from imitation to innovation in a critical
sector.15
An unheralded aspect of China’s green energy shift over the course of the past decade is
that as the scale of its wind and solar PV markets expanded, so efficiencies improved and
costs were driven down spectacularly, as evidenced in Figure 10.1. These cost reductions
have not been confined to China, but through globalization they have diffused to the rest
of the world. The result has been an unprecedented boom in building renewable energy
industries in newly industrializing countries, particularly in those pursuing ‘late’ latecom
er development strategies.
10.4 Opportunities for Late Latecomers to Ap

ply Green Industrial Policies to Drive Develop
ment
An excellent example of a green economy initiative in a ‘late’ latecomer that makes abun
dant economic and business sense may be found in Ethiopia, a country that is industrializ
ing fast, as the ‘China of Africa’. The country has a clear industrial policy focused on man
ufacturing, which is concentrated on a series of industrial parks located around the coun
try. There are clear strategies in place to have these parks linked to logistical infrastruc
ture, such as the new (Chinese-financed) railway from the capital Addis Ababa to the port
of Djibouti on the Red Sea, which serves as a conduit to the outside world for the land
locked country.
These industrial parks offer investors many advantages, including shared capital infra
structure and common provision of energy and resources inputs. For textile and clothing
firms (the majority of invested enterprises in the industrial parks) water is a critical re
source—and some of the parks are offering an advanced zero liquid discharge facility, or
Page 10 of 20
otherwise advanced water treatment and recycling facilities that come with park admis
sion. In one park, water is treated both physically and chemically to remove pollutants
and impurities in the form of a sludge that is solar dried and then (p. 277) transported off-
site to a local cement plant, where it is absorbed as an input. This creates linkages across
manufacturing centres as well as the closure of a loop linking the textiles and cement in
dustries. And from the perspective of the firms involved in the industrial parks, it reduces
the costs of water as input, since the water utilized in the park is 90 per cent recycled,
providing ‘grey water’ for all enterprises at costs well below those that would apply to
water drawn from underground supplies or from local lakes and rivers. And the country is
saved the environmental damage that would flow from otherwise ‘free’ discharge of pol
luted water into local waterways.16
This is an example of an environmental initiative that enhances the water security of the
country, utilizing a technologically advanced water treatment and recycling system that is
as yet found in few advanced countries. It is thus a case of a government agency (in this
case, the Industrial Parks Development Corporation) acting as a public entrepreneurial
agent and introducing technoeconomic ‘leapfrogging’ as a matter of industrial policy in
Ethiopia. And once installed and operational, the initiative can be replicated by other late
late developers elsewhere in Africa, Asia, or Latin America. The more it is replicated, the
more the costs can be expected to decline, through the enhanced efficiencies and special
ization that flow from market expansion. Far from a cost that must be imposed on a devel
oping country through an ethic of ‘zero growth’, it is in fact a business opportunity that
offers profits in one country after another through saving water as a precious resource by
recirculating and treating the waste water as it is generated. As a resource-saving initia
tive it recovers its public investment through the exports facilitated by firms locating
their operations in the industrial park involved. This is the circular economy in action.
10.5 Why Would any Country Wish to Continue

with Fossil-Fuelled Development?
Compare the gains to be won through the greening approaches discussed above with
maintaining the status quo based on fossil fuels and the linear economy. To pursue a con
ventional fossil-fuel pathway, a country would have to jeopardize its energy security by
maintaining dependence on oil, gas, or coal imports, at arbitrary prices and as an increas
ing burden on the balance of payments. As the world approaches peak oil or peak gas in
individual oil and gas fields (as it has done repeatedly in successive fields) so the de
mands for more extreme, dangerous, and costly extraction and transport processes multi
ply. As everyone knows, oil and gas deposits are arbitrarily scattered around the world,
and the discovery of deposits represents a windfall opportunity. (p. 278) By contrast, the
absence of these fossil-fuel resources (as in the case of Japan) makes the country exces
sively dependent on imports from a few sensitive countries, thus reducing energy securi
ty. In the case of Japan in the 1930s, America’s tightening grip on Japan’s oil supplies
eventually led to the Pearl Harbor attacks and the bloodbath of the Pacific War. Oil wars
Page 11 of 20
were the curse of the twentieth century, and they promise to be an even worse curse in
the twenty-first unless industrializing countries succeed in weaning themselves off fossil
fuels and shifting decisively towards renewable energy sources. And then there is the
huge environmental and pollution load associated with the oil and fossil-fuel industries
themselves—the dirtiest industries on the planet. Oil leakages are the bane of oil produc
tion everywhere, quite apart from the shocking oil pollution episodes involved in tanker
collisions and accidents, of which the Exxon Valdez oil spill may be taken as exemplary.
One common refrain in the anti-renewables discourse is that it would cost too much, or
require too many resources, to build a manufacturing-based energy system to match the
reach and scale of the existing fossil fuel-based system. Given the costs and suicidal
trends associated with continuing with fossil fuels, this is a strange argument. But let us
meet it head on. Would the costs of building manufacturing industries to produce all the
solar PV cells, all the wind turbines, and all the batteries needed for a 100 per cent tran
sition to a renewables future exceed the costs involved in continuing to invest in refiner
ies, mines, oil drilling platforms, tankers, pipelines, and distribution systems as at
present? And don’t forget to add in the hospital and health care costs incurred in treating
the victims of fossil fuel-related toxic poisoning and death—including lung cancer, bron
chitis, and other debilitating conditions.
Several research projects have been devoted to proving that the costs—both financial and
resource-based—of building a 100 per cent renewables-based energy system are contain
able—and that the transition is therefore feasible and practicable.17 No further credibility
can be attached to fears that we might run out of silicon, or that we would have to cover
the world’s deserts with black silicon panels. No such estimates are available as yet of ef
fecting a global transition from the linear economy to the circular economy—but they
should be conducted as a matter of public urgency.18
Viewed from this perspective, one has to wonder why any country would wish to perse
vere with the fossil fuel and linear economy status quo. And then reason dawns: this is
not a rational choice made by well-informed countries, but an outcome of the incumbents
continuing with their ‘business as usual’ and their extraction of rents from their past in
vestments. It is good to know that our future is in such safe hands. (p. 279)
10.6 Greening as Enhancing Energy and Re

source Security: Working with the Differential
Principle
If both a ‘climate perspective’ and a ‘green growth perspective’ end up favouring the de
carbonization of the economy and reducing its ecological footprint, what then is the dif
ference between them? This is an important and legitimate question—to which there is an
important and legitimate answer. The ‘climate perspective’ sets a standard of zero carbon
emissions as the ultimate goal, so that any single step towards this goal has to be viewed
as minor, until the goal is close. Or it sets a standard of ‘zero growth’ in order to reduce
Page 12 of 20
the economy’s ecological footprint—so that, again, any individual step towards reducing
growth has to be viewed as a minor achievement.
By contrast, the green growth strategy is all about the process of greening, not the end
result. So even a small initiative to shift the energy system towards renewables can be
counted as a positive move that, for example, improves energy security and generates lo
cal employment, which of course are positives in themselves and help to cement support
for the overall green growth strategy. Likewise, a move towards recirculating resources,
such as finding a way to turn an unwanted output into an input to another industrial
process (closing an industrial loop) can be counted as a positive move that enhances re
source security and reduces the waste generation problem, while also boosting manufac
turing linkages, employment, and profitability.
So, while China is continuing to burn a lot of coal—as numerous articles remind us—the
fact is that China is moving its energy system in a profoundly green direction, one step at
a time. It is the moving edge that is greening, with the energy system as a whole slowly
moving to become greener until a tipping point can be expected to be reached where the
entire system would be greener rather than blacker. China’s resource productivity is also
low (meaning that it generates a lot of waste). But China is taking active steps to improve
resource productivity, with moves to close industrial loops contributing to the construc
tion of a circular economy. And a tipping point will be reached when the circular flows
outnumber the linear flows, and the whole economy becomes more resource conserving
(through circularity) than resource wasting.
In my 2015 book, Greening of Capitalism, I discussed this issue under the heading of the
‘differential principle’.19 In mathematical terms, complex systems are described by their
differential equations, and the ‘point of change’ is captured by the differential. The most
succinct way of capturing how the system is changing is to find an expression for the dif
ferential. If a change in the system dynamics is required, the simplest way to (p. 280) ef
fect it is via a change in the expression for the differential. So, translating these ideas in
to real-world examples of complex physical or technoeconomic systems involves focusing
change efforts on the point where the system is already undergoing change—which in
technoeconomic systems means focusing on the point where investment is occurring.
One can try to change a large, complex system by means of an absolute change—one
whole system replaced by another. Or one can try (with more likelihood of success) to
change the system at the point of change, where the forces of change are already in evi
dence. Once business and technological systems have been built, they are highly resistant
to change. But at the point where investment is being contemplated, change can occur by
substituting one plan for another, with a stroke of the pen.
So one approach to greening industrial strategy is to focus on a system-wide absolute,

like carbon emissions, and uphold zero carbon emissions (the clean energy economy) as
the goal. This might be a legitimate goal—but as such it is quite unachievable, because
complex technoeconomic systems can change only in small steps. And this lack of achiev
ability of the absolute goal undermines confidence and support—which is why we have
Page 13 of 20
seen so little overall progress in meeting global climate targets. Another system-wide ab
solute might be to aim for the steady state (zero growth)—which again is a non-achiev
able goal in any practical sense. But that does not prevent zero-growth advocates from
protesting about a large country’s inability to conform to their absolutist expectations.
By contrast, the green growth strategy has more modest goals which are in fact achiev
able. As the steps towards achieving them accumulate, so the system starts to change in
discernible ways that attract more and more support. Investment in one small change,
when viewed as a step in the right direction, attracts more such investment.
This is indeed the huge advantage of the capitalist business system. It allows change to
occur incrementally, one investment at a time. The big changes needed are the province
of the state—as in the case of the developmental state. There is no harm in reminding
ourselves why the capitalist system has achieved unrivalled supremacy in the modern
world.
10.7 Concluding Comments

In this chapter, I have argued that by extending the scope of industrial policy from nar
row economic to broader technoeconomic considerations, encompassing energy and re
source flows, the restructuring of industrializing economies may be accelerated and at
the same time brought closer to a state of balance between economy and ecology. Using
examples from India, China, and late latecomers from around the world, the case has
been made that at the scale of transformation now under way, traditional reliance on the
fallback options of fossil fuels and linear resource flows is no longer feasible. These tradi
tional options run into problems of severe environmental spoliation: in the (p. 281) energy
domain, through particulate pollution as well as the pollution associated with continuing
extraction of fossil fuels; and in the materials/resources domain through resource exhaus
tion and waste accumulation. Worse still, these traditional options run up against geopo
litical limits when, for example, countries seek to extend their supply lines utilizing coun
tries that are subject to social conflict or even civil war. These geopolitical limits to
growth now outweigh the physical limits that were so much in vogue when the world was
wrestling with ‘limits to growth’ in the 1970s.20
Green initiatives being taken in India, China, and late latecomers, far from being viewed
as ‘special cases’, should be viewed as the general case that all countries industrializing
in the twenty-first century must engage with. There is a common view that ‘advanced’
countries constitute the general case in economics, and ‘developing’ countries a special
case. But as argued by Dudley Seers in the 1960s, and by Storm in the 2000s, the general
case is in fact the one where structural shifts are under way, and where dynamics have
greater salience than statics.21 In advanced countries, incumbents are concerned to pro
tect rents—but in the developing countries, particularly very large countries like India
and China, latecomer firms are seeking to create profits by emulating and replicating the
patterns of activity observable in the advanced world—and to do so at lower cost.
Page 14 of 20
In the case of greening, there are countless examples of such imitation and emulation,
with latecomers drawing on the huge pool of knowledge and technology available in the
advanced world. Think of Chinese firms scaling up solar PV wafer production to a mass
production industry for the first time, on their way to becoming world leaders in this sec
tor. Or Chinese wind-turbine manufacturers like Goldwind adopting such innovations as
PMDD and scaling them up for the first time, thus facilitating the global expansion of
wind power from onshore to offshore. There are also cases of leapfrogging, where the in
dustrializing countries see market opportunities and grasp them, even when this means
going beyond imitation to innovation. The case of China going it alone to UHV grid opera
tion as a means of transporting vast quantities of electric power from the interior of the
country to the eastern seaboard comes to mind as an outstanding example.
Indeed, there is a case for abandoning the term ‘greening’ altogether, once it is under
stood that the initiatives involved in enhancing energy security and resource security are
in fact well-recognized cases of industrial strategy at work. The perspective adopted in
this chapter is that greening initiatives are going to accumulate, driven by cost reduc
tions and Schumpeterian creative destruction, until the point where almost all technoeco
nomic initiatives are, axiomatically, regarded as green. At that point, there will indeed be
no need to discuss ‘green’ industrial policy as a special case—because by then it will have
become the general case. (p. 282)
References
Akoijam, Amitkumar Singh and Venni V. Krishna (2017) ‘Exploring the Jawaharlal Nehru
National Solar Mission: Impact on Innovation Ecosystem in India’, African Journal of
Science, Technology, Innovation and Development 9(5): 573–85.
Altenburg, Tilman and Dani Rodrik (2017) ‘Green Industrial Policy: Accelerating Structur
al Change towards Wealthy Green Economies’, in Tilman Altenburg and Claudia Assmann
(eds) Green Industrial Policy: Concept, Policies, Country Experiences. Geneva/Bonn: UN
Environment/German Development Institute (DIE), pp. 1–20.
Ambec, Stefan (2017) ‘Gaining Competitive Advantage with Green Policy’, in Tilman Al
tenburg and Claudia Assmann (eds) Green Industrial Policy: Concept, Policies, Country
Experiences. Geneva/Bonn: UN Environment/German Development Institute (DIE), pp.
38–49.
Bagchi, Amiya K. (2010) Colonialism and the Indian Economy. New Delhi: Oxford Univer
sity Press.
Cosbey, Aaron (2013) ‘Green Industrial Policy and the World Trading System’, Issue Brief
No. 17. Stockholm: ENTWINED/Winnipeg, Canada: IISD (International Institute for Sus
tainable Development). Available at https://www.iisd.org/library/green-industrial-pol
icy-and-world-trading-system.
Page 15 of 20
Delucchi, Mark A. and Mark Z. Jacobson (2011) ‘Providing All Global Energy with Wind,
Water, and Solar Power, Part II: Reliability, System and Transmission Costs, and Policies’,
Energy Policy 39(3): 1170–90.
Fay, Marianne, Stefane Hallegatte, Adrien Vogt-Schilb, Julie Rozenberg, Ulf Narloch, and
Tom Kerr (2015) Decarbonizing Development. Washington, DC: World Bank.
Hallegatte, Stefane, Marianne Fay, and Adrien Vogt-Schilb (2013) ‘Green Industrial Poli
cies: When and How?’. Policy Research Working Paper No. 6677. Washington, DC: World
Bank.
Jacobson, Mark Z., Mark A. Delucchi et al. (2017) ‘100% Clean and Renewable Wind, Wa
ter and Sunlight All-sector Energy Roadmaps for 139 Countries of the World’, Joule 1:
108‒21.
Karp, Larry and Megan Stevenson (2012) ‘Green Industrial Policy: Trade and Theory’.
Kumar, Charles R., Vinod Kumar, and M. A. Majid (2019) ‘Wind Energy Programme in In
dia: Emerging Energy Alternatives for Sustainable Growth’, Energy & Environment 30(7):
1135‒89.
Luetkenhorst, Wilfried, Tilman Altenburg, Anna Pegels, and Georgeta Vidican (2014)
‘Green Industrial Policy: Managing Transformation under Uncertainty’. Discussion Paper
No. 28/2014. Bonn: German Development Institute (DIE: Deutsches Institut für Entwick
lungspolitik).
Mathews, John A. (2015) Greening of Capitalism: How Asia is Driving the Next Great
Transformation. Stanford, CA: Stanford University Press.
Mathews, John A. (2016a) ‘Global Trade and Promotion of Cleantech Industry: A Post-
Paris Agenda’, Climate Policy 17(1): 102‒10.
Mathews, John A. (2016b) ‘Competing Principles Driving Energy Futures: Fossil Fuel De
carbonization vs. Manufacturing Learning Curves’, Futures (84A): 1‒11.
Mathews, John A. (2016c) ‘Latecomer Industrialization’, in Erik S. Reinert, Jayati Ghosh,

and Rainer Kattel (eds) Handbook of Alternative Theories of Economic Development.
Cheltenham: Edward Elgar, pp. 613‒36.
Mathews, John A. (2017) Global Green Shift: When CERES Meets GAIA. London: Anthem
Press. (p. 283)
Mathews, John A. (2019) ‘The Green Growth Economy as an Engine of Development: The
Case of China’, in Roger Fouquet (ed.) Handbook on Green Growth. Cheltenham: Edward
Elgar, pp. 325‒42.
Page 16 of 20
Mathews, John A. (2020) ‘The Greening of Industrial Hubs’, in Arkebe Oqubay and Justin
Yifu Lin (eds) The Oxford Handbook of Industrial Hubs and Economic Development. Ox
Mathews, John A. and Hao Tan (2014a) ‘Manufacture Renewables to Build Energy Securi
ty’, Nature 513(7517): 166‒8.
Mathews, John A. and Hao Tan (2014b) ‘A 10 Trillion Watt “Big Push” to Decarbonize the
World’s Electric Power’, Journal of Sustainable Energy Engineering 2(2): 87‒100.
Mathews, John A. and Hao Tan (2016) ‘Circular Economy: Lessons from China’, Nature
531(7595): 440‒2.
Mathews, John A., Hao Tan, and Mei-Chi Hu (2018) ‘Moving to a Circular Economy in Chi
na: Transforming Industrial Parks into Eco-industrial Parks’, California Management Re
view 60(3): 157‒81.
Meadows, Donella H., Dennis L. Meadows, Juergen Randers, and William W. Behrens
(1972) The Limits to Growth: A Report to the Club of Rome. Washington, DC: Potomac As
sociates/Universe Books.
Padilla, Emilio (2017) ‘What can Developing Countries Gain from a Green Transforma
tion?’, in Tilman Altenburg and Claudia Assmann (eds) Green Industrial Policy: Concept,
Policies, Country Experiences. Geneva/Bonn: UN Environment/German Development In
stitute (DIE), pp. 22–37.
Parthasarathi, Prasannan (2011) Why Europe Grew Rich and Asia Did Not: Global Eco
nomic Divergence. Cambridge: Cambridge University Press.
469‒91.
Schmitz, Hubert, Oliver Johnson, and Tilman Altenburg (2015) ‘Rent Management: The
Heart of Green Industrial Policy’, New Political Economy 20(6): 812‒31.
Seers, Dudley (1963) ‘The Limitations of the Special Case’, Oxford Bulletin of Economics
and Statistics 25(2): 77–98.
Shadikhodjaev, Sherzod (2018) ‘The Greening of Industrial Policy’, in Sherzod Shadikhod

jaev (ed.) Industrial Policy and the World Trade Organization. Cambridge: Cambridge Uni
Storm, Servaas (2011) ‘Economics for the “General Case”’, Development and Change
42(2): 669‒77.
Storm, Servaas and Naastepad, C. W. M. (2005) ‘Strategic Factors in Economic Develop

ment: East Asian Industrialization 1950–2003’, Development and Change 36(6): 1059‒94.
Page 17 of 20
UNEP (2011) Towards a Green Economy: Pathways to Sustainable Development and

Poverty Eradication. Nairobi, Kenya: United Nations Environment Programme.
Webster, Ken (2017) The Circular Economy: A Wealth of Flows. Cowes: Ellen Macarthur
Foundation.
Xu, Yi-chong (2019) ‘The Search for High Power in China: State Grid Corporation of Chi
na’, in Loren Brandt and Thomas G. Rawski (eds) Policy, Regulation and Innovation in
China’s Electricity and Telecom Industries. Cambridge: Cambridge University Press, pp.
221–61.
Zeng, Xianlai, John A. Mathews, and Jinhui Li (2018) ‘Urban Mining of e-Waste is Becom
ing More Cost-effective than Virgin Mining’, Environmental Science and Technology
52(8): 4835‒41.
Notes:
(1) See UNEP (2011) for the Green Economy report.
(2) For accounts of green industrial policy, which generally start with the challenge of de
carbonization to mitigate climate change, see such works as Shadikhodjaev (2018), Al
tenburg and Rodrik (2017), Ambec (2017), Padilla (2017), Fay et al. (2015), Schmitz et al.
(2015), Luetkenhorst et al. (2014), Rodrik (2014), Cosbey (2013), Hallegatte et al. (2013),
or Karp and Stevenson (2012).
(3) For my own contributions to these topics, see Mathews (2020) to Mathews (2015) in
reverse order.
(4) For a sophisticated account of East Asian industrialization strategies and the reasons
for their success, see Storm and Naastepad (2005).
(5) The alternative approach is to discuss various contingent instances of green industrial
policy, spanning such instruments as carbon taxes, cap and trade schemes and emission
allowances, energy subsidies (and reduction of fossil-fuel subsidies), environmental la
belling, and WTO-related instruments such as exemptions under GATT Article XX. For re
cent discussion of some of these instruments, see Shadikhodjaev (2018).
(6) The rise in renewables is relentless. Total renewables capacity reached 2.4 trillion
watts in 2018, according to REN21’s report ‘Renewables 2019’. We are now well into the
‘terawatt’ era of the renewables transition.
(7) See the study of urban mining in Beijing conducted with my Chinese collaborators, Dr
Zeng and Professor Li (Zeng et al., 2018).
(8) Consider, for example, the UN-inspired Green Climate Fund, established following the
2015 Paris Climate Agreement, at: https://www.greenclimate.fund/who-we-are/about-the-
fund.
Page 18 of 20
(9) See characterization of the Namibian green bond from the Climate Bonds Initiative, at:
https://www.climatebonds.net/files/files/2018-12%20NA%20Bank%20Windhoek.pdf.
(10) See the reports from the Climate Bonds Initiative, at: https://www.climatebonds.net/
resources/reports/green-bonds-market-summary-h1-2019.
(11) The literature on India’s development has recently been throwing off its ‘cultural
cringe’ with respect to Western industrialization, and is now reclaiming India’s strong
economic performance right up to the nineteenth century before colonial depredations
dismantled its sources of competitiveness. See, for example, Bagchi (2010) or
Parthasarathi (2011). On India’s pursuit of green policies, see the description and analy
sis by Simran Talwar and myself as at November 2017, ‘India’s green shift to renewables:
How fast is it happening?’, Energy Post, 22 November 2017, at: https://energypost.eu/in
dias-green-shift-to-renewables-how-fast-is-it-happening/.
(12) On India’s National Solar Mission, see recent treatments such as Akoijam and Krish
na (2017) or Kumar et al., (2019).
(13) See the description of the case (as at 28 February 2018) by the WTO, at: https://
www.wto.org/english/tratop_e/dispu_e/cases_e/ds456_e.htm.
(14) On China’s greening of its energy system, with emphasis on the way that green addi
tions to the power system now outrank black additions, see the successive articles by my
self and Hao Tan in Asia-Pacific Journal, including the most recent update, ‘The Greening
of China’s Energy System Outpaces its Further Blackening: A 2017 Update’, by John
Mathews and Carol X. Huang, with comments from Mark Selden and from Thomas Raws
ki, 1 May 2018, at: https://apjjf.org/2018/09/Mathews.html.
(15) See the account of China’s UHV innovation as a leapfrogging strategy in Xu (2019).
(16) The author visited this industrial park, Bole Lemi Phase II, in Addis Ababa, in October
2019, at the invitation of the IPDC.
(17) A prominent project devoted to demonstrating the feasibility of a 100 per cent renew
ables transition is that conducted by Mark Jacobson and Mark Delucchi at the University
of California, with a growing band of collaborators; see, for example, Delucchi and Jacob
son (2011) and Jacobson, Delucchi et al. (2017).
(18) The Ellen Macarthur Foundation in the United Kingdom has come closest to conduct
ing global cost-examination studies of the circular economy; see Webster (2017).
(19) See Mathews (2015: 270 n. 20): ‘A more general version of this idea might be called
the differential principle, in that it identifies the point of change of a system as the point
at which it is most susceptible to effective intervention in moving the system to a new tra
jectory.’
(20) The reference here is to the influential 1972 report to the Club of Rome, ‘The Limits
to Growth’ (Meadows et al., 1972).
Page 19 of 20
(21) See Seers (1963) for the original argument, and Storm (2011) for a more recent elab
oration.
John A. Mathews
John A. Mathews is professor emeritus in the Faculty of Business and Economics at

Macquarie University, Sydney. He was a professor of strategy at Macquarie Graduate
School of Management for twenty years, retiring from active teaching in 2018. From
2009–12 he held concurrently Enr Chair of Competitive Dynamics and Global Strate
gy at LUISS Gardo Carli University in Rome. For the past several years he has fo
cused on the greening of industry with an emphasis on the role of China. The year
2017 saw the publication of his Global Green Shift by Anthem Press, London. In July
2018 he was awarded the bi-annual Schumpeter Prize in recognition of his work and
most recent book.
Page 20 of 20
Globalization Narratives and Industrial Policy

Daniel E. Esser and James H. Mittelman
Competing narratives of globalization constitute parameters that both constrain industri

al policymaking and provide opportunities to negotiate alternatives. Such representations
of globalization are powerful precisely because they can become objectified as policy.
They are also compelling because the literature on industrial policy has so far largely left
them out. Focusing on the BRICS (Brazil, Russia, India, China, and South Africa), this
chapter analyses how elite narratives of globalization expressed in verbal and written
texts bear on industrial policies. It documents how the narratives of hyper-globalization
and de-globalization have gradually been supplanted by three more specific storylines: (1)
globalization as inexorable and malleable; (2) regionalization as a process that can re
fract global neo-liberalism; and (3) digital industrialization as a promise for industry-dri
ven development yet a catalyst for inequality in low- and middle-income countries. The
chapter shows how these storylines form policy openings that can be leveraged to pursue
creative, home-grown industrialization strategies.
Keywords: globalization, industrial policy, narratives, policy space, Fourth Industrial Revolution, neo-liberalism,
regionalization, BRICS
IN the contemporary era, industrial strategies are crafted amid competing narratives of
globalization. These narratives represent the parameters that both constrain industrial
policymaking and provide an opportunity to negotiate alternatives. Such representations
are powerful precisely because they can become objectified as policy. They are also com
pelling because the literature on industrial policy, taken as an ensemble, largely leaves
them out, interesting exceptions notwithstanding.1 The aims of this chapter are to fill in
this blind spot and enlarge the orbit of inquiry.
Mindful of the fluidity of discourses, we will analyse how elite narratives of globalization,
expressed in verbal and written texts, bear on industrial strategies. Our goal is to parse
out what these stories reveal. We will highlight ways in which the dominant narrative of
hyper-globalization frames analysis and action. Specifically, we will probe the manner in
which narratives of globalization are both shaped by and manifest in globalization indices
as well as in statements and speeches by senior political leaders of the five BRICS coun
Page 1 of 29
tries: Brazil, Russia, India, China, and South Africa. We will also show how the unifying
discourse of hyper-globalization has been challenged by an opposing narrative, de-global
ization, which, though less apparent in the data analysed here, is increasingly intertwined
with the former. Over the years, these narratives have gradually been supplanted by
three more specific storylines, to be traced in the pages ahead.
For the purposes of this study, the context is low- and middle-income countries in
(p. 285)
the global South from the 1990s to the present. These years are punctuated by the transi
tion to a post-Cold War order and adjustments to neo-liberalism, a set of ideas and a poli
cy framework centred on deregulation, liberalization, and privatization.
11.1 The Approach Adopted in this Chapter

To investigate the elective affinities between globalization narratives and industrial
strategies, this chapter adopts a critical political-economy perspective supported by origi
nal data. Our critical approach upsets certainties in extant knowledge sets. Two branches
of political economy and social theory are especially pertinent for examining industrial
strategies: our work bridges insights from materialist and poststructuralist traditions,
with their respective emphases on material power and discursive power. The rationale for
linking them has its roots in classical political economy and classical social theory. Both
traditions emerged during the social upheavals accompanying the rise of industrial capi
talism in Western Europe from the late eighteenth to the early twentieth centuries, as
chronicled by Smith, Ricardo, and Marx to Durkheim, Weber, and others. Our framework
also draws on feminist and postcolonial theorists’ writings on standpoint analysis, that is,
seeing global problems from different positions in social and geographical hierarchies
(Tickner, 2014).
Spoken and written words are central to this endeavour. Words condense and frame infor
mation. They overestimate some aspects of socio-economic phenomena and underesti
mate others. While words can plant ideas and generate action, they may also be a mas
querade, disguising meaning and intentions. Whether used as a form of deception or to
induce action, they are instruments of political economy that produce real consequences.
Moreover, words reflect what the French sociologist and public intellectual Pierre Bour
dieu (1984) famously called habitus: the deep-seated habits, skills, and dispositions that
serve as cornerstones of institutionalized social hierarchies. Bourdieu’s notion of habitus
is important to analyses of industrial policy because it can be read as drawing attention to
narrative voices that animate international practices. Expanding Bourdieu’s conceptual
ization, we argue that how international actors talk about industrialization, name strate
gies for bolstering it, and attach labels to its variants, seep into mindsets. The language
chosen is used to kindle or confound practice.2 The (p. 286) vocabularies are terrain for
contesting policy. They invite us to derive a sense of what mistakes to correct and where
to go: in short, how to do better policymaking.
Page 2 of 29
In dwelling on the power of verbal and written acts, we will attempt to pick up on the
subtlety of these ambiguities and anomalies. While empirical digging beneath public
statements to investigate hidden transcripts is beyond the scope of this inquiry about In
dustrial Globalization Talk,3 this chapter takes but a modest, exploratory step in that di
rection. It is our hope that it can serve as a starting point for multipronged projects com
bining interviews, ethnographies of organizations, and other types of field research that
would delve deeply into dominant‒subaltern relations and the institutions of civil society
amid the changing global division of labour and power (GDLP).
Conceptually, the narratives of hyper-globalization and de-globalization serve as nonbina

ry markers, poles on a spectrum, with many variants in between. Of course, other dis
courses are circulating. But we focus on the two widely shared narratives because of
their frequency in both the literature and the empirical data examined for this research.
It must also be stressed that they should not be construed as a dichotomy. In the coming
pages, we acknowledge sub-narratives.
In addition, we presuppose that world regions, multilateral agencies, and individual coun
tries are saturated with their external environment: in particular, by globalizing forces, al
beit accompanied by counterforces such as resurgent nationalism and authoritarian pop
ulism.
Methodologically, we offer textual analysis of intersubjective framing on the basis of doc

umentary research. Evidence is garnered from governmental and intergovernmental re
ports issued between 2003 and 2019, a period spanning the 2008 global financial crisis
and the Eurozone debt crisis, the early years of modest recovery, and the rise of counter-
globalist forms of national populism in different world regions. Our research integrates
qualitative and quantitative data from think tanks and private corporations with state
ments and reports by multilateral organizations and BRICS leaders. These are drawn
from online media outlets and the World Economic Forum (WEF, a private, non-profit
body that brings together global elites at annual gatherings usually held in Davos,
Switzerland), keynote speeches and reports by the five members of the BRICS Business
Council, and annual reports of the sole worldwide specialized agency devoted to industri
al development, the United Nations Industrial Development Organization (UNIDO). All
the documents were collected between January and June 2019 through extensive Internet
searches and by accessing publicly available online repositories.
Our empirical analysis thus proceeds selectively. As Fernando Santiago (see Chapter 26,
this volume) demonstrates, the BRICS are a heterogeneous grouping with regard to their
productivity and integration in the global political economy. This variation notwithstand
ing, they are a visible expression of ongoing shifts in the GDLP. Admittedly, some of the
data in our study are patchy—they are not as (p. 287) prodigious as we would like.4
Nevertheless, we deem them representative of national and international elite narratives
of globalization as they relate to industrial policy. Whereas additional documents generat
ed by governments and multilateral organizations could broaden our work, we propose
avenues for future inquiry in section 11.6 of this chapter.
Page 3 of 29
By grounding this research in empirical findings, we seek to gain conceptual clarity,

throw light on contending narratives, and inform industrial policies in the global South.
The chapter therefore proceeds as follows. After advancing our theoretical framework
and tracking the rival narratives of hyper-globalization and de-globalization, we turn to a
first level of generality. It is the more wide-ranging narrative and maps the ways in which
globalization indices epitomize hyper-globalization. The second level evolved from the
first and describes specific trends, particularly narrative framing of industrial policy by
the leaders of multilateral organizations and the BRICS. Today’s narratives have their
roots in the hyper-globalist and de-globalization paths that continue to be frequently trav
elled.5
The two levels of narrativity alike embody an epistemology: a theory of how to know the
world. It creates a problematic about what is ‘askable’ and how to go about responding,
thereby influencing what is to be said and done. Supposedly a means, it becomes the ob
jective of policymaking. To pull together these themes, we conclude by offering reflec
tions on the emphases and silences in the data, inferring opportunities on this basis for
industrial policymaking, and suggesting a research itinerary.
11.2 Theoretical Perspectives

To link ideas, symbols, and language with material forces, Antonio Gramsci (1971) formu
lated the concept of ideological hegemony: a fit between consensus and coercion in which
consensus is the dominant element. It would seem that neo-liberalism and developmental
ism constitute ideological projects in a Gramscian sense (Mkandawire, 2001). In the post
colonial era, these ideologies formed the prevailing wisdom about the role of the state in
much of Africa and elsewhere (Whitfield et al., 2015). Their implications for industrial pol
icy have been unambiguous: a broad consensus, with economic and diplomatic coercion
in the backdrop, is reflected in narratives about industrializing by forerunners and late
comers, catching up, climbing a ladder, realizing comparative advantages, and leapfrog
ging (Chang, 2002; Nayyar, 2013; Behuria and Goodfellow, 2019). Within this consensus,
there are of course serious internal differences.
Intervening in these debates, Akira Suehiro (2008: 319) cautions about the pitfalls of
‘thinking in old-fashioned catch-up terms’ and notes that it is difficult to shed this (p. 288)
mindset. Suehiro traces the process of ‘technology formation’—that is, upgrading learn
ing capabilities in industry and bolstering linkages between industrialization and trade.
His conclusion dovetails with our argument in the previous paragraph: ‘The core of the
matter is ideological, not technological’ (Suehiro, 2008: 323; emphasis in original).
To be sure, industrialization requires even more than these frames suggest. The chal
lenge is not to climb a ladder rung by rung to a pre-ordained destination decided by the
organizers of a race, mostly based in affluent countries. The ground beneath the mythical
track and hypothetical ladder can be unstable, even tremulous. The landscape is pro
duced by historical structures. It cannot be gleaned from analogical reasoning, which
may spread misconceptions and generate faulty narratives. To the extent that analogies
Page 4 of 29
depict structured hierarchies, they facilitate analysis; but history is more complex than
collective metaphors could possibly capture.
For example, the image of climbing a ladder is a linear view of capitalist development. In
fact, the route is circuitous; the process, dysrhythmic. The imaginary of ascendance omits
zigzagging and swerving. It is mistaken to neglect multidirectional tendencies that can be
deeply contradictory. Nonetheless, the conventional wisdom is binary. It intimates verti
cality: upward or downward movement. This picture deflects attention from the impor
tance of horizontal and lateral moves—spreading operations, building networks, and
forming alliances. In short, it induces two-dimensional thinking when multidimensional
analysis is needed to grasp the messiness of specific contexts.
Falling into these narrative traps can come about through the mimetic repetition of key
words, a pivotal point to which we will return. In view of this relentless drumbeat, our at
tention centres on how stakeholders frame industrialization. The focus on the discursive
moves by which industrial strategies are established, maintained, modified, or abandoned
is crucial because a substantial corpus of literature on industrial policy privileges eco
nomic factors and understates the salience of intersubjectivity and social forces.
In an attempt to countervail economistic propensities, one of us (Mittelman, 2011: 187‒

204) has sought to advance the Smithian/Ricardian international division of labour (IDL)
theories, which emphasize the importance of the specialization of function and the com
parative advantage of trading products for which costs are relatively low. Efforts to sup
plement them in a new international division of labour (NIDL) describe the devolution of
certain manufacturing industries from advanced capitalist countries to the global South
(Fröbel et al., 1980). This spatial reorganization of production, which began in the second
half of the twentieth century, merits fresh analysis since global monopolization further
concentrates industries geographically (see Chapter 8, this volume) argue. Beyond the
NIDL, the concept of the GDLP incorporates structural power, constituted by a mix of ob
jective and intersubjective factors that transcend economic spheres. It links big powerful
institutions such as transnational corporations (TNCs) to social structure: the intersec
tionality of class, gender, generations, race and ethnicity, and sexual orientation.
Building on this conceptualization, we maintain that imaginaries of globalization embod

ied in competing narratives render some strategies more promising than others. (p. 289)
The menu of narratives provides options for industrial policy and begets societal impacts,
shaping countries’ positions in the global economy and their sociopolitical development.
To move beyond structuralism and economism in mainstream narratives (with important

correctives, e.g. Hirschman, 1958, 1971, [1981] 2008; Amsden, 1989, 2001), we want to
explore alternative approaches. The pay-off should lie in explanatory power as well as
suggestions for regulatory reform. Towards these ends, let’s now endeavour to disentan
gle global narratives.
Page 5 of 29
11.3 Hyper-Globalization and De-Globalization

Widely deployed by governance agencies as well as by scholars and other policy intellec
tuals, the narrative of hyper-globalization embraces propellants in the global political
economy. It is characterized by an acceleration in cross-border flows of capital, technolo
gy, population, and cultural products. Its purveyors call attention to the extent to which
the speed and reach of contemporary economic globalization differ significantly from the
pace and expanse of earlier phases.
The hyper-globalization narrative emerged in the 1990s and coincided with the inception
of the concept of global governance. Simultaneously pioneered by scholars such as James
Rosenau and Ernst-Otto Czempiel (1992) and policy intellectuals in international organi
zations, the umbrella concept of global governance premises increasing complexity in the
GDLP. In various formulations, hyper-globalization and global governance are kneaded to
gether. Taking into account Asian experiences with industrialization during the Cold War,
heterodox thinkers affirm that the developmental state must play a major role in animat
ing economic development by nurturing domestic industries to the point that their inser
tion into global value chains produces profits outweighing the initial investment and the
costs.
In its early renderings, the notion of hyper-globalization depicted a borderless world, one
with heightened denationalization and a proliferation of actors (e.g. Ohmae, 1999; cf.
Dicken, 2015). Globalization was said to be an inexorable process that spurs capital accu
mulation. For some authors (e.g. Dollar and Kraay, 2004), deep integration in the global
economy trends towards faster growth and poverty reduction and constitutes a win‒win
scenario. Yet questions about inequality on a world scale and the ethics of the GDLP
abound: whose rules and which discourses are inscribed in it? And who is responsible for
global economic governance?
Other authors have attempted to sharpen the hyper-globalist claim, positing that with
growing deterritorialization, the old Westphalian state system is increasingly porous. As
an organizing concept, the Westphalian principle has limited utility in accommodating
nongovernmental activities and accounting for ways that the state itself and sovereignty
are being reconstituted. New actors have come on the scene with (p. 290) a backlash
against globalizing forces, in particular from social movements that have mounted
protests targeting multilateral agencies such as the World Trade Organization (WTO), the
World Bank, and the International Monetary Fund. It has become apparent that the
North‒South divide and centre/periphery distinction lack sufficient explanatory value for
elucidating mobility in parts of the global South and persistent marginalization in some
areas (see Held et al., 1999: 3‒5; Scholte, 2005).
In the second decade of the twenty-first century, analysts updated their data, but many
observers cling to the construct of hyper-globalization. Researchers based in internation
al institutions such as the United Nations Conference on Trade and Development (UNC
TAD, 2018: iv‒v, ix, xiii, 50) and at think tanks, among them Arvind Subramanian and
Page 6 of 29
Martin Kessler (2013) at the Peterson Institute for International Economics and Yuefen Li
(2018: passim) at the South Centre, still regard rising global flows as indicators of the
staying power of hyper-globalization. This is despite nuanced readings of features of glob
alization—for example, the changing meaning of sovereignty and citizenship (Sassen,
1996, 1998)—and trenchant critiques by certain trade economists. Dani Rodrik (2018), for
instance, notes a groundswell against the hyper form of globalization, efforts to level the
playing field, and the rise of populisms, partly as a reaction to the impact of neo-liberal
capitalism.
In the current capitalist era, the tides of globalization tack back and wash forward. It
would be impossible to trace a neat sequential progression. Yet analysts can chart toggles
between retreats and advances, tensions and challenges, that spawn reconstituted narra
tives.
From the 1990s, scholar-activists began to forge the counter-narrative of de-globaliza

tion. An avant-garde book titled Delinking (1990) by Samir Amin, an Egyptian-French in
tellectual, laid the groundwork for careful research on this theme carried out in myriad
contexts. Amin distinguishes delinking from autarky and withdrawal from the worldwide
industrial, trade, and financial systems. Delinking is a strategy for capturing control of
the national economy—an auto-centric programme for reconstructing the postcolonial
economy. Meanwhile, the aim at the international level is to work with allies to shape a
polycentric system of power.
Amin and like-minded thinkers such as Walden Bello (2005), a Filipino professor and for
mer member of the House of Representatives of the Philippines, have put forward ideas
for transforming a political economy with due regard for the specificity of individual
countries in the global South. In Karl Polanyi’s sense (1957), the goal is to re-embed the
economy and institutions in society rather than to allow the economy to drive society. Ac
cordingly, in Amin and Bello’s usage, the lodestar of de-globalization is disengaging from,
and then selectively redialling into, the global political economy.
Seemingly similar to these propositions, populists have formulated nationalist economic

agendas. They argue in favour of restructuring terms of trade, levying tariffs, and protect
ing the domestic economy. In Indian Prime Minister Narendra Modi’s words: ‘As opposed
to globalization, the forces of protectionism are emerging. Their intent is to not only safe
guard themselves from globalization, but also to alter the natural flow of (p. 291)
globalization’ (Government of India, 2018). This thrust mirrors similar developments in
the global North where diverse economic nationalists, religious fundamentalists, and rad
ical environmentalists would use the national state as a shield or barrier to constrict glob
alization. But major differences between the de-globalization narrative and populist
stances on the right are palpable. For instance, while the former propagates the idea of
decoupling to address economic inequalities associated with soft regulation of global cap
ital flows, the latter considers it mainly to reinvigorate nationalist discourses and policies.
Moreover, whereas the former stresses governments’ regulatory power vis-à-vis global fi
nancial capital as a critical component of a just GDLP, the latter calls for a reduction in
Page 7 of 29
human rather than financial flows even if reducing international mobility is detrimental to
industrialization because it restricts the movement of labour and ideas.
In light of these currents, we will next look at empirical evidence for how contemporary
elite narratives have evolved and, one might claim, materialize as practices and policies.
We begin with how globalization indices produced by different nongovernmental and cor
porate actors provide scaffolding for dominant narratives of globalization.
11.4 Narratives in Action I: Globalization

Indices
Both narratives—hyper-globalization and de-globalization—offer interpretations of un
even development and subtle shifts in the global political economy. The most recent ag
gregate data for the period since the 2008 financial crisis show neither a large rise nor a
substantial decline in the overall level of globalization. Nor do the data suggest a pattern
wherein countries are voluntarily delinking from globalizing forces. However, the indices
rest on the taken-for-granted premise that countries can be ranked and compared by the
same measures.
In response, we find that the production of global indices itself incorporates the prepon
derance of but one narrative: hyper-globalization. To sustain this proposition, we juxta
pose two global indices, the KOF Globalization Index issued by the KOF Swiss Economic
Institute (2018a) and the DHL Global Connectedness Index DHL (2018) compiled by New
York University and the Barcelona-based IESE business schools.6
According to the 2018 KOF Globalization Index (KOF Swiss Economic Institute,
(p. 292)
2018b), worldwide globalization increased between 1990 and 2007 but slowed during the
2008 financial crisis and in the recession that ensued. Disaggregating globalization by di
mension, the index suggests that despite a slight upward tick in 2016, aggregate econom
ic globalization has flat-lined.7 Although financial globalization continued to mount, trade
integration reportedly receded. Yet this overall deceleration has left the hyper-globalist
narrative largely intact, as we will indicate in section 11.5 of this chapter. While the social
dimensions of globalization increased slightly before evening out since 2013, global infor
mation flows and political globalization vectored upward.
The 2018 KOF Globalization Index lists Switzerland as the most globalized country, fol
lowed by other small European countries such as the Netherlands and Belgium, whereas
landlocked and island countries in the global South like the Central African Republic and
the Comoros are ostensibly the least globalized. The BRICS rank considerably below
small European countries. Brazil occupies the 100th spot in the 2018 KOF Overall Global
ization Index; Russia, 49th; India, 96th; China, 79th; and South Africa, 59th. As stated by
the organizations producing the ranking (KOF Swiss Economic Institute, 2018b; Bertels
mann Stiftung, 2018), one factor that explains these ratings is country size. The BRICS
are relatively large in terms of population and land mass, and small countries and their
Page 8 of 29
neighbours tend to be more interdependent than many big ones. Second, not only is the
BRICS’ level of intra-regional integration less than that of the most highly ranked small
countries, but greater economic strictures (trade barriers, capital controls, etc.) in the
BRICS are likely to be another element in this ordering (Weiß et al., 2018). As the relative
ease of flows within the European Union illustrates, comparative regionalism may help
account for these differences. In this vein, regionalism may be construed both as a com
ponent of globalization and a response and challenge to it, an issue that we will revisit.
Third, the KOF Globalization Index demonstrates that advanced economies in general are
more tied to global capital flows than are emerging economies. For China and the other
BRICS, domestic flows still exceed cross-border transactions.
The DHL Global Connectedness Index measures globalization in terms of the depth and
breadth of cross-border flows of trade, capital, information, and people. The most recent
DHL Index (Altman et al., 2018) indicates that for the world as a whole, the level of global
connectedness crested at a record high in 2017. Although not entirely sharing (p. 293)
KOF’s determination that globalization is declining, the DHL Index attests that on actual
levels of global connectedness, the world is, in fact, not as globalized as many observers
believe. The DHL finds that a larger portion of flows of trade, capital, information, and
people is domestic rather than international. Equally, in contrast to the KOF Globalization
Index, the DHL Index reveals that this is apparent for not only countries on the low end of
tallies of globalization indicators but also for those at the top. Notwithstanding the ways
in which technological globalization compresses time and space, sovereignty and distance
continue to constrain cross-border flows.
The DHL Index ranks the Netherlands as leading all countries in global connectivity. It is
followed in the top ten by mostly European countries in the following order: Singapore,
Switzerland, Belgium, the United Arab Emirates, Ireland, Luxembourg, Denmark, the
United Kingdom, and Germany. They hardly constitute a homogeneous grouping as the
list comprises a city-state (Singapore), a quasi-city-state (Luxembourg), and relatively
large national territories (UAE, Germany). The DHL Index also clusters knowledge-based
industrial economies such as Singapore, Switzerland, and Denmark with service-based
ones like the United Kingdom and Ireland, and the oil revenue-dependent UAE. Moreover,
it evinces substantial variation in flows over time, across locales, between countries, and
within regions. And despite complaints in Washington about the scale and impact of glob
alization on the US domestic economy, the DHL Index reports that the United States lags
behind many other advanced countries in this regard, ranking it 30th among 169 coun
tries in overall global connectedness. Considering that, of the major global tech players,
fourteen are based in the United States, three in the European Union, three in China,
four elsewhere in Asia, and only one in Africa (Jomo and Chowdry, 2018, drawing on UNC
TAD, 2018), the index omits important dimensions of global aggregations of power. Global
connectedness varies considerably by both country and region; it cannot be reduced to
one universal pattern.
Page 9 of 29
But the deeper issue about these indices is not just that they offer nominally factual in
sights. Their use betrays a philosophical embrace of positivism that resonates with the
tenets of neo-liberalism. The underlying assumption is that data may be objectively mea
sured, are independent of subjective processes, and that the findings—rankings—can be
replicated and are falsifiable. In objectifying the measures, complex social and political
dynamics are reduced to numbers. The presupposition is that brute data are ‘given’, not,
as critical theorists would have it, produced. The question of production begets questions
not ordinarily asked in the field of industrial policy: Who authorizes the production of
measures, who selects the producers, who pays them, and to whom are they accountable?
If these searching questions are sidelined, are the measures—indicators—an ideological
construct adopted by myriad policymakers? In this sense, dominant narratives may be un
derstood as ideologies of power. And for this reason, they are theoretically and empirical
ly intriguing.
The empirical evidence adduced above shows that the levels of global connectedness lie
somewhere between what the enthusiasts of hyper-globalization claim and what the pro
ponents of de-globalization seek. In other words, even if globalization trends were really
measurable, neither narrative would adequately depict the actualities. To pursue (p. 294)
these issues, let us now pivot to how these two narratives and their variants are reflected
in other forms of policy texts and explore what this implies for industrialization strategies
in the global South.
11.5 Narratives in Action II: Three Storylines

In segueing from globalization trends to industrialization strategies amid competing nar
ratives of globalization, it is important to be cognizant of narrative entrapment. Ob
servers can be enthralled with imagery not of their own making. These images often take
the form of language, mental pictures of social phenomena, and paradigms that filter
ideas (Mittelman, 2011: 143, citing Shotter, 1993: 26‒31).
Much like speeches and statements by heads of state and ministers at global conferences,
multilateral organizations’ flagship reports are political documents that need to be read
as such (Esser and Williams, 2014). They are variegated and do not necessarily present
coherent policy positions, but rather exemplify the dynamics of global policymaking. They
also reaffirm dominant global narratives, such as the Millennium Development Goals and
their 2015 successors, the Sustainable Development Goals, that project authority, purport
mandates, and seek legitimacy. This is not to suggest that all stories are directly action in
ducing or option excluding (Rasche and Esser, 2007). But, to underscore a point made at
the outset of this chapter, storylines have the potential to be converted into strategies and
policies.
Page 10 of 29
11.5.1 Globalization as Inexorable yet Malleable
The first storyline scrutinized here frames globalization as unavoidable on the one hand,
but susceptible to unilateral interventions by some national governments on the other.
Thus, in a 2017 speech, President Xi Jinping presented the Chinese perspective as recog
nizing economic globalization as ‘a double-edged sword’ (Xi, 2017) and alluded to ‘a time
when China also had doubts about economic globalization’ (Xi, 2017). Xi nonetheless
posited ‘that integration into the global economy is a historical trend’ since ‘the global
economy is the big ocean that you cannot escape from’ (Xi, 2017). Weighing into ‘a heat
ed debate on economic globalization [amid] a world of contradictions’, he concluded that
for China, the best way forward is to ‘act pro-actively and manage economic globalization
as appropriate so as to release its positive impact and rebalance the process of economic
globalization’ (Xi, 2017). Xi also stressed that ‘this round of industrial revolution will pro
duce extensive and far-reaching impacts such as growing inequality, particularly the pos
sible widening gap between return on capital and return (p. 295) on labour…This is the
biggest challenge facing the world today. It is also what is behind the social turmoil in
some countries’ (Xi, 2017).
In a similar vein, speaking at a 2019 conference in Washington, DC, Brazilian Vice Presi
dent Hamilton Mourão attributed an ‘international environment of greater instability and
competition’ to the ‘triumph of neo-liberal ideas’ (Wilson Center, 2019). In his view: ‘To
day all nations, even the most developed ones, are kept in a permanent state of
alert’ (Wilson Center, 2019). So, too, Rob Davies, South Africa’s former minister of trade
and industry, warned about ‘the extreme uncertainty of the current and immediately fore
seeable global environment’ (Department of Trade and Industry [DTI], 2017: 3), specifi
cally the ‘volatile mix of geopolitical uncertainties and risks and the emergence (and ap
parently growing traction) of populist backlashes in the United States and much of Eu
rope, partly fuelled by the unchecked exclusion and inequality which the current model of
globalisation has engendered’ (DTI, 2017).8 And during the 2018 WEF meeting, India’s
prime minister paraphrased Mahatma Gandhi: ‘I do not want that the walls and windows
of my house are closed from all sides…But at the same time it will not be acceptable to
me that this wind blows away my feet off the ground’ (Government of India, 2018).
It thus appears that even though BRICS leaders embrace a wide range of political ideolo
gies, they converge on a narrative of economic globalization as a pervasive force subject
to a set of pressures on their societies. They largely agree that the rise of protectionist
stances as outgrowths of chauvinistic populism risks squandering development opportuni
ties from the flow of goods and human capital and achieves little in terms of mitigating
the pernicious social effects of unfettered globalization. These impacts and options are
recognized across the global South, from Latin America to East Asia, with important dif
ferences in the political implications.
Meanwhile, a joint communiqué by the BRICS’ trade ministers ‘recognized the impor
tance of preserving policy space to promote industrialization, industrial upgrading, and
value addition as a core pillar for structural transformation and sustainable development
Page 11 of 29
and BRICS countries’ integration into the global economy’ (BRICS, 2016a: 2). This need
to defend and reclaim national policy space was reiterated during the eighth BRICS Sum
mit, which affirmed ‘measures that support greater participation, value addition, and up
ward mobility in Global Value Chains of [BRICS] firms including through the preservation
of policy space to promote industrial development’ (BRICS, 2016b: 5).
This narrative shift towards a more nuanced assessment of globalization’s enabling and
deleterious effects that extend beyond the economic sphere, along with the appeal of se
lective protectionism by countries confident in its successful implementation, (p. 296) be
comes equally apparent in UNIDO’s annual reports since 2003.9 While recognizing ‘mar
ginalization in today’s globalized world’ (UNIDO, 2003, 2004, 2005), UNIDO nonetheless
counselled that ‘[i]ncreased globalization and market liberalization provide great oppor
tunities for developing countries and economies in transition to trade their way out of
poverty’ (UNIDO, 2006: 41).
The emphasis changed in the aftermath of the 2008 global financial crisis. According to
UNIDO (2009: 2), ‘[a] number of existing and emerging megatrends—including…unequal
globalization—converged and together wreaked havoc on the poorest countries that are
home to the bottom billion’. Although the same report noted a ‘multitude of gains to be
achieved from globalization’ (2009: 68), this formulation soon switched to another frame:
‘rapid changes brought about by globalization’ (UNIDO, 2010: 56), which was cast as a
‘new hurdle’ (UNIDO, 2011: 62) for the least developed countries. Four years later, UNI
DO (2015: 13) averred, ‘While globalization and trade liberalization offer undeniable op
portunities…they also expose markets to rapid changes in technologies, consumer prefer
ences, and the pressures of competition.’ The first storyline shows how narratives of glob
alization evolve and expand. Initially schematic frames become more multifaceted. Narra
tives of globalization spread globally but are also localized and modified to fit specific na
tional and regional conditions and concerns.10 This evolution across multiple scales of
policymaking is central to comprehending the second storyline emerging from the data.
11.5.2 Regionalization as a Process that can Refract Global Neo-liber

alism
The second storyline spotlights the promises of globalization as a crucial element in re
gional integration. Let us again illustrate with a statement by Chinese President Xi:
We need to achieve a coordinated development of Asian countries…This will help

bring down the logistic costs, create demand and employment, give play to our
comparative and latecomer advantages, secure favourable position [sic] in the
global supply chain, industrial chain, and value chain, raise our comprehensive
competitiveness, and usher in a new situation in Asian development. (People’s Re
public of China [PRC], 2014)
This regional focus on Asia by the region’s largest economic power parallels the South
African minister of trade and industry’s reframing globalization as African (p. 297) region
alization. Arguing that South Africa had ‘been supporting the industrialization of the
Page 12 of 29
African continent’ (Chatham House, 2017) and reflecting his government’s pursuit of a
‘broader-based industrialization [to] enhance[e] the participation of historically disadvan
taged people and marginalized regions in the mainstream of the industrial
economy’ (Government of India, 2012: 154), he underlined that South Africa’s govern
ment ‘want[ed] to see the development of value chains on a regional basis’ (Chatham
House, 2017; see also Africa Report, 2018). Brazilian Vice President Mourão made similar
claims concerning the role of Brazil in Latin America when he stated: ‘Our region wants a
better place on the great geopolitical chessboard of globalization and global value
chains’ (Wilson Center, 2019).
Another facet of this second storyline is BRICS members states’ persistent emphasis on
inter-regional integration. An early example is the 2011 Sanya Declaration issued at the
third BRICS Summit, which pledged to ‘support infrastructure development in Africa and
its industrialization within framework [sic] of the New Partnership for Africa’s Develop
ment (NEPAD)’ (BRICS, 2011: 4). Two years later, the fifth BRICS Summit ‘took place un
der the overarching theme, “BRICS and Africa: Partnership for Development, Integration
and Industrialisation”’(BRICS, 2013: 1). In addition, both summits adopted resolutions ‘to
increase…engagement and cooperation with non-BRICS countries’ (BRICS, 2013: 2). This
commitment has been reiterated consistently (e.g. BRICS, 2016b).
Extending the storyline to the nongovernmental realm, a statement from the 2013 BRICS
Business Forum reinforced the regional focus on Africa ‘in order to contribute to its de
velopment and to expand trade links between Africa and BRICS’ (noted in DTI, 2013: 1).
BRICS Business Council chairman Onkar Kanwar complemented the emphasis on African
infrastructure development by demanding joint BRICS economic zones (Kanwar, 2013)
and greater ‘intra-BRICS trade using local currencies’ (Kanwar, 2014). And in 2016, a
joint statement by the council reaffirmed this commitment by promoting the ‘develop
ment of inter-regional business ties’ (BRICS Business Council, 2016: 2). This practical em
phasis on the modalities of regional as well as inter-regional integration stands in con
trast to the final storyline emanating from the data: namely, the importance of digital
technologies for industrial development in the global South amid a continuing dearth of
actionable strategies.
11.5.3 Digital Industrialization: Distant Promise or Catalyst of

Inequality?
The articulation of the third storyline hinges on what some have termed the Fourth Indus
trial Revolution (4IR) and its implications for the global South.11 In this (p. 298) connec
tion, Klaus Schwab, the founder and executive chairman of the WEF, maintains that the
evolution of globalization and the genesis of 4IR are interrelated. He holds that the first
industrial revolution brought steam trains, steamships, and the industrialization of weav
ing and mining; the second, the modern assembly line, the automobile, and the airplane;
and the third, the computer and early digitalization. The fourth is marked by artificial in
Page 13 of 29
telligence (AI), autonomous vehicles, and the Internet of things: the integration of multi
ple technologies into one whole system (Schwab, 2017, 2019).
BRICS leaders have embraced the 4IR narrative. For instance, the 2014 Fortaleza Decla
ration capping the BRICS’ sixth Summit stated: ‘We believe that ICTs [information and
communications technologies] should provide instruments to foster sustainable economic
progress and social inclusion’ (BRICS, 2014: 10). At a 2016 regional WEF meeting, Chi
nese Premier Li Keqiang argued: ‘The advent of a new round of technological and indus
trial revolution has provided a historical opportunity for … the upgrading of traditional
industries’ (PRC, 2016). And one year later, President Xi (2017) reiterated this narrative
in his remarks at the annual meeting in Davos, urging world leaders to ‘seize opportuni
ties presented by the new round of industrial revolution and digital economy’. Echoing
the Fortaleza Declaration, Indian Prime Minister Modi expounded at the 2018 WEF meet
ing: ‘Technology and digital revolution increases [sic] the likelihood of…new solutions,
which could help us to tackle the old problems of poverty and unemployment in a new
manner.’ So, too, the Russian Federation’s minister of trade and industry referred to the
present era as ‘a time when manufacturing must adapt to the demands and the opportuni
ties created by the Fourth Industrial Revolution’ (Roscongress Foundation, 2018). Along
these lines, the WEF’s Centre for the Fourth Industrial Revolution Network pursues a
mission to hone governance frameworks and protocols and has started to work on
projects in various domains, such as in AI and machine learning (WEF, 2019: 21).
However, in the data analysed in this chapter, assertions about the potential of digitaliza
tion for the global South remain largely hypothetical. True, mutual gains seem likely
when, for instance, South Korea and Singapore join forces to facilitate product develop
ment and marketization for start-ups (Korea.net, 2018). Yet, how national governments
lacking influence on the world stage and in countries with weak links to global value
chains are to engage digitalization is an open question, one acknowledged in the global
South.
Addressing this issue at a BRICS Business Council meeting in 2019, South Africa’s minis
ter of international relations and cooperation recognized ‘the challenges arising from the
Fourth Industrial Revolution’ (Sisulu, 2019). Remarkably, a 2015 UNIDO report was the
only document in our database that presented a discordant view: namely, that new tech
nologies are no panacea for industrial development, for they have ‘often failed to trans
late into concrete growth opportunities’ (UNIDO, 2015: 27).
But what can governments of industrializing countries do to level the 4IR playing field? A
joint initiative by the Indian and South African delegations to the WTO is emblematic of
future opportunities as well as current limitations. Acting on their shared position that
‘[t]ariffs are instruments of industrial policy and have implications (p. 299) for capital ac
cumulation, technological change, productivity growth, and employment’ (Government of
India, 2012: 154) and citing a UN study on the implications of growing global trade in e-
commerce (UNCTAD, 2019), the two delegations advocated an end to the WTO’s 1998
moratorium on imposing customs duties on electronic transmissions in order to render
Page 14 of 29
the latter a source of domestic revenues (WTO, 2019). Urging the global regulatory body
‘to understand the ruinous impact of digitization on SMEs [small and medium-sized enter
prises] in developing countries and LDCs [least developed countries]’ (WTO, 2019: n.p.),
the two signatories argued:
The moratorium will negatively impact the efforts of many developing countries,
which are laggards as far as digital industrialization is concerned, to industrialize
digitally. It could also undermine existing industries. Tariffs play an important role
in protecting infant domestic industries from more established overseas competi
tors until they have attained competitiveness and economies of scale. (WTO, 2019:
n.p.)
The joint memo concludes that ‘for designing such policies [to protect infant digital indus
tries], it is extremely important for developing countries to preserve policy and regulatory
space in the WTO’ (WTO, 2019: n.p.). Like statements in favour of selective protectionism
that exemplify the first storyline’s ongoing evolution, such calls for industry-specific pro
tectionism are at odds with the neo-liberal vision of 4IR as a global process unfolding
preferably without significant regulatory constraints. On the one hand, they show that
some governments in the global South value the potential of digitalized trade and of in
dustries for building technological scaffolding. On the other hand, they clearly recognize
that unregulated global digital industrialization is unlikely to foster economic growth and
welfare in countries where entrepreneurial capital has often been restrained, requisite in
frastructure is still nascent, and productivity, whether digital or analogue, is held back by
limited knowledge transfers.
Taken together, the three storylines provide a wide range of examples of how the narra
tives of hyper-globalization and de-globalization are adopted and used, but also supplant
ed and challenged. In this chapter’s final section, section 11.6, we interpret the emphases
and silences apparent in the data and, on this basis, sketch opportunities for industrial
policymaking in the global South. We also offer suggestions for how to expand the eviden
tiary support for these openings and, in a preliminary manner, provide a multipronged re
search agenda.
11.6 Next Steps

One of the otiose stories is about the ‘flying geese’ model, introduced by Kaname Akamat
su (1962) in the 1930s in Japanese and first appearing in an English-language publication
in 1962. This catchphrase depicts followers in formation, pursuing the pattern set by lead
countries, enabling the others to ascend the hierarchy of (p. 300) industrialization. The
metaphor is used to describe the IDL in the Pacific region, where the United States was a
lead country; Japan is said to have played catch-up, and other Asian countries have fallen
in line. But one must take cognizance of how structural constraints operate. The crux of
the matter is that neo-liberal globalization typically reduces national policy space and de
taches economic reforms from social policy.12 In the changing GDLP, policymakers can ill
afford to be swayed by the familiar story of boosting manufacturing by relying on exter
Page 15 of 29
nally driven growth and creating agglomeration economies. This agenda is derived from
the Cold War era, with the clustering of production in East Asia during the first wave of
late industrialization. The contemporary context is far different. Strategic thinking must
come to grips with vast Chinese investment in manufacturing (see Oqubay and Lin, 2019).
Rising costs of skilled labour in China are precipitating continuous restructuring in manu
facturing and shifts in global commodity chains, resulting in the redistribution of industri
al employment and redeployment to new export destinations.
At present, transitions from industrial to digital economies, thus far limited in the global
South, along with more advanced knowledge inputs, appear to be impelling a parametric
transformation in the global economy. While a few middle-income countries are preparing
to benefit from digitalization, many are not. For some of them—among others, Malaysia—
preparation involves studies, even blueprints, with little actual implementation. But the
implications in terms of training and bringing skills and capital from their diasporas are
stark.
Another central question is how low- and middle-income countries can take part in this
global transformation without being relegated to the role of its clients: takers rather than
makers in the GDLP. For these countries, foregoing the potential of recalibrating their po
sitions in the global political economy, with reduced transaction costs, market expansion,
and technological learning and upgrading, represents an opportunity loss. Yet, without
extolling the virtues of digitalization, as do vogue iterations of the hyper-globalist narra
tive, it would be wrong to underestimate the skewed structures of participation.
By now, policymakers should be crafting new strategies for more fully engaging the
knowledge structure to meet the challenges of digital industrialization. Phenomena such
as industrial robots in manufacturing and the concentration of control of digital data un
der the auspices of a handful of big tech firms, the world’s largest ones headquartered in
the United States, are patent. With the exchange of scientific and technological informa
tion, big tech companies based in the global North withhold know-how, especially with
digital industries that are developing algorithms and AI. Obfuscation employed by these
firms to protect their trade secrets takes the (p. 301) form of abstruse language, eu
phemisms, corporate jargon, and esoteric acronyms. In the United States and some other
countries, trade law, with its many loopholes and carve-outs (e.g. for intellectual proper
ty), buttresses the use of secrecy in many kinds of disputes and reinforces barriers to en
try. And in the name of national security, governments seek to deny access to information
about their technological operations so that actors with whom they have adversarial rela
tions cannot acquire it. Faced with this challenge, certain countries are trying to secure a
measure of control, Rwanda among them; they have designed national policies for build
ing digital skills and exercising sovereign rights over the flow of their national data. A
reason why some actors and not others are making strides lies in the particularity of their
domestic political and transnational economic coalitions.
The knowledge community could make an important contribution to achieving this objec
tive. However, the narratives and the data scrutinized above say little about this
Page 16 of 29
community’s role in industrialization. The only exception in our dataset was the 2014 For
taleza Declaration, which affirms the BRICS’ commitment to ‘working together with the
ICT industry, civil society, and academia in order to realize the ICT-related potential op
portunities and benefits for all’ (BRICS, 2014: 10). Although university-based research
and training can be vital resources for industrialization strategies, the linkages between
knowledge production in the global South and industry are generally weak. Rodrigo Aro
cena, a professor of mathematics and former vice-chancellor of the University of the Re
public of Uruguay, and his co-authors drive this point home, calling for the relationship to
be strengthened (Arocena and Sutz, 2001; Arocena, Göransson, and Sutz, 2018). These
ties offer possible gains in advanced knowledge and, beyond copycat technologies, per
haps breakthrough technologies. As emphasized above, national variation is evident. In
some cases, the quality of higher education, research institutes in particular, is such that
cooperation with domestic industries can bear fruit only if it is highly selective and active
ly nurtured by governments’ strategic investments.
Realizing the potential of industry‒university collaborations (IUCs) in the global South

therefore requires clearing hurdles that add to a list of related challenges arising from
IUCs in the global North (Rybnicek and Königsgruber, 2019). Moreover, spending on re
search and development as a proportion of gross domestic product in the global South is
well below the levels of domestic resource mobilization in most developed countries. In
keeping with the neo-liberal trend in the global North, basic research, which is the basis
of applied work and can catalyse innovations, is typically assigned low priority (Arocena
and Sutz, 2001: 1227, 1231; Mittelman, 2018). Practical results and immediate pay-off
have taken precedence. Also, external transfers of technology from sellers or donors to
receivers of imported knowledge often become more a one-way process than a two-way
street.
Surely fundamental and applied research can be rendered complementary rather than
compartmentalized. The knowers and doers can work together or be one and the same. In
this effort, the knowers would strive to better understand the mechanisms of translating
narratives of industrialization into policy, and the doers’ job is to use this (p. 302) under
standing to improve practice. Their mutual goal is to enable local actors to be makers
rather than takers of industrial policies so as to achieve societal well-being.
On this route, attempts to align knowledge production and a high level of training to meet
the challenges of industrial globalization are often hobbled by mistrust of the internation
al experts whose agencies sponsor development projects and the local officials who try to
steer them. This suspicion is not irrational fear. It is based on experience: a combination
of pronounced inefficiencies, endemic corruption, and class antagonism reflecting chang
ing power relations between capital and labour. In this sense, industrial policies can run
aground on ‘technologies of power’ (Foucault, 1984: 178).
In order to arrive at a sober assessment of the prospects for realignment, it is important

to note that knowledge institutions themselves, universities in particular, are beset with a
plethora of problems. Some of them are structural. These include demographic trends,
Page 17 of 29
with a growing population of the young in the global South seeking higher learning; re
ductions in state spending per student with few exceptions, mainly in Asia; and lack of
qualified staff equipped with PhDs. Added to this are matters of political and economic
governance: an increase in executive power and the decline of collegial decision-making
within the academy; mounting costs of maintaining and operating higher education insti
tutions; and pressure from the public and politicians for enhanced performance (Mittel
man, 2018).
From the vantage point of industry, universities are prone to scholasticism. Universities
are seen as being preoccupied with broad philosophical debates, arcane subjects, ab
stract theorizing with little relevance to the down-to-earth needs of the industrial sector,
and their own inbred protocols: the habitus of the academic trade. The application of
Bourdieu’s concept of habitus to this dimension of industrial policy lends perspective to
persistent reservations about closing the gap between industrialists and university-based
scholars, and helps explain why many opportunities for realigning knowledge communi
ties and industry are not pursued.
That said, industry and the academy should be reminded to proceed by respecting one
another’s distinctive missions. And if the goal is to serve mutual interests so as to meet
the needs of the local population in an equitable manner, an antidote to the distance be
tween industry and higher education is to introduce tangible incentives for collaboration
between them. Government authorities could allocate funds to institutions partly on the
basis of what in the Nordic countries is known as the ‘third mission’—knowledge transfer
and societal service. True, increasing incentives comes with a cost and requires money.
Yet mobilizing revenue to invest in this kind of innovation is likely to have a substantial
pay-off.
But to what end? It is not that a gap separates competitors in a race for an imagined fin
ish line. The so-called laggards are not in a race of their own making. The challenge is to
design home-grown industrialization strategies, not to import or download them from
standard global scripts. The goal ought to be increasing autonomy. This calls for captur
ing a larger share of control over globalization and thereby repositioning in the GDLP
(Mittelman and Norani Othman, 2001). Implementation hinges on not only resources but
resourcefulness. It entails not only inventions but inventiveness.
More specifically, the three evolved storylines outlined above point to narrative
(p. 303)
traps as well as opportunities for policymakers in the global South. Putting aside whether
the representations of industrial strategies plotted in prior decades were apt when they
were conceived, the operative consideration now is that customary forms of industrializa
tion in the global South, along with the older narratives that portray them, are historical
ly and theoretically interesting, yet outmoded. The current conjuncture in the world econ
omy is marked by transformations in capitalism. Among them, the decline of manufactur
ing relative to service industries and a shift from industrial capitalism to digital capital
ism are prominent.
Page 18 of 29
11.7 Leveraging Storylines as Policy Space

In our times, the first storyline suggests that national policymakers ought to continue es
chewing protectionist language in the global arena. But this does not mean that selective
industry protection is politically unfeasible. In fact, subtle narrative modifications in re
sponse to regional hegemons’ pivot towards protectionism provide an opportunity for na
tional actors to derive lessons from historical cases of industrial policy. The task is to con
sider locally and regionally workable approaches to industrial midwifery by which domes
tic industries are selectively nurtured and shielded internationally during their initial
stages of growth. Economic powers other than the United States and China have already
adapted their industrial policies accordingly, albeit with a focus on the latter component
of economic midwifery. For instance, India’s unfolding industrial policy enacts Prime Min
ister Modi’s stance on protecting the domestic economy from global pressures. And his
government’s collaboration with South Africa to mobilize state revenues from digital com
merce signifies that India is not alone in this protectionist push. Although policymakers in
countries at the bottom of the GDLP may have less room for manoeuvre, the global tide
for strategic protection is indeed rising.
Another promising move is to reclaim policy space by leveraging the second storyline,
that is, globalization-as-regionalization, both on the intra-regional and inter-regional lev
els. The idea is to replace the logic of catch-up with more flexible scenarios of regional
and inter-regional integration. Conceivably, the fragmentation of the global economic sys
tem may lessen international pressure on industrializing countries to comply with global
norms. The example of India and South Africa jointly arguing their case against the exten
sion of the WTO’s e-commerce moratorium is emblematic of this power shift. In this case,
new tactical alliances among industrializing countries appear more viable than hereto
fore.
But one more word of caution: promoting regionalism carries different pay-offs for the
BRICS than for small, landlocked states. An empirical focus on the BRICS member
(p. 304) states leaves out the latter. Future research could complement our country cases
by integrating data-enabling analyses of how national governments forging industrial pol

icy in the proximity of the BRICS have framed globalization and, on this basis, assess
prospects for regionalism.
Next, an extension of this research could home in on multilateral organizations whose

missions are to promote regional development: among them, the Inter-American, African,
and Asian Development Banks and the UN Department of Economic and Social Affairs’ re
gional commissions. More granular tracing of the evolution of narratives within the com
pass of international institutions and national governments could yield deep insights into
the incentives and disincentives shaping the formation of globalization narratives as they
bear on industrial policy.
Third, inquiry should extend beyond elite discourses so that it encompasses those of the
most vulnerable subjects of narrative power. A bottom-up perspective would round out
Page 19 of 29
what we have attempted in this chapter. Narratives from below are complementary to our
analysis, especially with changes in industrial relations. Studies at the micro level would
be particularly useful in this regard.
Fourth, while undertaking micro studies, there should be no mistaking the macro issue
concerning the parameters in which industrial policy is being forged. In the interregnum
in which we find ourselves, industrial capitalism is rapidly losing ground to a new form of
capitalism, only hinted at but not elaborated above. Digitalization and knowledge work
are gaining sway at the expense of prior generations of technology and wage labour. The
implications for industrial policy in the context of development have yet to be drawn in
any systematic manner. As intimated, in an attempt to adapt during this transitional peri
od, discourse brokers in government and at the WEF are promoting 4IR and have dissem
inated a spate of reports, white papers, and articles on the implications of 4IR for corpo
rations, technology, governance, and academia (Kagermann et al., 2013; Drath and
Horch, 2014; Brettel et al., 2014: 37; Cotteleer and Sniderman, 2017). The extent to
which the 4IR business model and its asymmetries (see Morgan, 2019; Andreoni, Chapter
14, this volume) create a false narrative about a global win‒win scenario while widening
the divide between the world’s data-rich and data-poor populations gives pause.
The Davos 2019 Manifesto clearly calls for a new social narrative in response to the now
widespread recognition of economic globalization’s pernicious effects on inequality. It
would seem that 4IR affords an opportunity to shape this narrative in order to influence
policy. For policymakers in the global South as much as for nongovernmental activists
globally, this is the time to challenge technological solutionism by subjecting the narra
tive to more critical scrutiny. If executed in a constructive, thoroughgoing way, this effort
could help mitigate socio-economic polarization both within and among countries and
world regions.
Finally, the point that warrants emphasis is that for countries at the low end of the GDLP,
achieving fruitful policies requires striving for homemade approaches to industrialization
by strategically seizing shifts in globalization narratives and resulting policy openings.
Contextualized combinations of the steps proposed above—industrial (p. 305) midwifery
enlisting the knowledge community, prioritizing basic as well as applied research, muster
ing resources for incentivizing knowledge transfers and societal service, and orienting
training programmes to foster capacities for resourcefulness and inventiveness—can opti
mize the prospects for industrialization. Each country must forge its own path. It is our
hope that deliberations over country-specific permutations of this portfolio will spur de
velopment through sound industrial policies amid globalizing forces.
Acknowledgements
For valuable research assistance, we are indebted to Ji Young Kwon and Luciana Storelli-
Castro. Thanks, too, to the reviewers, especially Peter Lawrence, of a preliminary version
of this chapter.
Page 20 of 29
References
Africa Report (2018) ‘Interview: Rob Davies, Minister of Trade and Industry, South
Africa’. Available at https://www.theafricareport.com/618/interview-rob-davies-min
ister-of-trade-and-industry-south-africa/.
Akamatsu, Kaname (1962) ‘A Historical Pattern of Economic Growth in Developing Coun

tries’, The Developing Economies 1(1): 3–25.
Altman, Steven A., Pankaj Ghemawat, and Phillip Bastian (2018) ‘DHL Global Connected
ness Index 2018: The State of Globalization in a Fragile World’. Deutsche Post DHL
Group. Available at https://www.logistics.dhl/content/dam/dhl/global/core/docu
ments/pdf/glo-core-gci-2018-full-study.pdf.
Amin, Samir (1990) Delinking: Towards a Polycentric World. London: Zed Books.
Amsden, Alice H. (2001) The Rise of ‘the Rest’: Challenges to the West from Late-Industri
alizing Countries. Oxford: Oxford University Press.
Arocena, Rodrigo and Judith Sutz (2001) ‘Changing Knowledge Production and Latin
American Universities’, Research Policy 30: 1221–34.
Arocena, Rodrigo, Bo Göransson, and Judith Sutz (2018) Developmental Universities in In
clusive Innovation Systems: Alternatives for Knowledge Democratization in the Global
South. Basingstoke: Palgrave Macmillan.
Behuria, Pritish and Tom Goodfellow (2019) ‘Leapfrogging Manufacturing? Rwanda’s At
tempt to Build a Services-led “Developmental State”’, The European Journal of Develop
ment Research 31: 581–603.
Bello, Walden (2005) Deglobalization: Ideas for a New World Economy. London: Zed
Books.
Bertelsmann Stiftung (2018) ‘Globalization Report 2018: What about the BRICS Coun
tries? Why Are They Not the Main Profiteers?’. A Global Economic Dynamics Blogpost.
Available at https://ged-project.de/ged-blog/improving-public-understanding-of-
economic-globalisation/globalization-report-2018-what-about-the-brics-coun
tries/. (p. 306)
Bourdieu, Pierre (1984) Distinction: A Social Critique of the Judgement of Taste. London:
Routledge.
Brettel, Malte, Niklas Friederichsen, Michael Keller, and Marius Rosenberg (2014) ‘How
Virtualization, Decentralization and Network Building Change the Manufacturing Land
Page 21 of 29
scape: An Industry 4.0 Perspective’, International Journal of Information and Communica

tion Engineering 8(1): 37–44.
BRICS (2011) ‘Sanya Declaration’. Sanya: Third BRICS Summit, 14 April.
BRICS (2013) ‘eThekwini Declaration’. eThekwini: Fifth BRICS Summit, 26 March.
BRICS (2014) ‘Fortaleza Declaration’. Fortaleza: Sixth BRICS Summit, 15 July.
BRICS (2016a) ‘Trade Ministers [sic] Communique’. New Delhi: Sixth Meeting of the
BRICS Trade Ministers, 13 October.
BRICS (2016b) ‘Goa Declaration’. Goa: Eighth BRICS Summit, 16 October.
BRICS Business Council (2016) ‘Joint Statement’. Moscow: BRICS Business Council, 6
April.
Burawoy, Michael et al. (2000) Global Ethnography: Forces, Connections, and Imagina
tions in a Postmodern World. Berkeley, CA: University of California Press.
Casa Rosada (2011) ‘Cena por el Día de la industria en Tecnópolis: Palabras de la Presi
denta de la Nación’. Available at https://www.casarosada.gob.ar/informacion/archivo/
25370-cena-por-el-dia-de-la-industria-en-tecnopolis-palabras-de-la-presidenta-
de-la-nacion.
Chatham House (2017) ‘Interview with Rob Davies, South African Trade Minister’. Lon
don: Chatham House. Available at https://www.chathamhouse.org/expert/comment/
interview-rob-davies-south-african-trade-minister.
Cho, Lee-Jay, Yoon Hyung Kim, and Chung H. Lee (eds) (2002) Industrial Globalization in
the Twenty-first Century. Seoul: Korea Development Institute.
Cotteleer, Mark and Brenna Sniderman (2017) ‘Forces of Change: Industry 4.0’, Deloitte
Insights. Available at https://www2.deloitte.com/insights/us/en/focus/industry-4–0/
overview.html?nc=1#what-is-industry-4.
Council on Foreign Relations (2018) ‘A Conversation with Cyril Ramaphosa’. Washington,

DC: Council on Foreign Relations, 24 September. Available at https://www.cfr.org/event/
conversation-cyril-ramaphosa.
Department of Trade and Industry (2013) ‘Joint Statement of the BRICS Business Forum’.
Durban: BRICS Business Forum, 26 March.
Department of Trade and Industry (2017) ‘Industrial Policy Action Plan: Economic Sec
tors, Employment and Infrastructure Development Cluster’. Pretoria: Department of
Trade and Industry.
Page 22 of 29
Dicken, Peter (2015) Global Shift: Mapping the Changing Contours of the World Economy.
7th edition. New York: Guilford Press.
Dollar, David and Aart Kraay (2004) ‘Trade, Growth, and Poverty’, Economic Journal
114(493): 22–49.
Drath, Rainer and Alexander Horch (2014) ‘Industrie 4.0: Hit or Hype?’, IEEE Industrial
Electronics Magazine 8(2): 56–8.
Esser, Daniel E. and Benjamin J. Williams (2014) ‘Tracing Poverty and Inequality in Inter
national Development Discourses: An Algorithmic and Visual Analysis of Agencies’ Annual
Reports and Occasional White Papers, 1978–2010’, Journal of Social Policy 43(1): 173–
200. (p. 307)
Feenstra, Robert C., Robert Inklaar, and Marcel P. Timmer (2015) ‘The Next Generation of
the Penn World Table’, American Economic Review 105(10): 3150–82.
Foucault, Michel (1984) The Foucault Reader, edited by Paul Rabinow. New York, NY: Pan
theon.
Freedom House (2018) Freedom in the World 2018. Washington, DC: Freedom House.
Available at https://freedomhouse.org/report/freedom-world/freedom-world-2018.
Fröbel, Folker, Jürgen Heinrichs, and Otto Kreye (1980) The New International Division of
Labour: Structural Unemployment in Industrialised Countries and Industrialisation in De
veloping Countries. Cambridge: Cambridge University Press.
Government of India (2012) The BRICS Report: A Study of Brazil, Russia, India, China,
and South Africa with Special Focus on Synergies and Complementarities. New Delhi: Ox
Government of India (2018) ‘Prime Minister’s Statement on the Subject “Creating a

Shared Future in a Fractured Word [sic]” in the World Economic Forum’. Delhi: Govern
ment of India, 23 January. Available at https://mea.gov.in/Speeches-Statements.htm?
dtl%2F29378%2FPrime_Ministers_Statement_on_the_subject_Creating_a_Shared_Future_in_
Gramsci, Antonio (1971) Selections from the Prison Notebooks, translated and edited by
Quintin Hoare and Geoffrey Nowell Smith. London: Lawrence and Wishart.
Gygli, Savina, Florian Haelg, Niklas Potrafke, and Jan-Egbert Sturm (2019) ‘The KOF
Globalisation Index–Revisited’, Review of International Organizations (published online).
Available at https://doi.org/10.1007/s11558-019-09344-2.
Held, David, Anthony G. McGrew, David Golblatt, and Jonathan Perraton (1999) Global
Transformations: Politics, Economics, and Culture. Stanford, CA: Stanford University
Press.
Page 23 of 29
Hirschman, Albert O. (1971) A Bias for Hope. New Haven, CT: Yale University Press.
Hirschman, Albert O. ([1981] 2008) Essays in Trespassing: Economics to Politics and Be
yond. Cambridge: Cambridge University Press.
Jomo, Kwame Sundaram and Anis Chowdry (2018) ‘Developing Countries Losing out to
Digital Giants’. Inter Press Service, 17 October. Available at http://www.ipsnews.net/
2018/10/developing-countries-losing-digital-giants/.
Kagermann, Henning, Wolfgang Wahlster, and Johannes Helbig (2013) ‘Recommendations

for Implementing the Strategic Initiative INDUSTRIE 4.0–Securing the Future of German
Manufacturing Industry’. Final Report of the Industrie 4.0 Working Group. Forschung
sunion. Munich: National Academy of Science and Engineering.
Kanwar, Onkar S. (2013) ‘Bridging Gaps with BRICS’, The Financial Express, Noida, Uttar
Pradesh, 24 September.
Kanwar, Onkar S. (2014) ‘BRICS Must Score a Goal in Fortaleza’, Hindustan Times, New
Delhi, 9 July.
KOF Swiss Economic Institute (2018a) KOF Globalisation Index. Eidgenössische Technis
che Hochschule Zürich. Available at https://www.kof.ethz.ch/en/forecasts-and-indica
tors/indicators/kof-globalisation-index.html.
KOF Swiss Economic Institute (2018b) ‘KOF Globalisation Index: Globalisation Lull Con
tinues’. Press release, 12 December. Available at https://www.kof.ethz.ch/en/news-and-
events/media/press-releases/2018/12/kof-globalisation-index-globalisation-lull-
continues.html. (p. 308)
KOF Swiss Economic Institute (2018c) ‘KOF Globalisation Index: Method of Calculation’.
Eidgenössische Technische Hochschule Zürich. Available at https://ethz.ch/content/
dam/ethz/special-interest/dual/kof-dam/documents/Globalization/2018/
Method_2018_2.pdf.
KOF Swiss Economic Institute (2018d) ‘KOF Globalisation Index: Variables Description’.
Eidgenössische Technische Hochschule Zürich. Available at https://ethz.ch/content/
dam/ethz/special-interest/dual/kof-dam/documents/Globalization/2018/
Definitions_2018_2.pdf.
Korea.net (2018) ‘Remarks by President Moon Jae-in at the Korea-Singapore Business Fo
rum’, 18 July. Available at http://www.korea.net/Government/Briefing-Room/Presi
dential-Speeches/view?articleId=161411.
Kwon, Ji Young (2019) ‘Background Paper on Globalization Trends’. Unpublished paper.
Page 24 of 29
Li, Yuefen (2018) ‘Assessment of South-South Cooperation and the Global Narrative on
the Eve of BAPA+40’, Research Paper No. 88. Geneva: South Centre.
Mittelman, James H. (2011) Contesting Global Order: Development, Global Governance,

and Globalization. London: Routledge.
Mittelman, James H. (2018) Implausible Dream: The World-class University and Repur
posing Higher Education. Princeton, NJ: Princeton University Press.
Mittelman, James H. and Norani Othman (eds) (2001) Capturing Globalization. London:
Routledge.
Mkandawire, Thandika (2001) ‘Thinking about Developmental States in Africa’, Cam

bridge Journal of Economics 25(3): 289–314.
Morgan, Jamie (2019) ‘Will We Work in Twenty-first Century Capitalism? A Critique of the
Fourth Industrial Revolution Literature’, Economy and Society. doi:
10.1080/03085147.2019.1620027.
Nayyar, Deepak (2013) Catch Up: Developing Countries in the World Economy. Oxford:
Ohmae, Kenichi (1999) The Borderless World: Power and Strategy in the Interlinked
Economy. New York: HarperCollins.
Oqubay, Arkebe and Justin Yifu Lin (eds) (2019) China‒Africa and an Economic Transfor
mation. Oxford: Oxford University Press.
People’s Republic of China (2014) ‘Connectivity Spearheads Development and Partner

ship Enables Cooperation, Address by Xi Jinping at Dialogue on Strengthening Connectiv
ity Partnership’. Beijing, 8 November. Available at http://www.china-embassy.org/eng/
zgyw/t1210466.htm.
People’s Republic of China (2016) ‘Premier’s Speech at the Opening Ceremony of the
Tenth Annual Meeting of the New Champions’. Tianjin: The State Council of the People’s
Republic of China. Available at http://english.gov.cn/premier/speeches/2016/06/28/
content_281475381438561.htm.
Polanyi, Karl (1957) The Great Transformation. Boston, MA: Beacon Press.
Rasche, Andreas and Daniel E. Esser (2007) ‘Managing for Compliance and Integrity in
Practice’, in Chris Carter, Stewart R. Clegg, Martin Kornberger, Stephan Laske, and Mar
tin Messner (eds) Business Ethics as Practice: Representation, Reflexivity and Perfor
mance. Cheltenham: Edward Elgar, pp. 107–27.
Rodrik, Dani (2018) ‘Populism and the Economics of Globalization’, Journal of Internation
al Business Policy 1(1–2): 12–33.
Page 25 of 29
Rodrik, Dani (2019) ‘Globalization’s Wrong Turn and How It Hurt America’, Foreign Af
fairs 98(3): 26–33. (p. 309)
Roscongress Foundation (2018) ‘Russia to Host the Global Manufacturing and Industriali
sation Summit 2019’. Available at https://roscongress.org/en/news/russia-to-host-
the-global-manufacturing-and-industrialisation-summit-2019/.
Rosenau, James N. and Ernst-Otto Czempiel (eds) (1992) Governance without Govern
ment: Order and Change in World Politics. Cambridge: Cambridge University Press.
Rybnicek, Robert and Roland Königsgruber (2019) ‘What Makes Industry–University Col
laboration Succeed? A Systematic Review of the Literature’, Journal of Business Econom
ics 89(2): 221–50.
Sassen, Saskia (1996) Losing Control? Sovereignty in the Age of Globalization. New York:
Columbia University Press.
Sassen, Saskia (1998) Globalization and its Discontents: Essays on the New Mobility of
People and Money. New York: New Press.
Scholte, Jan Aart (2005) Globalization: A Critical Introduction. 2nd edition. New York: Pal
grave Macmillan.
Schwab, Klaus (2017) The Fourth Industrial Revolution. New York: Crown Business.
Schwab, Klaus (2019) ‘Globalization 4.0: A New Architecture for the Fourth Industrial
Revolution’, Foreign Affairs, 16 January. Available at https://www.foreignaffairs.com/
articles/world/2019-01-16/globalization-40.
Scott, James C. (1990) Domination and the Arts of Resistance: Hidden Transcripts. New
Haven, CT: Yale University Press.
Shotter, John (1993) Cultural Politics of Everyday Life: Social Constructionism, Rhetoric
and Knowing of the Third Kind. Toronto: University of Toronto Press.
Sisulu, Lindiwe N. (2019) ‘Keynote Address at the Mid-term Meeting of the BRICS Busi
ness Council’, Johannesburg: BRICS Business Council, 4 April.
Subramanian, Arvind and Martin Kessler (2013) The Hyperglobalization of Trade and Its
Future. Washington, DC: Peterson Institute for International Economics.
Suehiro, Akira (2008) Catch-up Industrialization: The Trajectory and Prospects of East
Asian Economies, translated by Tom Gill. Honolulu: University of Hawai’i Press.
Sunkel, Osvaldo and Michael Mortimore (2001) ‘Transnational Integration and National
Disintegration’, in Björn Hettne, András Inotai, and Osvaldo Sunkel (eds) Comparing Re
gionalisms: Implications for Global Development. Basingstoke: Palgrave, pp. 54–92.
Page 26 of 29
Tickner, J. Ann (2014) A Feminist Voyage through International Relations. Oxford: Oxford
University Press.
United Nations Conference on Trade and Development (2018) Trade and Development
Report 2018: Power, Platforms and the Free Trade Delusion. New York: UNCTAD.
United Nations Conference on Trade and Development (2019) ‘Growing Trade in Elec
tronic Transmissions: Implications for the South’. UNCTAD Research Paper No. 29. New
York: UNCTAD.
United Nations Industrial Development Organization (2003–6, 2009–11, 2015) Annual Re
ports. Vienna: UNIDO.
Weiß, Johann, Andreas Sachs, and Heidrun Weinelt (2018) 2018 Globalization Report:
Who Benefits Most from Globalization? Gütersloh: Bertelsmann Stiftung. Available at
https://ged-project.de/research/studies/globalization-report-2018-who-benefits-
most-from-globalization/.
Whitfield, Lindsay, Ole Therkildsen, Lars Buur, and Anne Mette Kjær (2015) The Politics
of Africa Industrial Policy: A Comparative Perspective. Cambridge: Cambridge University
Press. (p. 310)
Williams, Raymond (1976) Keywords: A Vocabulary of Culture and Society. London:

Croom Helm.
Wilson Center (2019) ‘We are Committed to Restoring Confidence in the Country and its
Institutions: A Conversation with His Excellency Hamilton Mourão, Vice President of the
Federative Republic of Brazil’, Washington, DC: Woodrow Wilson International Center for
Scholars, 9 April. Available at https://www.wilsoncenter.org/sites/default/files/
event_transcript_-_vice_president_mourao_at_wc_4.9.2019_0.pdf.
World Economic Forum (2019) ‘Globalization 4.0: Shaping a New Global Architecture in
the Age of the Fourth Industrial Revolution’. Geneva: WEF. Available at http://
www3.weforum.org/docs/WEF_Globalization_4.0_Call_for_Engagement.pdf.
World Trade Organization (2019) ‘Work Programme on Electronic Commerce: The E-com
merce Moratorium and Implications for Developing Countries’, Communication from the
Indian and South African delegation. Geneva: World Trade Organization.
Xi, Jinping (2017) ‘Jointly Shoulder Responsibility of our Times, Promote Global Growth’.
Davos: World Economic Forum, 17 January. Available at http://www.china.org.cn/
node_7247529/content_40569136.htm.
Notes:
(1) Yet literature in cognate fields such as cultural theory and media studies gives ample
attention to keywords and speech acts. The acclaimed social critic Raymond Williams
(1976), for instance, argues that the uses and implications of keywords must be explicat
Page 27 of 29
ed historically. He demonstrates that the language in vogue is constituted as lived experi

ence in diverse contexts, open to challenge, and subject to appropriation for gainful pur
poses.
(2) The impact of the preponderant use of the English language on the formation of global
narratives is discussed elsewhere (Mittelman, 2011: especially 229 n. 6 and 232) and falls
outside the purview of this chapter.
(3) As in James C. Scott’s book (1990) on hidden and public transcripts. For an exemplar
of how to discover and decode imaginaries, see Michael Burawoy et al. (2000). On osten
sible representations of industrial globalization, Cho, Kim, and Lee’s collection of essays
(2002) offers ample evidence.
(4) See n. 7 this chapter.
(5) As, for example, in Dani Rodrik’s (2019) use of this construct for examining dilemmas
in the current conjuncture.
(6) Secondary data accuracy could be a concern for our analysis since the KOF Globaliza
tion Index notes that ‘not all data are available for all countries and all years [and] miss
ing values within a series are imputed using linear interpolation’ (KOF Swiss Economic
Institute, 2018c). Yet we are more focused on construct validity than data quality. The
KOF Globalization Index, for instance, measures ‘de facto Cultural Globalization’ through
trade in cultural goods and personal services, international trademark applications, and
McDonald’s and IKEA stores relative to the size of countries’ populations. ‘De jure Cultur
al Globalization’ is proxied by the ratio of girls to boys enrolled in primary education in
public and private schools, Feenstra et al.’s (2015) human capital index (based on the av
erage years of schooling per country and an assumed rate of return to education), and
Freedom House (2018) data. The latter comprise quantifications of aspects of freedom of
expression and belief, associational and organizational rights, rule of law, and personal
autonomy and individual rights (KOF Swiss Economic Institute, 2018d). While creative,
these proxies omit crucial dynamics of cultural globalization. For instance, they do not
capture the international spread of ideas (e.g. political traction gained by de-globaliza
tion), meanings (e.g. rules and norms of global trade), and values (e.g. conditions under
which national governments may choose to violate trade agreements in order to protect
their citizens).
(7) For definitions of the dimensions of globalization and explanations of the methodology
employed, see Savina Gygli et al. (2019). We are also drawing on Kwon (2019).
(8) One year later, South African Prime Minister Cyril Ramaphosa also highlighted the
‘rise of trade protectionism’ as one of the country’s main foreign policy challenges (Coun
cil on Foreign Relations [CFR], 2018).
(9) We have avoided terms such as ‘success’ and ‘failure’ that recur in the literature on in
dustrial policy, for they are normative and reflect sociopolitical hierarchies.
Page 28 of 29
(10) Almost three years after the global financial meltdown, former Argentinean President
Cristina Kirchner pointed to the economic potential inherent in national industrialization
amid contemporary forces of globalization. In a speech to industrialists, she said: ‘Today I
see globalization, which for years looked like a threat to me, as an immense opportunity
for the Argentinean Republic’ (Casa Rosada, 2011). Despite her government’s persistent
nationalistic policies at the time, Kirchner hence felt it opportune to reverse her previous
ly protectionist rhetoric.
(11) Introduced in 2011 as ‘Industrie 4.0’ to enhance competitiveness in manufacturing,

the German federal government incorporated this parlance in its High-Tech Strategy for
2020 and empanelled a working group to advise on specific policies to implement the
plan.
(12) Along these lines, Osvaldo Sunkel and Michael Mortimore (2001) contend that the fly
ing-geese model does not obtain in the manufacturing sector in Latin America, where no
single country has taken the lead and where TNCs and foreign direct investment have not
fulfilled a developmental role. Rather than flying geese, countries in this region have
been, in Sunkel and Mortimore’s words, ‘sitting ducks’.
Daniel E. Esser
Daniel E. Esser is an associate professor in the American University’s School of Inter

national Service in Washington, DC. He holds MSc and PhD degrees from the London
School of Economics and Political Science. His current book project focuses on so
ciopolitical processes forging perceptions of legitimacy where democratic institu
tions are fractured or absent. He also continues to pursue scholarly interests in the
politics of sub-national governance and communities’ responses to violence and ter
ror. His research frequently takes him to Latin America and South Asia, and its find
ings have been published in leading international journals and two Oxford Hand
books.
James H. Mittelman
James H. Mittelman is distinguished research professor and university professor

emeritus at the American University. Previously, he was professor and dean at
Queens College, City University of New York; professor and dean, Graduate School of
International Studies (today the Korbel School), University of Denver; and Pok
Rafeah chair, National University of Malaysia. He received the International Studies
Association’s Distinguished Scholar award in international political economy, and is
an honorary fellow at the Helsinki Collegium for Advanced Studies. His books in
clude The Globalization Syndrome: Transformation and Resistance (Princeton Uni
versity Press, 2000) and Implausible Dream: The World-Class University and Repur
posing Higher Education (Princeton University Press, 2017).
Page 29 of 29
Grand Challenges, Industrial Policy, and Public Value

Mariana Mazzucato and Rainer Kattel
Policymakers are increasingly embracing the idea of using industrial and innovation poli
cy to tackle societal ‘grand challenges’ such as the climate emergency. In this article we
argue that challenge-led policies require a new conceptual and analytical framework that
has at its core the idea of confronting the direction of growth—growth that is, for exam
ple, more inclusive and sustainable. Such framework should focus on public value cre
ation through market co-shaping rather than simply fixing market failures. This is both a
question of theory as well as of policy practice. In theory, challenge-driven industrial and
innovation policy questions both established neoclassical as well as evolutionary con
cepts. In policy practice, directed policies require rethinking what is meant by vertical in
dustrial policies. In this article we argue that through well-defined goals, or more specifi
cally ‘missions’, that are focused on solving important societal challenges, policymakers
have the opportunity to determine the direction of growth by making strategic invest
ments and coordination actions across many different sectors and nurturing new industri
al landscapes.
Keywords: grand challenges, mission-oriented innovation policy, public value, inclusive growth, sustainable
growth
12.1 Introduction
TWENTY-FIRST-century policymaking is increasingly defined by the need to respond to
major social, environmental, and economic challenges. Sometimes referred to as ‘grand
challenges’, these include threats like climate change, demographic, health, and well-be
ing concerns, as well as the difficulties of generating sustainable and inclusive growth.
Against this background, policymakers are increasingly embracing the idea of using in
dustrial and innovation policy to tackle these ‘grand challenges’. Examples of challenge-
led policy frameworks include the United Nation’s Sustainable Development Goals (SDGs;
Borrás, 2019), the European Union’s Horizon Europe research and development pro
Page 1 of 29
gramme (Mazzucato, 2018a), and the UK’s 2017 Industrial Strategy White Paper (HM
Government, 2018).
Challenge-driven policy frameworks are emerging in parallel to well-established modern

ization1 and competitiveness2 frameworks. While modernization, and in particular com
petitiveness frameworks, rely on the idea that government should first and foremost fix
market failures,3 a challenge-driven agenda does not have such clearly defined theoreti
cal origins and analytical lenses. As Richard Nelson argued in 1977 in his seminal book
The Moon and the Ghetto, getting man to the moon and back is not (p. 312) the same as
solving the problem of ghettos in American cities. Put differently, the nature of our knowl
edge about socio-economic challenges differs from our perception of strictly technical
challenges. We can discover answers to technical puzzles; socio-economic issues do not
have a single correct discoverable solution. Such issues require continuous discussion,
experimentation, and learning.
We believe challenge-led growth requires a new conceptual and analytical framework

that has at its core the idea of confronting the direction of growth with growth that is, for
example, more inclusive and sustainable. Such a framework should focus on market shap
ing and market co-creating (Mazzucato, 2016). This is a question of both theory and poli
cy practice. In theory, challenge-driven innovation policy questions both established neo
classical and evolutionary concepts (Schot and Steinmueller, 2018). In policy practice, di
rected policies require rethinking what is meant by ‘vertical policies’.
Industrial policies have always been composed of both a horizontal and a vertical ele
ment. Horizontal policies have historically been focused on skills, infrastructure, and edu
cation, while vertical policies have focused on sectors like transport, health, energy, or
technologies. These two traditional approaches roughly embody differing schools of eco
nomics: neoclassical economics-inspired horizontal policies focusing on supply-side fac
tors and inputs; and evolutionary economics-inspired policies putting emphasis on de
mand-side factors and systemic interactions (Nelson and Winter, 1974; Hausmann and
Rodrik, 2003 for a synthesis). Although certain sectors might be more suited to sector-
specific vertical strategies, the ‘grand challenges’ expressed in SDGs are cross-sectoral
by nature and hence we cannot simply apply a vertical approach to them. Both neoclassi
cal and evolutionary approaches to industrial policy have relied on the idea that the best
policy outcome is economy-wide development, without specifying its nature. In policy this
has led to managing economies according to GDP growth rates, competitiveness indices
and rankings, or other macro indicators (e.g. exports, patents) (Drechsler, 2019). Yet,
many SDGs are only indirectly related to the economy and hence many of the key issues
around SDGs have not been theorized in the context of innovation and industrial policy
(see, e.g. Zehavi and Breznitz, 2017).
In this chapter we argue that through well-defined goals, or more specifically ‘missions’,
that are focused on solving important societal challenges, policymakers have the opportu
nity to determine the direction of growth by making strategic investments, coordinating
actions across many different sectors, and nurturing new industrial landscapes that the
Page 2 of 29
private sector can develop further (Mazzucato, 2017; Mazzucato and Penna, 2016). The
result would be an increase in cross-sectoral learning and macroeconomic stability. This
‘mission-oriented’ approach to industrial policy is not about top-down planning by an
overbearing state; it is about providing a direction for growth, increasing business expec
tations about future growth areas, and catalysing activity—self-discovery by firms (Haus
mann and Rodrik, 2003)—that otherwise would not happen (Mazzucato and Pérez, 2015).
It is not about de-risking and levelling the playing field, nor about supporting more com
petitive sectors over less (Aghion et al., 2015), since the market does not always know
best, but about tilting the playing field in (p. 313) the direction of the desired societal
goals, such as the SDGs. However, we argue, to achieve this requires a new analytical
framework based on the idea of public value and a policymaking framework aimed at
shaping markets in addition to fixing various existing failures. Indeed, we argue that if we
want to take grand challenges such as the SDGs seriously as policy goals, market shaping
should become the overarching approach followed in various policy fields.
12.2 From Market Failure to Market Shaping

The dominant approach to public policy is derived from neoclassical economic theory, in
particular microeconomic theory and welfare economics.4 This approach emphasizes the
idea that, given certain assumptions, individuals pursuing their own self-interest in com
petitive markets gives rise to the most efficient outcomes (Samuelson, 1947; Mas-Colell et
al., 1995: 539‒40). Efficiency is understood in a utilitarian sense, whereby an activity is
efficient if it enhances someone’s welfare without making anyone else worse off (so-called
Pareto efficiency). Under these conditions, the role of government intervention is in prac
tice often limited to addressing instances where the market is unable to deliver Pareto-ef
ficient outcomes.
These ‘market failures’ arise when there are information asymmetries, transaction costs
and frictions to smooth exchange, non-competitive markets (e.g. monopolies) or externali
ties whereby an activity harms another agent not directly connected with the market
transaction (e.g. pollution), or coordination and information failures hamper investment
(Rodrik, 1996).
In the 1960s, public-choice theory considered how the actions of agents (voters, bureau
crats, politicians) involved in policy could be considered from an economic efficiency per
spective, as those agents, including government agents, were assumed to be self-interest
ed (Buchanan and Tullock, 1962). While in markets the existence of competition and the
profit motive tends to enforce efficient decision-making, in collective decision-making
processes (i.e. politics and public administration) the same disciplining framework does
not exist. Policymaking is thus subject to capture by certain interest groups, in particular
those most able to influence policymakers for reasons of power or money. In public ad
ministration, the lack of competitive pressures leads to ‘bureau-maximizing’ behaviour,
whereby departments and agencies look after their own survival rather than the ‘common
good’.
Page 3 of 29
Public-choice theory argues that even where there are clear examples of market failure, it
is not always the case that government intervention would result in a more efficient out
come. Rather, there could also be ‘government failure’, whereby decisions (p. 314) aimed
at improving welfare make things even worse than they would have been under condi
tions of market failure (Le Grand, 1991). For policymaking processes such an approach
creates a bias towards inaction. If the default assumption is that the market will find the
best outcome, even if it doesn’t the overriding concern is that government intervention
may worsen existing outcomes; the default prescription for government policy is to main
tain the status quo. There is a danger that analytical frameworks become focused more
on justifying and measuring the non-failure of public policies than on the attainment of
wider policy goals.
In development theory and practice, the market-failure-based approaches coalesced in

the 1990s around the so-called Washington Consensus policies focused on deregulation,
opening up domestic markets, and relying on foreign direct investments and exports to
drive economic transition and growth (Williamson, 2002). The Washington Consensus
main assumption was that as all development problems are of the same nature, the solu
tions are bound to be the same as well. This removes the question of directionality of
growth away from domestic policymaking and leaves global markets in charge (Kattel et
al., 2016).
The market failure perspective also creates a particular orientation towards innovation,
industrial policy, and structural economic change.5 While certain elements of innovation
policy, in particular early-stage R&D, can be considered to be public goods and thus a
case for public policy provision can be justified, in the main it is assumed that the private
sector is the more efficient innovator, possessing greater entrepreneurial capacity and
better able to take risks given the pressure created by competition. In contrast, the state
is viewed as risk averse and in danger of creating government failure if it becomes too in
volved in industrial policy by ‘picking winners’. Its role is to level the playing field for
commercial actors—mostly through supply-side inputs such as better skills or the removal
of market frictions—and then get out of the way. This has led to rather diverse debates
and the development of policy practices aimed at finding ever more precise policy targets
through better measurement of failures and of the impact of policies trying to fix those
failures. Instead, policy discussions in particular should focus on ‘heterodox’ policy ap
proaches that recognize both market and government imperfections and failures—as well
as the fact that it is impossible or even undesirable to attempt to remove all of them at
once—and the need for policies that support scale economies, dynamic learning effects,
and cross-sectoral spillovers (Rodrik, 2009).
In order to expand such heterodox notions to grand challenges, we need a posi

(p. 315)
tive theory of public value that begins with a notion of the public good not as a correction
to a failure, but as an objective in itself—an objective that can only come about if linked to
a process through which value is created. In this sense a new building block is needed to
guide and legitimize public policy. As indicated by Kenneth Arrow (1962), while a market
failure approach can be utilized to understand why, for example, private firms underin
Page 4 of 29
vest in R&D, it is not so useful for guiding public investment in R&D, because of the in
herent uncertainty involved in the outcomes of such investment. Indeed, Arrow called for
alternative approaches to analysing public investment and policies for innovation.
Critically, the market-failure justification implies that pure private markets/private goods
can exist independently of public or collective action. While the role of institutions is ad
mitted (North, 1991), the role of different voices coming together to form the notion of
the public itself is left mainly to sociology, not economics. Nelson notes that ‘there is no
satisfactory normative theory regarding the appropriate roles of government in a mixed
economy’ (1987: 556) and no theory that captures the complex variety of institutional
arrangements that people have developed to solve collective problems. Just as pure pub
lic goods are rare, so too are pure private goods. Babysitters or sharing everyday appli
ances such as lawnmowers involves no government intervention or regulations, but does
involve collective or ‘public’ negotiation. Hence the ‘market failure’ dichotomy is not par
ticularly useful.
We propose an alternative approach, which begins with the notion of public value as col
lectively generated by a range of stakeholders including the market, the state, and civil
society. Key here is the emphasis on value creation at the core: not ‘public’ value but val
ue itself—with a clear delineation of the role of the different actors that are central to its
formation. While in economics value is, in essence, created inside businesses and only fa
cilitated by the public sector, in this view value is co-created and requires a stakeholder
understanding of capitalism itself. This view draws on the work of Elinor Ostrom (2005),
who shows that a radical state/private division is, to use her word, barren. In developed
economies there are many types of organizations. Non-partisan government regulators,
state-funded universities, and state-run research projects, for example, are quite differ
ent. Besides, the crude binary state/private division fails to capture the many ways in
which all institutions create and destroy value. In addition, Ostrom’s (1990) emphasis on
pooled common resources and her interest in shaping systems so that they take into ac
count collective behaviour, can help shape new policy tools.
This more collective view also benefits from a different understanding of the market it
self, with the market as an outcome of the interactions of individuals, firms, and the state,
as discussed in the work of Karl Polanyi (1957), and ‘embedded economies’, as discussed
by Granovetter (1985). If value is created collectively, a first question becomes: what ca
pabilities, resources, and capacities are needed for this value to be created inside all the
different organizations, including those in the public sector, private sector, and civil soci
ety? In the same way that a theory of private value creation (p. 316) benefits from a re
source-based theory of the firm (Penrose, 1959), so does a public-value notion. Indeed, it
is by sidelining the notion of value as only created in business and facilitated or redistrib
uted by the public sector that the question of capabilities is missed. The work by Teece
(1996) on the dynamic capabilities of the firm becomes equally necessary for the public
sector, as we have argued elsewhere (Kattel and Mazzucato, 2018).
Page 5 of 29
A collective theory of value creation requires understanding by all actors of investment

and production capacity. Indeed, as discussed by Mazzucato and Sekera (2019), a theory
of public value needs to also understand the productive capacity and capabilities of the
state. And if the state loses that capacity it will lose its absorptive capacity—and hence be
unable to understand technological and market opportunities (Cohen and Levinthal,
1990).
Similarly, instruments like taxation are no longer about correcting externalities, but about
creation itself. Adam Smith’s notion of the free market was free from rent and this dis
tinction between rent and profits requires tools to incentivize creation, not extraction, of
value (Mazzucato, 2018a). Thinkers such as Ricardo, Mill, and even Adam Smith recog
nized that unfettered markets were often inefficient and prone to capture by special inter
ests, and could have negative distributional outcomes without ongoing intervention by
the state. In particular, there was a recognition between productive profits and economic
rents that represented unearned income deriving from arbitrary control over resources.
These authors argued that the primary role of taxation, for example, rather than internal
izing externalities caused by identified market failures, should be to tax away rents accru
ing from the monopolistic ownership of factors of production, in particular land (Mazzu
cato, 2018b: 39‒45; Ryan-Collins, 2018: 37‒64). In the classical view, rents did not accrue
from market ‘imperfections’ as in market failure theory, but from the inherent imbalances
in economic and political power that characterized dynamic capitalist economies.
Thus, the focus is on the economic and political processes, institutions, and conditions
that enable (public) value creation—and equally on how to counter (public) value extrac
tion—across sectors and economies (Mazzucato, 2018a). The role of the state is key here,
since it is the only institution with the power to shape markets and direct economic activi
ty in socially desirable directions—or ‘missions’—to achieve publicly accepted outcomes
(Mazzucato, 2013, 2016). Similarly, many government interventions enable markets to
function, such as legal codes, public policies, antitrust policies, university scientists, and
physical infrastructure (Nelson, 1987: 550).
Public value should thus be as much concerned with the direction of growth (and the
macroeconomic implications) as with the microeconomic structure of government agen
cies. The question should be how to shape and co-create markets, not just how to correct
them. Industrial and innovation policy should be focused both on fixing existing market
failures and, equally importantly, on shaping future markets to deliver public value.
(p. 317)
12.3 Learning from History: The Evolution of

Mission-oriented Policies
While economists have had difficulties with the normative theory of the role of govern
ment in the mixed economy, as noted by Nelson above, governments have at various
points in history attempted to implement challenge-driven, mission-oriented policies. This
Page 6 of 29
section gives a brief overview of these policies and draws some key lessons for public val
ue-based market-shaping theory and policies.6
The idea of mission-oriented policies has its root in the idea of modernization, which of
course is not a ‘modern’ idea at all. Even if we are today accustomed to equating modern
ization with Westernization, what we call the modern state and bureaucracy have ar
guably Asian and specifically Chinese origins (Fukuyama, 2011). What matters for our
context is the religious-cultural idea of the ‘mandate of heaven’ (particularly applicable in
late Imperial China) under which rulers must govern well and provide for the people; non-
fulfilment of this ‘mandate’ was a legitimate reason to overthrow the ruler. The counter
part to mandate of heaven in Western culture is the idea of ‘reason of state’, originating
with Giovanni Botero’s eponymously titled book from 1591, Della Ragion di Stato—num
ber five on the bestseller list of economics books published before 1850 (E. Reinert et al.,
2017; E. Reinert, 2007; S. Reinert, 2011)—and justifying policies (Botero explicitly in
cludes economic policies) on the grounds of what today is called ‘national interests’.
These two ideas coalesce around developmental states of the late eighteenth and early
nineteenth centuries with (proto-)missions of catching up, finding their practical toolbox
in Alexander Hamilton’s Report on the Subject of Manufactures in 1791 and their theoret
ical embodiment in Friedrich List’s Das Nationale System der politischen Oekonomie
(1841). Mandate of heaven and national interest offer ideational backdrops to what can
be called a ‘duty to catch up’ as an overarching policy challenge that subsumed under it a
variety of policy missions, from building up knowledge base (e.g. reforming universities)
to creating trade relations and social policy (the latter is particularly crucial for
Bismarck’s Germany, including for the evolution of economics as a science through the
debates around the ‘social question’ of the 1870s; see Drechsler, 2016). The German
catching-up story is especially noteworthy, not only for the country’s significant invest
ments in development resulting in impressive actual catching up and, in many instances,
forging ahead of England and other industry leaders of the time, but also for a wealth of
institutional innovations such as central banks as underwriters of private investment in
industry and multiple welfare-state insurance schemes.
The first generation of mission-oriented policies, the ‘developmental state’, relied on ex
pert meritocracy in public organizations accompanied, however, by constant renewal
(p. 318) and rejuvenation of organizational configurations (Karo and Kattel, 2015). As ex
emplar institutions we can look at what is called System Althoff in German higher educa
tion and research policy (named after Friedrich Althoff, a top civil servant in the Prussian
Ministry of Culture responsible for hugely successful university reforms; vom Brocke,
1991), and at MITI (Ministry of International Trade and Industry) in Japan as quintessen
tial development agencies with ‘embedded autonomy’ (Evans, 1995). While both institu
tions are often seen as representing Weberian bureaucracy at its finest (with merit-based
recruitment and promotion systems and rule-based organization of work), both enjoyed
high-level political support but also relied on what can be called wide-ranging charismat
ic networks, built and nurtured by top civil servants (Karo, 2018).
Page 7 of 29
These policies were wide-ranging, demanding endurance and sacrifice in the name of na
tional catching up and pride, which served as the key sources for normative framing of
public value and as the backdrop for mission/priority selection. A version of these policies
is also to be found in the Soviet Union and other planning-based policies of the post-Se
cond World War era (Freeman, 1987; Chandrashekar, 2016; Chibber, 2003). Indeed, the
latter can be considered an intermediate form between the first and second generations
of mission-oriented policies, particularly in their more successful forms such as Commis
sariat du Plan in France, with its origins in mixed enterprises of the 1920s and its heyday
in the 1960s. Schonfield argues that the success of French planning was also pivotal for
Kennedy’s ‘un-American’ fervour for setting targets for long-term economic growth
(1966: 72‒3). The French planning culture, with its focus on achieving a ‘more complete
view of man’ (quoted in Schonfield, 1966: 227), makes it clear that the point of ‘planning
is thus in part an ethical one: it imposes choices about the use of resources other than
those which the market would produce’ (Schonfield, 1966: 227). The planning exercises
were thus driven by an idea of public value and socio-economic challenges.
The second generation of mission-oriented policies are the well-known policies and public
agencies of the 1940s‒1960s concerning military and space technologies in the United
States and major Western European economies (Ergas, 1987; Soete and Arundel, 1993).
As Alvin Weinberg, for eighteen years the director at Oak Ridge National Laboratory,
which was part of the Manhattan Project, argues, these mission-oriented policies were
about ‘big science’ deployed in the ‘national interests’ (Weinberg, 1967: 132). Thus, on
the one hand, the national interest of the first generation of mission-oriented policies be
came much more clearly defined through defence and military aims; on the other hand, as
we shall see, the translation of these public-value framings into civilian policy areas
failed.
Organizationally, these policies were often implemented by single national/public labora

tories with a concrete mission to solve particular technological problems, not to engage
in basic science, which also meant that their facilities were built up more hierarchically
than universities (Weinberg, 1967). Weinberg argues that partly due to accomplishing
missions, partly due to pressure from researchers, many of these organizations lost focus
on the original missions and moved towards basic research (p. 319) (Weinberg, 1967: 136;
see also Nelson et al., 1967: 169). Similar problems emerged in the Western European
countries where the basic policy assumption was that ‘research results constitute an
undirected potential’ (Krupp, 1985: 51) and that it was up to the private sector to ‘find’
the direction of innovation. The basic research policies were in reality supplemented by
multiple civilian mission-oriented policies in the form of large-scale funding for example,
nuclear energy and transportation (magnetic trains, supersonic aviation) (see Gummett,
1991). This was perhaps the key challenge for the second generation of mission-oriented
policies, and specifically, implementing agencies: how to redeploy former military re
sources around new, civilian missions (Weinberg, 1967: 134‒5; Nelson et al., 1967: 3).
While first-generation mission-oriented policies—catching-up policies—relied on a wide
range of constantly renewed organizations that hired expert civil servants and had strong
political support, the second generation of mission-oriented policies had a much more
Page 8 of 29
heroic vision of dynamic change. Missions were built around single agencies with high-
profile managers in charge of them (Weinberg 1967: 134; also Lambright et al., 1985).
This ambition—in terms of both the problems these organizations took on and the scale of
investment—brought both massive successes and spillovers (Block and Keller, 2011; Maz
zucato, 2013), but also played a crucial role in the demise of this generation of mission-
oriented policies.
As suspected by Weinberg and later documented by Ergas (1987), many mission-driven

research laboratories could not create sensitivity and flexibility around their purpose,
particularly when it came to taking up new emerging, less technological and more social,
challenges such as pollution and the decay of inner cities (Nelson, 1977). The seeming
lack of success in translating the successes of military R&D and its procurement into the
civilian realm also played a significant role in changing policy attitudes (Pavitt and Walk
er, 1976). Similarly, planning exercises, the siblings of mission-oriented policies, often did
not lead to successful outcomes. As documented by Schonfield (1966), economic and in
dustrial policy failures in the United Kingdom and elsewhere in the 1960s, particularly in
contrast to their successes in France, were due both to low political commitment to long-
term planning (not just business-cycle management) and to lack of proper capabilities
within planning organizations. The idea was to pick the willing: ‘deliberately selecting a
few promising firms who seem willing and able to move ahead fast, and then giving them
every encouragement in the form of large contracts, financial help, and other
favours’ (Schonfield, 1966: 111) and with handpicked membership in modernization com
missions (Schonfield, 1966: 98).
Furthermore, one of the key factors in the demise of mission-oriented policies and indus
trial planning in Europe was the emergence of the European Economic Community, in
which each country had rather different planning styles and capabilities (Schonfield,
1966). In the late 1960s and 1970s, instead of a common European style of industrial
planning and mission-oriented policies emerging, rather a gridlock of plans and missions,
and policy cultures remained in place (Schonfield, 1966: 141; also 133). The results of
this could be seen in the fate of the European electronics and semiconductor industries:
they could not compete individually with the US companies (p. 320) but neither could a
European industry emerge as national policy cultures remained dominant (Schonfield,
1966: 374‒5; Dosi, 1981).
The end of this era saw the emergence of (general-purpose) technology foresight exercis
es and a search for visions which, particularly in East Asian economies, was accompanied
by the idea of leapfrogging international competitors rather than just catching up with
them. In essence, mission-oriented policies were slowly replaced by the search for future
technologies and preparing economies for their diffusion (Rothwell and Zegveld, 1981).
At the same time, however, the end of the era denotes the emergence of the market-fail
ure-based approach to (innovation) policy that came to dominate the policy arena, along
with New Public Management reforms, in the late 1980s, resulting in privatization of pub
lic laboratories, the emergence of new arms-length funding agencies (such as research
councils), a focus on commercializing and marketizing research (for example, with com
Page 9 of 29
petitive grant systems), and cost-efficiency practices in policy evaluation (Gummett,

1991; Boden et al., 1998). This contributed to the demise of the directionality of innova
tion as a policy agenda, and to the dominance of market-failure-based value framings in
innovation and industrial policies.
The third generation of mission-oriented policies and organizations, which today is in the
ascendant, has multiple drivers and a somewhat heterogeneous set of actors:
a) Multilateral organizations such as the European Union have been prominent in

urging the development of ‘new’ missions around sustainability and other decidedly
socio-economic (as opposed to solely technological) issues (Soete and Arundel, 1993).
b) Large private philanthropies such as the Gates Foundation and others have sought
out specific problems (e.g. diseases) to solve and have focused not only their funding
but also important networks on those problems as missions.
c) Bottom-up social movements have been able to focus the directionality of re
search, for example, ACT UP’s impact on HIV research and its increased funding
(Leadbeater, 2018). Similarly, Germany’s Energiewende would have never happened
without the green movement (Fagerberg, 2018).
In contrast to previous generations of mission-oriented policies, the current manifestation

does not have a ‘dominant design’ regarding its public-value framing or its governance
system. There is, however, a distinct focus on increasing the social responsiveness of sci
ence and innovation. Rather than focusing on a specific sector (such as energy) or tech
nology (such as nuclear), as was often the case in the previous generation, current at
tempts are characterized by cross-sectoral focus ‘by design’ (Mazzucato, 2018a). Along
side social responsiveness, citizens and social engagement have moved into the arena of
mission-oriented policies. Finally, experimentalism is seen as a key feature of mission-dri
ven policies and organizations, which is reflected in randomized controlled trials being
embraced by philanthropists and social enterprises at the one extreme, and by service de
sign principles of prototyping within various public agencies at the other.
To sum up, we can draw the following lessons from the history of mission-oriented poli
cies. First, there is a taxonomy of missions, from the socio-economic missions of (p. 321)
the first generation to the technological missions of the second, and socio-technological
missions of the current (third) generation. Each type of mission-oriented policies implies
different public-value framings and capabilities to design, implement, and evaluate mis
sions. The directionality of the innovation systems is engendered by different ideational
contexts: first-generation policies were driven by catching up as a national and often also
nationalistic mission; second-generation policies were driven by national security needs
and the technological arms race; and third-generation policies gain their urgency and
purpose from ‘intractable’ socio-economic challenges and social movements connected to
these challenges (e.g. various green movements).
Second, among factors determining the success or failure of previous generations of mis
sion-oriented innovation policies were investments both in R&D capabilities (e.g. re
search laboratories) and in market-shaping capabilities (e.g. procurement practices of
Page 10 of 29
military organizations). This complementarity within and between mission-oriented poli

cies and other economic policies plays an important role in the success of missions.
Third, missions are about setting concrete directions, which of course must be picked,
that is, chosen strategically. The choice is not whether or not to pick but how to do so:
picking directions is not the same thing as ‘picking winners’, in the sense of picking indi
vidual firms or sectors. It is about deciding that a transformation must occur in society—
and making it happen. The direction will require different missions, which in itself pro
vides a focus for the different actors and sectors to collaborate. Thus, missions require
‘picking the willing’: those organizations across the economy (in different sectors, includ
ing both the public and private sphere) that are willing to engage with a mission relevant
to society.
Fourth, with the focus on the market-making, rather than the market-fixing, role of mis
sions, it also becomes clear why public investment by mission-oriented institutions has
been required along the entire innovation chain, and not just upstream basic research.
Better understanding of the distribution of public agencies, their positioning across the
innovation chain, and the balance between directive and bottom-up interactions is a key
area for future study.
12.4 Instrumentalizing Missions as Industrial

Policy
In order to instrumentalize public-value-based policy frameworks as industrial and inno
vation policy tackling grand challenges such as the SDGs, we need, first, to be able to jus
tify the kind of market creation and mission-oriented directionality that was required for
innovations such as the Internet and nanotechnology and is now required to address
today’s societal challenges (Mazzucato, 2016). Second, without the state as a lead in
vestor and market creator, failure-based approaches do not provide insights into the type
and structure of public-sector organizations needed to provide the depth and breadth of
high-risk investments. Third, as long as policy is seen merely as an (p. 322) ‘intervention’,
rather than a key part of the market creation and shaping process, the criteria used to as
sess public investments will inevitably be problematic. Finally, by not describing the state
as a lead risk-taker and investor in this process, failure-based approaches have avoided
the key issue of the distribution of risks and rewards between the state and the private
sector.
Thus, a policy framework for market-shaping activities by the public sector should offer
answers to the following questions, for which we have devised the acronym ROAR:7
(1) How can public policy be understood in terms of setting the direction and route
of change; that is, shaping and creating markets rather than just fixing them (Routes
of directionality)?
Page 11 of 29
(2) How should public organizations be structured so they accommodate the risk-tak
ing, explorative capacity, and capabilities needed to envision and manage contempo
rary challenges (Organizations)?
(3) How can this alternative conceptualization be translated into new dynamic indi
cators and evaluation tools for public policies, beyond the static microeconomic cost–
benefit analysis and macroeconomic appraisal of crowding in/crowding out that stem
directly from the market failure perspective (Assessment)?
(4) How can public investments along the innovation chain result in the socialization
not only of risks, but also of rewards, enabling smart growth to also be inclusive
growth (Risks and rewards)?
While the questions may seem broad, it is their potential connections and internal coher
ence that can help build a market-creation policy framework—and a practical toolkit. Poli
cies that aim to actively create and shape markets require indicators that assess and
measure the performance of a policy along that particular transformational objective. The
state’s ability and willingness to take risks, embodied in transformational changes, re
quires an organizational culture and dynamic capabilities that welcome the possibility of
failure and experimentation and are rewarded for successes so that failures (which are
learning opportunities) can be covered and the next round financed.
12.5 R: Routes and Directionality—A Mission-

oriented Approach
A key success of past market-shaping innovation policies, such as the mission-oriented
policies of the Moonshot era, has been setting a clear direction for problems to be
(p. 323) solved (e.g. going to the moon and back in one generation), which then required
cross-sectoral investments and multiple bottom-up solutions, some of which inevitably

fail. Too much top-down can stifle innovation and too much bottom-up can make it disper
sive with little impact.
A crucial difference between classical mission-oriented policies of the Cold War era and
modern missions is that the latter are focusing on areas such as managing the impact of
technological advance and artificial intelligence on the labour market; adapting to chang
ing demographics and an ageing population; or making the transition to a low-carbon
economy (European Commission, 2011; Kattel and Mazzucato, 2018). Taking up the chal
lenge posed by Richard Nelson in his seminal Moon and the Ghetto (1977), modern-day
mission-oriented policies focus not on technological challenges alone but rather on areas
traditionally the responsibility of public services, such as education or the welfare state,
and entail changes across various economic and policy sectors. Germany’s Energiewende
policy, for instance, aims to combat climate change, phase out nuclear power, improve en
ergy security by substituting renewable sources for imported fossil fuel, and increase en
ergy efficiency. By providing a direction to technical change and growth across different
sectors, Energiewende is tilting the playing field in the direction of a desired socio-eco
nomic goal. Importantly, it is not just about growing ‘green sectors’—it has required many
Page 12 of 29
sectors, including traditional ones such as steel, to transform themselves, and is leading
to changes in patterns of production, services, and consumption of energy. In other
words, its spillovers are as much technological as social and behavioural (see Fagerberg,
2018, for a discussion).
Policies tackling grand challenges should thus be broad enough to engage the public, en
able concrete missions, and attract cross-sectoral investment, and remain focused enough
to involve industry and achieve measurable success. By setting the direction for a solu
tion, missions do not specify how to achieve success, but rather stimulate the develop
ment of a range of different solutions to achieve the objective. In other words, missions
guide entrepreneurial self-discovery (Foray, 2018). As such, a mission can make a signifi
cant and concrete contribution to meeting a Sustainable Development Goal (SDG) or
grand challenge.
The criteria for selecting missions adopted by the European Commission, after wide
spread stakeholder consultation based on the ‘Missions Report’ (Mazzucato, 2018a), are
that they should:
• be bold and address societal value;

• have concrete targets: you know when you get there!
• involve research and innovation: technological readiness over limited time frame;
• be cross-sectoral, cross-actor, and cross-disciplinary;
• involve multiple competing solutions and bottom-up experimentation.
To illustrate, take SDG 14: ‘Conserve and sustainably use the oceans, seas and marine re
sources for sustainable development.’ This could be broken down into various missions,
for example, ‘A plastic-free ocean’ (Figure 12.1). This could stimulate (p. 324) research
and innovation in methods of clearing plastic waste from oceans or in reducing the use of
plastics, innovation in new materials, research on health impacts from micro-plastics, be
havioural research and innovation to improve recycling, or drive public engagement in
cleaning up beaches. Each of these areas can be broken down into particular ‘projects’.
Page 13 of 29
Figure 12.1 A mission-oriented approach to cleaning

the oceans
Source: Mazzucato (2018a: 24).
12.6 O: Organizational Capabilities in the Pub

lic Sector
A key concern should be to establish skills/resources, capabilities, and structures that can
increase the chances that a public organization will be effective, both at learning (p. 325)
and at establishing symbiotic partnerships with the private sector, and will ultimately suc
ceed in implementing mission-oriented and transformative policies. Public and private or
ganizations must re-rethink their roles when working together. Public‒private partner
ships have often limited the public part to de-risking the private part. This ignores the ca
pabilities and challenges involved in public-sector risk-taking. De-risking assumes a con
servative strategy that minimizes the risks of picking losing projects, but does not neces
sarily maximize the probability of picking winners, which requires the adoption of a port
folio approach to public investments (Rodrik, 2014). In such an approach, the success of a
few projects can cover the losses from many projects, and the public organization in ques
tion also learns from its loss-making investments (Mazzucato, 2013). Here, matching fail
ures with fixes is less important than having an institutional structure that ensures that
winning policies provide enough rewards to cover the losses, and that losses are used as
lessons to improve and renew future policies.
One can argue that the community of Schumpeterian scholars never followed up the call
by Nelson and Winter (1982) for public policy to be matched by bold new organizational
structures in the public sector: ‘The design of a good policy is, to a considerable extent,
the design of an organizational structure capable of learning and of adjusting behaviour
in response to what is learned’ (1982: 384). Indeed, there is no equivalent in the litera
Page 14 of 29
ture to ‘dynamic capabilities of the firm’ for the public sector. Developmental state re
search looking at the success of the East Asian Tigers argued that public-sector capaci
ties and capabilities can be best developed by talent, recruited and motivated by Weber
ian meritocratic recruitment and career management that makes working for government
either financially competitive and/or culturally even more rewarding/prestigious than
working in the private sector. Evans and Rauch (1999) cemented these ideas through a
quantitative analysis that tested the importance of some of the ‘Weberian’ elements (mer
it-based recruitment and career systems) in a much broader sample of countries as a
whole (see also Rauch and Evans, 2000; Evans, 1998). This is best captured by Chalmers
Johnson and his concept of the developmental state: a country with a predominant policy
orientation towards development, supported by a small and inexpensive elite bureaucracy
centred around a pilot organization, such as the Ministry of International Trade and In
dustry (MITI) in Japan, and with sufficient autonomy (limited intervention by the legisla
ture and judiciary) (Johnson, 1982: 305‒20).
What is missing, however, in the Weberian framework of capacities are the evolutionary
dynamics: why do specific constellations of capacities become more successful than oth
ers?
Teece and Pisano define dynamic capabilities of the firm by their evolutionary nature.
The term ‘dynamic’ refers to the shifting character of the environment; certain
strategic responses are required when time-to-market and timing is critical, the
pace of innovation accelerating, and the nature of future competition and markets
difficult to determine. The term ‘capabilities’ emphasizes the key role of
(p. 326)
strategic management in appropriately adapting, integrating, and re-configuring

internal and external organizational skills, resources, and functional competencies
towards a changing environment.
(Teece and Pisano, 1994: 1)
We argue that challenge-driven public policies need to be based on a similarly evolution

ary understanding of capabilities in the public sector.
We propose that twenty-first-century missions require the following set of dynamic capa
bilities in the public sector in order to engender mission-oriented policies (Kattel and
Mazzucato, 2018).
First, key to our premise is that grand challenges can only be solved through dynamic
public‒private partnerships, but these have been constrained by the notion of public ac
tors as, at best, fixing markets. A market-co-creating role requires the state to have capa
bilities for leadership and engagement: missions can all too quickly become either just
fashionable labels on ‘business-as-usual’ practices or too rigid top-down planning exercis
es. Thus, capabilities to engage with a wide set of social actors and to show leadership
through bold vision are vital at a time of high ‘democratic deficit’ in many developed
countries (see also ESIR, 2017). Some of the grand challenges contest ‘the way of life’ as
Page 15 of 29
we know it (e.g. suburbanization accompanied by congested transportation systems). Ca

pabilities to encourage bottom-up engagement mean a capacity to set missions but also to
leave enough space for contestation and adaptability.
Second, on the level of policy, finding coherent policy mixes (instruments and funding)
and coordination capabilities is fundamental to the success of today’s mission-oriented
policies. As these missions are not just about technological solutions but also have strong
socio-political aspects, experimentation capabilities matter perhaps more than before.
Equally important are evaluation capabilities that do not simply rely on market-failure-
based approaches (e.g. cost‒benefit analysis) but can also integrate user research, social
experiments, and system-level reflection (Lindner et al., 2016; Rip, 2006).
Third, administrative capabilities need to rely on a diversity of expertise and skills from
engineering to human-centric design: organizational forms that mix unrelated knowledge
areas (e.g. in urban mobility and planning, lifestyles are just as important as new energy
storage systems; see Grillitsch et al., 2017) and organizational fluidity (e.g. cross-depart
mental teams) seem to be fundamental for managing new missions (OECD, 2017).
12.7 A: Assessment and Evaluation

One of the key challenges of applying a public value-based framework to policymaking is
the relationship between policy assessment, appraisal, and evaluation. Influenced by the
market failure framework, today’s dominant evaluation practices are usually based
on allocative efficiency and some form of ex ante cost‒benefit analysis (CBA).8
(p. 327)
Costs (including the costs of potential government failure) are usually defined by their op
portunity cost, that is, the value that reflects the best alternative use a good or service
could be put to (include a do-nothing/business-as-usual option), with all else (including all
other prices) assumed equal, and with market prices usually the starting point for the
analysis (see, e.g. HM Treasury, 2018: 6). Post-intervention policy evaluation then seeks
to verify whether the estimates were correct and whether the market failure was ad
dressed.
To enable market-type price comparison of interventions whose return will vary in terms
of time, CBAs typically make use of a ‘discount rate’ that reflects the time preference of
users of the service for having money now rather than in the future. After adjusting for in
flation and discounting, costs and benefits can be added together to calculate the net
present value (NPV) of different policy options. In recognition of the problem of externali
ties, some attempt has been made in recent years to incorporate the wider costs to soci
ety of particular policy actions, for example through monetizing certain social or ecologi
cal externalities in a ‘social cost‒benefit analysis’ (SCBA) or ‘social cost-effectiveness
analysis’ (SCEA). However, the overall framework remains rooted in the idea that creat
ing a ‘market price’ for interventions will enable the most accurate decision to maximize
welfare and public value. CBA and NPV are mostly aimed at preventing costly govern
Page 16 of 29
ment failures; by their very nature, they cannot tell us very much at all about proactive
market creating and shaping.
This limitation is of crucial importance. Market-shaping policies, such as missions, aim to

accelerate innovation, creating new technologies and radically changing the prices, avail
ability, and existence of goods and services. Their central purpose is to transform underly
ing relationships, a wide range of prices and the broader environment (OECD, 2015). The
‘all else (including prices) being equal’ assumption underlying cost‒benefit analysis be
comes problematic in such circumstances.
By always comparing the policy intervention with the status quo and emphasizing short-
term risks, CBA approaches encourage decision makers to prefer small-scale, marginal
interventions (Allas, 2014: 89). Yet there is considerable evidence that innovation systems
exhibit increasing returns or an S-curve-type effect, in which shifting incentives across
multiple sectors may be more likely to achieve such increasing returns (Mazzucato,
2017). So, arguably, if there is to be any bias around innovation policy it should be in
favour of large-scale interventions. Furthermore, the strong emphasis on risk assessment/
optimism bias is likely to mitigate against the creation of a mission-oriented approach
where failure is viewed as a learning process integral to the achievement of important
technological breakthroughs (Mazzucato, 2013).
CBA-type analyses derived from market failure theory are concerned with allocative or
distributive efficiency, which involves making the best use of (fixed) resources at a fixed
point in time. Dynamic efficiency involves making the best use of resources to (p. 328)
achieve changes over time and so is concerned with innovation, investment, improve
ment, and growth—including, perhaps most importantly, the creation of new resources
(technologies) and shifting technology frontiers (Kattel et al., 2018). Missions are, by defi
nition, concerned with dynamic efficiency, since they aim to accelerate innovation and
transformational change.
When allocative efficiency frameworks are applied to dynamic efficiency problems, the
analysis risks are either irrelevant or actively unhelpful.
Aside from considerations of efficiency, given the importance of dynamics over time for
market-shaping policies, it is important to define a concrete target and objectives. In oth
er words, it must be possible to say definitively whether the policy has been achieved or
not. Technological missions such as putting a man on the moon had obvious end points
which made evaluation easier. However, modern grand challenges are more long term
and their end points less easy to define.
12.8 R: Risks and Rewards

But this raises a more fundamental question: how to make sure that, like private venture
capital funds, the state can reap some return from the successes (the upside), in order to
cover the inevitable losses (the downside) and finance the next round of investments.9
Page 17 of 29
This is especially important given the path-dependent and cumulative nature of innova
tion. Returns arise slowly; they are negative in the beginning and gradually build up, po
tentially generating huge rewards after decades of investment. Indeed, companies in ar
eas like ICT, biotechnology, and nanotechnology had to accept many years of zero profits
before any returns were in sight. If the collective process of innovation is not properly
recognized, the result will be a narrow group of private corporations and investors reap
ing the full returns of projects which the state helped to initiate and finance.
So who gets the reward for innovation? Some economists argue that returns accrue to the
public sector through knowledge spillovers (new knowledge that can benefit various ar
eas of the economy), and via the taxation system due to new jobs being generated and
taxes being paid by companies benefiting from the investments. But the evolution of the
patenting system has made it easier to take out patents on upstream research, meaning
that knowledge dissemination can effectively be blocked and spillovers cannot be as
sumed. The cumulative nature of innovation and the dynamic returns to scale (Nelson and
Winter, 1982) mean that countries stand to gain significantly from being first in the devel
opment of new technologies.
At the same time the global movement of capital means that the particular country or re
gion funding initial investments in innovation is by no means guaranteed to reap (p. 329)
all the wider economic benefits, such as those relating to employment or taxation. In
deed, corporate taxation has been falling globally, and corporate tax avoidance and eva
sion rising. Some of the technology companies that have benefited the most from public
support, such as Apple and Google, have also been among those accused of using their in
ternational operations to avoid paying tax (Johnston, 2014). Perhaps most importantly,
while the spillovers that occur from upstream ‘basic’ investment, such as education and
research, should not be thought of as needing to earn a direct return for the state, down
stream investments targeted at specific companies and technologies are qualitatively dif
ferent. Precisely because some investments in firms and technologies will fail, the state
should treat these investments as a portfolio, and enable some of the upside success to
cover the downside risk.
In particular, there is a strong argument to be made that, where technological break

throughs have occurred as a result of targeted state interventions benefiting specific
companies, the state should reap some of the financial rewards over time by retaining
ownership of a small proportion of the intellectual property it had a hand in creating. This
is not to say that the state should ever have exclusive licence or hold a large enough pro
portion of the value of an innovation to deter its diffusion (and this is almost never the
case). The role of government is not to run commercial enterprises; it is to spark innova
tion elsewhere. But by owning some of the value it has created, which over time has the
potential for significant growth, government can generate funds for reinvestment into
new potential innovations. With the adoption of a portfolio approach to public investment
in innovation, success from some projects can help cover the losses from others. In this
way, both risks and rewards are socialized (Mazzucato, 2016).
Page 18 of 29
There are many examples of public organizations that have strategically considered the
distribution of risks and rewards. At times, they have granted licences to private firms
willing to invest in upgrading publicly owned technologies, offering the opportunity for
public and private sectors to share both risks and rewards. For example, NASA has some
times captured the returns to its inventions, while private partners gained on the value
added in case of successful commercialization (Kempf, 1995). There are other examples of
state-owned venture capital activity generating royalties from public investments (in Is
rael, see Avnimelech, 2009) or equity (in Finland via Sitra), and the more pervasive use of
equity by state development banks (e.g. in Brazil, China, and Germany; see Mazzucato
and Penna, 2016).
Policy instruments for tackling risk/reward issues combine supply- and demand-side
mechanisms geared to enabling public value creation through symbiotic public‒private
partnerships (‘active’) (Lazonick and Mazzucato, 2013) and mechanisms blocking value
extraction (‘defensive’).
Rewards can be distributed either directly through profit sharing (via equity, royalties) or
indirectly through conditions attached that focus more on the market-shaping role. The
latter may involve conditions on the reinvestment of profits, conditions on pricing, or con
ditions on the way that knowledge is governed.
This list is not meant to be exhaustive, but rather, to illustrate that there are multiple ex
periences in handling policy instruments that, implicit or explicitly, (p. 330) enable consid
eration of issues like value extraction and allowing government to capture a share of the
value it helped to generate. The latter, in particular, have been adopted by different types
of agencies, at different stages of the innovation chain but mainly downstream, involving
different types of partners (e.g. firm size) and industries. However, they have not always
been adjusted to the specificities of different economic, industrial, and legal settings. Ab
sent a framework that more clearly informs these policies, decisions on these matters
have sometimes been made unintentionally and haphazardly, inviting both government
and systemic failures.
The prospect of the state owning a stake in a private corporation may be anathema to
many parts of the capitalist world, but given that governments are already investing in
the private sector, they may as well earn a return on those investments (something even
fiscal conservatives might find attractive). The state need not hold a controlling stake, but
it could hold equity in the form of preferred stocks that get priority in receiving divi
dends. The returns could be used to fund future innovation (Rodrik, 2015). Politicians and
the media have been too quick to criticize public investments when things go wrong, and
too slow to reward them when things go right.
Thus, more thought should be given, not so much to the problem of ‘picking winners’, as
to how to reward the winning investments so they can both cover some of the eventual
losses (inevitable in the innovation game) and raise funds for future investments. Going
hand in hand with this consideration is the need to rethink how public investments are
accounted for in national income accounting. Investments in innovation are different to
Page 19 of 29
current expenditure. The latter does not add to balance-sheet assets; the former does,
and is potentially productive investment in the sense that it creates new value (Mazzuca
to and Shipman, 2014). When setting limits to fiscal deficits, it is therefore necessary to
distinguish public debt contracted for investment in R&D and infrastructure (value-creat
ing investments) from public debt contracted for (public or private) consumption. In this
sense, financial and accounting reforms should be regarded as a prerequisite for any suc
cessful smart and inclusive growth plan.
Finally, regarding the government as lead risk-taker helps to debunk fundamental as
sumptions behind the theory of shareholder value that underpins the exorbitant rewards
earned by senior executives in recent years. Share options as part of pay packages have
been a key feature of modern capitalism, and especially, a key driver of the inequality be
tween the top 1 per cent of income earners and the rest (Piketty, 2014). Share options are
boosted when share prices rise, and prices often rise through ‘financialized’ practices
such as share repurchase schemes by companies (Lazonick, 2014). Focusing on boosting
share prices is justified on the grounds of the theory of shareholder value, which holds
that shareholders are the biggest risk takers in a company because they have no guaran
teed rate of return (while workers earn set salaries, banks earn set interest rates, etc.).
That is, they are the residual claimants (Jensen, 1986).
(p. 331) But this assumes that other agents do have a guaranteed rate of return. As we
have argued throughout this chapter, it is precisely because what the state does is not
just facilitate and de-risk the private sector, but also take major risks, that there is no
guarantee of success in its investments, which have historically also played a crucial role
in enabling wealth creation. The fact that a key driver of inequality has been linked with a
problematic understanding of which actors are the greatest risk takers implies that com
batting short- termism and speculative forms of corporate governance requires not only
reforming finance and corporate governance, but also rethinking the models of wealth
creation upon which they are based (Mazzucato, 2018b).
12.9 Conclusion
Market-shaping, mission-oriented approaches to policy allow us to reconsider how to jus
tify ambitious policies that aim to transform landscapes rather than fix problems in exist
ing ones. This approach to policy raises challenges in terms of how to nurture organiza
tional structures that can manage such policies, and how to appraise and evaluate the
market-shaping effect of the policies. This approach would help capture the potential for
policy to create spillover effects across many sectors of the economy and alter the level of
investment and the broader trajectory of economic growth.
What are some of the possible concerns with this type of approach? One concern is
around the setting of missions and the direction of the market shaping in the first place.
Clearly governments can and do become captured by particular interest groups which
limit their ability to both establish missions and follow through on them. The challenges
of climate change and inequality are obvious examples. Government subsidies continue to
Page 20 of 29
favour vested interests (for example fossil-fuel energy firms), while taxation policy
favours labour saving (increasing unemployment or underemployment) over resource sav
ing (supporting decarbonization), despite governments signing up to treaties committing
themselves to different policy directions. And, of course, democracy is no guarantee that
societal missions such as climate change will be adopted globally, as the current adminis
trations in the United States and Brazil clearly demonstrate.
However, arguably, these are the outcomes of governments not doing enough to shape
markets to support social and environmental policy goals in the first place. Hopefully the
ideas in this chapter can help meet that challenge.
Acknowledgements
This chapter is based on our previous work: Mazzucato (2018a and 2018b), Kattel and
Mazzucato (2018), Mazzucato and Ryan-Collins (2019), and Mazzucato, Kattel, and Ryan-
Collins (2019). The research for the article has been partially funded by the Horizon2020
project GROWINPRO, http://www.growinpro.eu.
References
Aghion, Philippe, Jing Cai, Mathias Dewatripont, Luosha Du, Ann Harrison, and Patrick
Legros (2015) ‘Industrial Policy and Competition’, American Economic Journal Macroeco
nomics, 7(4): 1–32.
Allas, Tera (2014) ‘Why a Fundamental Shift in Innovation Policy is Necessary’, in Mari
ana Mazzucato and Caetano C. R. Penna (eds) Mission-oriented Finance for Innovation.
London: Policy Network, pp. 85–94.
Arrow, Kenneth J. (1962) ‘The Economic Implications of Learning by Doing’, The Review
of Economic Studies 29(3): 155–73.
Avnimelech, Gil (2009) ‘VC Policy: Yozma Program 15-Years Perspective’. SSRN Scholarly
Paper No. ID 2758195. Rochester, NY: Social Science Research Network. Available at
https://papers.ssrn.com/abstract=2758195.
Block, Fred and Matthew R. Keller (eds) (2011) State of Innovation: The U.S.
Government’s Role in Technology Development. London: Paradigm.
Boden, Rebecca, Philip Gummett, Debora Cox, and Kate Barker (1998) ‘Men in White
Coats…Men in Grey Suits’, Accounting, Auditing & Accountability Journal 11(3): 267–91.
Borrás, Susana (2019) ‘Domestic Capacity to Deliver Innovative Solutions for Grand So
cial Challenges’, in Stone, D. and Moloney, K. (eds) Oxford Handbook on Global Policy and
Transnational Administration. Oxford: Oxford University Press, pp. 182–99.
Buchanan, James M. and Gordon Tullock (1962) The Calculus of Consent, Vol. 3. Ann Ar
bor, MI: University of Michigan Press.
Page 21 of 29
Chandrashekar, C. P. (2016) ‘Development Planning’, in Erik S. Reinert, Jayati Ghosh, and

Rainer Kattel (eds) Handbook of Alternative Theories of Economic Development.
Cheltenham: Edward Elgar, pp. 519–32.
Chibber, Vivek (2003) Locked in Place: State Building and Late Industrialization in India.
Coase, Ronald H. (1960) ‘The Problem of Social Cost’, Journal of Law and Economics 3(1):
1–44.
Cohen, Wesley M. and Daniel A. Levinthal (1990) ‘Absorptive Capacity: A New Perspec
tive on Learning and Innovation’, Administrative Science Quarterly 35(1): 128–52.
Dosi, Giovanni (1981) ‘Technical Change and Survival: Europe’s Semiconductor Industry’.
Sussex European Papers No. 9, Sussex European Research Centre, University of Sussex.
Drechsler, Wolfgang (2016) ‘The German Historical School and Kathedersozialismus’, in

Erik S. Reinert, Jayati Ghosh, and Rainer Kattel (eds) Handbook of Alternative Theories of
Economic Development. Cheltenham: Edward Elgar, pp. 109–23.
Drechsler, Wolfgang (2019) ‘Kings and Indicators: Options for Governing without Num
bers’, in Markus J. Prutsch (ed.) Science, Numbers and Politics. Cham: Palgrave Macmil
lan.
Ergas, Henry (1987) ‘Does Technology Policy Matter?’, in Bruce Guile (ed.) Technology
and Global Industry: Companies and Nations in the World Economy. Washington, DC: Na
tional Academies Press, pp. 191–245.
ESIR (Expert Group on the Economic and Societal Impact of Research and Innovation)
(2017) ‘Towards a Mission-oriented Research and Innovation Policy in the European
Union’, an ESIR Memorandum.
European Commission (2011) ‘Green Paper–From Challenges to Opportunities: Towards a

Common Strategic Framework for EU Research and Innovation Funding’. Brussels: Euro
pean Commission. (p. 333)
Evans, Peter (1995) Embedded Autonomy: States and Industrial Transformation.

Evans, Peter (1998) ‘Transferable Lessons? Re‐examining the Institutional Prerequisites

of East Asian Economic Policies’, The Journal of Development Studies 34(6): 66–86.
Evans, Peter and James E. Rauch (1999) ‘Bureaucracy and Growth: A Cross-national
Analysis of the Effects of “Weberian” State Structures on Economic Growth’, American
Sociological Review 64(5): 748–65.
Fagerberg, Jan (2018) ‘Mission (Im)Possible? The Role of Innovation (and Innovation Poli
cy) in Supporting Structural Change and Sustainability Transitions’. Working Paper on In
Page 22 of 29
novation Studies No. 20180216, TIK Centre for Technology, Innovation and Culture, Uni
versity of Oslo.
Foray, Dominique (2018) ‘Smart Specialisation Strategies as a Case of Mission-oriented

Policy: A Case Study on the Emergence of New Policy Practices’, Industrial and Corporate
Change 27(5): 817–32.
Freeman, Chris (1987) Technology Policy and Economic Performance: Lessons from
Japan. London: Pinter.
Fukuyama, Francis (2011) The Origins of Political Order: From Prehuman Times to the
French Revolution. New York: Farrar, Straus, and Giroux.
Granovetter, Mark (1985) ‘Economic Action and Social Structure: The Problem of Embed
dedness’, American Journal of Sociology 91(3): 481–510.
Greenwald, Bruce C. and Joseph E. Stiglitz (1986) ‘Externalities in Economies with Imper
fect Information and Incomplete Markets’, The Quarterly Journal of Economics 101(2):
229–64.
Grillitsch, Markus, Björn Asheim, and Michaela Trippl (2017) ‘Unrelated Knowledge Com
binations: Unexplored Potential for Regional Industrial Path Development’. Papers in In
novation Studies, Lund University, CIRCLE–Center for Innovation, Research and Compe
tences in the Learning Economy, 2017/10.
Gummett, Philip (1991) ‘The Evolution of Science and Technology Policy: A UK Perspec
tive’, Science and Public Policy 18(1): 31–7.
Hausman, Ricardo and Dani Rodrik (2003) ‘Economic Development as Self-discovery’,

Journal of Development Economics 72(2): 603–33.
HM Government (2018) ‘Industrial Strategy: Building a Britain Fit for the Future’, De
partment of Business, Energy and Industrial Strategy. Available at https://www.gov.uk/
government/publications/industrial-strategy-building-a-britain-fit-for-the-future.
HM Treasury (2018) ‘The Green Book: Central Government Guidance on Appraisal and
Evaluation’. Available at https://www.gov.uk/government/publications/the-green-
book-appraisal-and-evaluation-in-central-governent ISBN 978-1-912225-57-6
PU2145.
Jensen, Michael (1986) ‘Agency Costs of Free Cash Flow, Corporate Finance, and
Takeovers’, American Economic Review 76(2): 323–9.
Johnson, Chalmers (1982) MITI and the Japanese Miracle: The Growth of Industrial Policy.
Stanford, CA: Stanford University Press.
Page 23 of 29
Johnston, David Cay (2014) ‘How Google and Apple Make their Taxes Disappear’,
Newsweek, 14 December. Available at http://www.newsweek.com/2014/12/26/how-
google-and-apple-make-their-taxes-disappear-291571.html.
Karo, Erkki (2018) ‘Innovation Policy and Mission-oriented Public Organizations in East
Asia’, Industrial and Corporate Change 27(5): 867–81.
Karo, Erkki and Rainer Kattel (2015) ‘Innovation Bureaucracy: Does the Organization of
Government Matter when Promoting Innovation?’. Papers in Innovation Studies, Lund
(p. 334) University, CIRCLE–Center for Innovation, Research and Competences in the
Learning Economy, 2015/38.
Kattel, Rainer and Mariana Mazzucato (2018) ‘Introduction: Mission-oriented Innovation

Policy and Dynamic Capabilities in the Public Sector’, Special Issue of Industrial and Cor
porate Change, 28(5): 787–801.
Kattel, Rainer, Jan Kregel, and Erik S. Reinert (2016) ‘Classical Development Economists
of the Mid-20th Century’, in Erik S. Reinert, Jayati Ghosh and, Rainer Kattel (eds) The El
gar Handbook of Alternative Theories of Economic Development. Cheltenham: Edward El
gar, pp. 336–51.
Kattel, Rainer, Mariana Mazzucato, Josh Ryan-Collins, and Simon Sharpe (2018) ‘The Eco
nomics of Change: Policy Appraisal for Missions, Market Shaping and Public Purpose’.
UCL Institute for Innovation and Public Purpose Policy Report.
Kempf, Robert (1995) ‘Exploitation of Rights from the US Space Program by NASA’, ESA
SP–378, p. 27. Intellectual Property Rights and Space Activities. Paris: European Space
Agency.
Krupp, Hans (1985) ‘Public Promotion of Innovation—Disappointments and Hopes’, in

Gerry Sweeney (ed.) Innovation Policies: An International Perspective. London: Pinter, pp.
48–79.
Lambright, W. Henry, Michael Crow, and Ralph Shangraw (1985) ‘National Projects in
Civilian Technology’, in J. David Roessner (ed.) Government Innovation Policy: Design, Im
plementation, Evaluation. Basingstoke: Macmillan, pp. 63–74.
Lazonick, William (2014) ‘Profits without Prosperity: Stock Buybacks Manipulate the Mar
ket and Leave Most Americans Worse Off’, Harvard Business Review, September: 46–55.
Lazonick, William and Mariana Mazzucato (2013) ‘The Risk–Reward Nexus in the Innova
tion–Inequality Relationship: Who Takes the Risks? Who Gets the Rewards?’, Industrial
and Corporate Change 22(4): 1093–128.
Leadbeater, Charles (2018) ‘Postscript: Movements with Missions Make Markets’, Indus
trial and Corporate Change 27(5): 937–42.
Page 24 of 29
Le Grand, Julien (1991) ‘The Theory of Government Failure’, British Journal of Political
Science 21(4): 423–42.
Lindner, Ralf, Stephanie Daimer, Bernd Beckert, Nils Heyen, Jonathan Koehler, Benjamin
Teufel, Philine Warnke, and Sven Wydra (2016) ‘Addressing Directionality: Orientation
Failure and the Systems of Innovation Heuristic: Towards Reflexive Governance’. Fraun
hofer ISI Discussion Papers Innovation Systems and Policy Analysis No. 52.
Mas-Colell, Andreu, Michael Dennis Whinston, and Jerry R. Green (1995) Microeconomic
Theory, Vol. 1. Oxford: Oxford University Press.
Mazzucato Mariana (2016) ‘From Market Fixing to Market-creating: A New Framework

for Innovation Policy’, Industry and Innovation 23(2): 140–56.
Mazzucato, Mariana (2017) ‘Mission-oriented Innovation Policy: Challenges and Opportu

nities’. UCL Institute for Innovation and Public Purpose Working Paper, IPP WP 2017-01.
Available at https://www.ucl.ac.uk/bartlett/public-purpose/publications/2018/jan/
mission-oriented-innovation-policy-challenges-and-opportunities.
Mazzucato, Mariana (2018a) ‘Missions: Mission-oriented Research and Innovation in the

European Union’. European Commission. Available at https://ec.europa.eu/info/sites/
info/files/mazzucato_report_2018.pdf.
Mazzucato, Mariana (2018b) The Value of Everything. London: Penguin. (p. 335)
Mazzucato, Mariana and Penna, C. C. (2016) ‘Beyond Market Failures: The Market Creat
ing and Shaping Roles of State Investment Banks’, Journal of Economic Policy Reform
19(4): 305–26.
Mazzucato, Mariana and Carlota Pérez (2015) ‘Innovation as Growth Policy’, in Jan Fager
berg, Staffan Laestadius, and Ben R. Martin (eds) The Triple Challenge: Europe in a New
Age. Oxford: Oxford University Press, pp. 229–64.
Mazzucato, Mariana and Josh Ryan-Collins (2019) ‘Putting Value Creation back into “Pub
lic Value”: From Market Fixing to Market Shaping’. UCL Institute for Innovation and Pub
lic Purpose, Working Paper Series No. 2019-05. Available at https://www.ucl.ac.uk/
bartlett/public-purpose/wp2019-05.
Mazzucato, Mariana and June Sekera (2019) ‘Rethinking Public Production: Producing
Public Value and Achieving Public Purpose’. IIPP Working Paper, forthcoming.
Mazzucato, Mariana and Shipman, A. (2014) ‘Accounting for Productive Investment and
Value Creation’, Industrial and Corporate Change 23(1): 1–27.
Page 25 of 29
Mazzucato, Mariana, Rainer Kattel, and Josh Ryan-Collins (2019) ‘Challenge-driven Inno
vation Policy: Towards a New Policy Toolkit’, Journal of Industry, Competition and Trade,
forthcoming.
Nelson, Richard R. (1977) The Moon and the Ghetto: An Essay on Public Policy Analysis.
New York: W. W. Norton.
Nelson, Richard R. (1987) ‘Roles of Government in a Mixed Economy’, Journal of Policy

Analysis and Management 6(4): 541–66.
Nelson, Richard R. and Sidney G. Winter (1974) ‘Neoclassical vs. Evolutionary Theories of
Economic Growth: Critique and Prospectus’, The Economic Journal 84(336): 886–905.
Nelson, Richard R. and Sidney G. Winter (1982) An Evolutionary Theory of Economic

Change. Cambridge, MA: Belknap Press of Harvard University.
Nelson, Richard R., Merton J. Peck, and Edward D. Kalachek (1967) Technology, Econom
ic Growth and Public Policy. Washington, DC: The Brookings Institution.
North, Douglass C. (1991) ‘Institutions’, Journal of Economic Perspectives 5(1): 97–112.
OECD (2015) ‘System Innovation: Synthesis Report’. Paris: OECD. Available at https://
www.innovationpolicyplatform.org/sites/default/files/general/
SYSTEMINNOVATION_FINALREPORT.pdf.
OECD (2017) Systems Approaches to Public Sector Challenges: Working with Change.
Paris: OECD.
Ostrom, Elinor (1990) Governing the Commons: The Evolution of Institutions for Collec
tive Action. Cambridge: Cambridge University Press.
Ostrom, Elinor (2005) Understanding Institutional Diversity. Princeton, NJ: Princeton Uni
versity Press.
Pavitt, Keith and William Walker (1976) ‘Government Policies towards Industrial Innova
tion: A Review’, Research Policy 5: 11–97.
Penrose, Edith (1959) The Theory of the Growth of the Firm. Oxford: Oxford University
Press.
Piketty, Thomas (2014) Capital in the Twenty-first Century. Cambridge, MA: Harvard Uni
versity Press.
Polanyi, Karl (1957 [1944]) The Great Transformation: The Political and Economic Origins
of our Time. Boston, MA: Beacon Press.
Rauch, James E. and Peter B. Evans (2000) ‘Bureaucratic Structure and Bureaucratic Per
formance in Less Developed Countries’, Journal of Public Economics 75(1): 49–71.
Page 26 of 29
Reinert, Erik S. (1995) ‘Competitiveness and its Predecessors: A 500-year Cross-national

Perspective’, Structural Change and Economic Dynamics 6(1): 23–42. (p. 336)
Reinert, Erik S. (2007) How Rich Countries Got Rich…and Why Poor Countries Stay Poor.
London: Constable.
Reinert, Erik S., Kenneth Carpenter, Fernanda A. Reinert, and Sophus A. Reinert (2017)
‘80 Economic Bestsellers before 1850: A Fresh Look at the History of Economic Thought’.
Working Papers in Technology Governance and Economic Dynamics No. 74. Available at
http://technologygovernance.eu/files/main//2017051103164242.pdf.
Reinert, Sophus A. (2011) Translating Empire: Emulation and the Origins of Political
Economy. Cambridge, MA: Harvard University Press.
Rip, Arje (2006) ‘A Co-evolutionary Approach to Reflexive Governance—and its Ironies’, in

Jan-Peter Voss, Dierk Bauknecht, and René Kemp (eds) Reflexive Governance for Sustain
able Development. Cheltenham: Edward Elgar, pp. 82–100.
Rodrik, Dani (1996) ‘Coordination Failures and Government Policy: A Model with Applica
tions to East Asia and Eastern Europe’, Journal of International Economics 40(1–2): 1–22.
Rodrik, Dani (2009) One Economics, Many Recipes. Princeton, NJ: Princeton University
Press.
469–91.
Rodrik, Dani (2015) ‘From the Welfare State to the Innovation State’, Project Syndicate.
January 14. Available at https://www.project-syndicate.org/commentary/labor-sav
ing-technology-by-dani-rodrik-2015-01?barrier=accesspaylog.
Rothwell, Roy and Walter Zegveld (1981) Industrial Innovation and Public Policy: Prepar
ing for the 1980s and 1990s. London: Pinter.
Ryan-Collins, Josh (2018) Why Can’t You Afford a Home? London: Polity Press.
Samuelson, Paul (1947) ‘Foundations of Economic Analysis’, Harvard Economic Studies,

80. Harvard University Press.
Schonfield, Andrew (1966) Modern Capitalism: The Changing Balance of Public and Pri
vate Power. Oxford: Oxford University Press.
Schot, Johan and W. Edward Steinmueller (2018) ‘New Directions for Innovation Studies:
Missions and Transformations’, Research Policy 47: 1583–4.
Soete, Luc and Anthony Arundel (eds) (1993) An Integrated Approach to European Inno
vation and Technology Diffusion Policy: A Maastricht Memorandum. Brussels: European
Commission.
Page 27 of 29
Stiglitz, Joseph E. (2010) ‘Government Failure vs. Market Failure: Principles of Regula
tion’, in Edward J. Balleisen and David A. Moss (eds) Government and Markets: Toward a
New Theory of Regulation. Cambridge: Cambridge University Press, pp. 13–51.
Teece, David J. (1996) ‘Firm Organization, Industrial Structure, and Technological Innova
tion’, Journal of Economic Behavior and Organization 31(2): 193–224.
Teece, David and Gary Pisano (1994) ‘Dynamic Capabilities of Firms: An Introduction’.
Working Paper No. WP-94-103. Laxenburg: IIASA.
Vom Brocke, Bernhard (ed.) (1991) Wissenschaftsgeschichte und Wissenschaftspolitik im

Industriezeitalter. Das “System Althoff” in historischer Perspektive. Hildesheim: Lax Ver
lag.
Weinberg, Alvin M. (1967) Reflections on Big Science. Oxford: Pergamon Press.
Williamson, James (2002) ‘What Washington Means by Policy Reform’. Available at

https://www.piie.com/commentary/speeches-papers/what-washington-means-poli
cy-reform.
Zehavi, Amose and Dan Breznitz (2017) ‘Distribution-sensitive Innovation Policies: Con
ceptualization and Empirical Examples’, Research Policy 46(1): 327–36.
Notes:
(1) See, for instance, the Blair government’s ‘Modernising Government’ White Paper from
1999, https://www.civilservant.org.uk/library/1999_modernising_government.pdf.
(2) The EU’s Lisbon Strategy from 2000 is perhaps the best-known example of this, see
https://portal.cor.europa.eu/europe2020/Profiles/Pages/TheLisbonStrategyinshort.aspx.
(3) On competitiveness, see Reinert (1995).
(4) This section builds on Mazzucato and Ryan-Collins (2019).
(5) Some eminent economists have rejected the market failure justification for policy in
tervention since the concept that markets by themselves lead to efficient outcomes is de
pendent on conditions—perfect information, completeness, no transaction costs or fric
tions—that have never been empirically demonstrated (Coase, 1960; Stiglitz, 2010).
Rather, markets are always incomplete and imperfect, and hence not ‘constrained Pareto-
efficient’, that is, they are never in a situation where a government (a central planner)
may not be able to improve upon a decentralized market outcome, even if that outcome is
inefficient (Greenwald and Stiglitz, 1986).
(6) This section builds on Kattel and Mazzucato (2018).
(7) This section builds on Mazzucato (2016).
(8) This section builds on Kattel et al. (2018).

Page 28 of 29
(9) This section builds on Mazzucato (2016).
Mariana Mazzucato
Mariana Mazzucato is professor of the economics of innovation and public value at

University College London (UCL), where she is founding director of the UCL Institute
for Innovation & Public Purpose (IIPP). She is winner of the 2018 Leontief Prize for
Advancing the Frontiers of Economic Thought and the 2019 All European Academies
Madame de Staël Prize for Cultural Values. She is author of many publications in
cluding The Entrepreneurial State: Debunking Public vs. Private Sector Myths (2013)
and The Value of Everything: Making and Taking in the Global Economy (2018). She
advises global policymakers on innovation-led growth.
Rainer Kattel
Rainer Kattel is an Estonian academic and science administrator. He is deputy direc

tor and professor of innovation and public governance at the Institute for Innovation
and Public Purpose, University College London.
Page 29 of 29
The Political Economy of Development Banking

Jonathan Di John
Because protectionism is now less relevant for supporting learning and innovation, devel
opment banking has assumed a greater role in industrial policy. This chapter presents the
economic reasoning as to why state-sponsored development banking corrects market fail
ures and creates productive capacities. Drawing on case studies, it explores how develop
ment banking enhances firm capacities to learn and innovate while also enhancing the
technical capacity of state bureaucracy to design and monitor state support of industrial
policy. The chapter makes an original contribution, discussing why and how economies
that encourage or mobilize diversified sources of long-run finance (rather than relying on
one state-owned development bank) enhance the effectiveness of their industrial policies.
Finally, it examines one of the world’s largest and well-run national development banks,
the Brazilian Development Bank (BNDES). This case study is useful for emphasizing the
broader political economy and macroeconomic context in which national development
banking takes place.
Keywords: development banks, industrial policy, political economy, long-run financing, late development, innova
tion, structural transformation, Brazilian Development Bank (BNDES), China Development Bank (CDB), German
Development Bank (KfW)
13.1 Introduction
NATIONAL development banks (NDBs) have played a dominant role in the provision of
long-run investment, ‘patient’ financing of catch-up in agriculture, industry, and infra
structure in the context of late development (Gerschenkron, 1962; Amsden, 2001; De
Aghion, 1999; UNCTAD, 2016), and in the context of innovation in advanced and emerg
ing economies (Mazzucato, 2015; Mazzucatto and Penna, 2016, 2018).1 For less devel
oped economies, late industrialization entails substantial risks (because many projects
are large scale and have long learning and maturation periods) that banks are often un
willing to undertake. Long-term finance requires maturity transformation, which involves
a risk that banks usually prefer to avoid (UNCTAD, 2016). For these reasons, NDBs are
Page 1 of 36
designed and mandated to fulfil the role of facilitating risk-taking and learning through
the provision of long-run development finance.
As Alexander Hamilton noted, in the context of post-colonial United States: ‘capital is

wayward and timid in lending itself to new undertakings…the State ought to excite the
confidence of capitalists who are ever cautious and sagacious, and in doing so, overcome
the obstacles that lie in the way of all experiments’ (Alexander Hamilton, (p. 338) Report
on Manufacturers, 1781). State-sponsored development banking providing low-cost, long-
run credit is thus central to socializing risk and inducing risk-taking and learning (Ams
den, 2001). Due to the lack of private-sector capacity and commitment to financing the in
herently risky ventures of late development and innovation, it is thus not surprising to
find that NDBs are often the only source of long-run credit in least-developing countries
(LDCs), and form one of the main instruments of both vertical and horizontal industrial
policy.
Surveys of the share of NDBs clearly indicate that they are not things of the past, nor rel
evant only to low-income contexts. A World Bank survey of NDBs found that 37 per cent
were established during the import-substitution years between 1946 and 1980, nearly
one-half (47 per cent) came into existence between 1980 and 2015 and 25 per cent since
2000, periods characterized by increasing globalization and the growing predominance of
neo-liberal policy reforms (De Luna-Martinez et al., 2017). Their presence in the financial
system thus remains significant, as they account for 25 per cent of total banking assets
around the world (De Luna-Martinez et al., 2017).2 This suggests that, irrespective of poli
cy orientation or global technological production patterns, policymakers across a diverse
set of economies at varying stages of development have found that private financial mar
kets do not deliver adequate long-term finance to support catch-up, innovation, and struc
tural transformation. These patterns contradict notions of a ‘life cycle’ of development
banks: the idea that a more active role for state-owned banks may be necessary, given the
lack of private capital markets and private entrepreneurship in less advanced stages of
development, but may become less important as economies transform and their financial
markets develop (Torres and Zeidan, 2016).
Moreover, the three largest NDBs in the world (China Development Bank (CDB), KfW
Group in Germany,3 and BNDES in Brazil) have substantially increased their asset and
loan disbursement as a share of GDP since 2000. As of 2015, these three NDBs had as
sets-to-GDP ratios in the range of 14‒18 per cent of GDP (Ferraz and Coutinho, 2019:
91).4
While there are several roles that NDBs can play in an economy,5 this chapter primarily
explores why and how they have played a dominant role in the provision of long-run in
vestment finance to support industrial policy and production strategies more generally.
The chapter’s main aim is to identify common technical and political economy factors that
have contributed to the effectiveness of national development banking. Prominent among
the factors that matter to such NDB effectiveness and their (p. 339) role in promoting dy
namic structural transformation include: developing sectoral technical expertise among
Page 2 of 36
staff to monitor projects and the capacity to crowd in financing (that is, to promote co-fi
nancing) for projects from other government agencies, private banks, and business
groups. To assist policymakers to identify political economy challenges, the chapter also
addresses why and when such factors fail to emerge. These political economy factors pro
vide a different view of less successful NDB performance from that of neo-liberal critics.
Section 13.2 considers the economic reasoning behind the argument that state-sponsored
development banking is central to socializing risk and inducing risk-taking and learning
as well as more innovative ventures across a range of LDCs, and indeed in advanced capi
talist countries. In doing so, it challenges the mainstream view that advocates financial
deregulation. Next, we draw on the historical experience of successful national develop
ment banking across a range of countries (Germany, Japan, South Korea, China, and
Brazil), to determine what effective development banking is and how it has contributed to
industrial policy across a range of cases in advanced and less developed countries. An im
portant message that emerges, and something that is neglected in the literature, is that
national development banks, when successful, can provide a focal point not only for en
hancing firm capacities to learn and innovate but also for enhancing the technical capaci
ty of the state bureaucracy to design and monitor state support of such activities. We
make an original contribution to the literature with a discussion of the implications of
whether economies develop single or multiple sources of long-run finance. In particular,
we discuss why and how economies that encourage or mobilize diversified sources of
long-run finance (rather than relying on one or a small number of state-owned develop
ment banks, which is common practice) enhance the prospects of their industrial policies
meeting the challenges of large-scale, long-gestating projects which entail risk and uncer
tainty. We next take a more extended look at one of the world’s largest national develop
ment banks, the Brazilian Development Bank (BNDES), the source of most of the Brazil
ian economy’s long-run financing. This case study is useful for examining issues in the
broader political economy and macroeconomic context in which national development
banking takes place. It also highlights some of the limitations of relying on a single
source of long-run financing, even when that source is one of the largest and most well-
run development banks in the world. Section 13.6 summarizes the general policy implica
tions of the link between development banking and industrial policy.
13.2 The Economic Rationale for Development

Banking
The general logic of subsidization to promote industrial production in the context of late
development is well known: investments (especially in high-productivity sectors where
scale and gestation period are large) entail significant risks. There are several (p. 340)
reasons for this. First, the existence of learning by doing (Arrow, 1962) implies that the
now more advanced countries have a ‘first-mover advantage’ of accumulated knowledge
compared with producers in late developers. Second, in advanced economies, frontier
firms have an implicit advantage because of the generally lower transaction costs of do
Page 3 of 36
ing business, which is the result of institutions being more predictable, and the quality of
physical infrastructure and education systems being higher (Abramovitz, 1986; North,
1990). Third, the risk inherent in undertaking industrial development in a late developer
in the context of relatively underdeveloped capital markets is generally too high to induce
investment in the absence of subsidization and protection.
The supply of state-sponsored, long-run credit in even advanced late developers is all the
more necessary because firms in OECD countries have further significant advantages.
First, there are indirect and direct mechanisms that support the technological advance of
their own frontier firms (Mazzucato, 2015). While R&D spending as a share of GDP may
be only slightly higher in advanced countries compared with late developers, the absolute
amounts of such spending are vastly higher in most OECD countries because GDP levels
are five to ten times higher. Second, the depth of financial markets makes the advanced
economies better able to withstand banking, financial, and balance-of-payments crises.
This reduces the level of business uncertainty over the long run. Third, the depth of finan
cial markets in OECD countries allows them to provide a diversity of long-run financing
sources, which increases the prospects of selecting winners in the uncertain and risky
world of innovation-driven investments. Finally, the high level of retained earnings in es
tablished Fortune 500 companies provides a vast supply of investment resources for un
dertaking R&D.
Gerschenkron (1962), who examined the late development process in Germany and Rus
sia, argued that catching up occurs by undertaking more capital-intensive investment in
individual plants, even though overall capital intensity of the backward economy is less.
In his analysis, investment banks and national development banks served as a functional
substitute for stock markets and commercial banking, both of which financed (along with
retained earnings) industrial investment in the forerunner country, Britain. Germany (as a
more advanced, but backward economy) relied on relation-based private investment
banking as the key to successful catch-up. Investment banks owned substantial shares in
industrial ventures. Russia (a much more backward economy) relied primarily on a state
development bank to finance industrialization. The key insight is that the stage of devel
opment is relevant to the type of development finance required, challenging the idea that
late developers should adopt ‘best practice’ of more advanced countries.
There are sound economic reasons why NDBs have enhanced the prospects of late devel
opment. Standard models of financing suggest that the private banking system is unlikely
to be able to provide long-run financing on its own for such risky ventures without state
coordination and guarantees (Stiglitz and Weiss, 1981; Dewatripont and Maskin, 1995; De
Aghion, 1999). The logic of the basic models is as follows. In a competitive banking sys
tem in a low-income economy, banks tend to underinvest in long-term projects. This is be
cause long-term projects involve large sunk costs (p. 341) requiring co-financing by sever
al banks. However, each bank will tend to provide a limited monitoring effort in the
knowledge that part of the marginal return from this effort will accrue to the other banks.
Insufficient monitoring jeopardizes project profitability, discouraging the co-financing of
long-term projects. This suggests a role for a coordinating agency in order to overcome
Page 4 of 36
‘free-rider’ problems and prevent short-termism. Given the inadequate private provision
of long-term finance, coordinating agencies are often sponsored by national governments.
The role of government-led and government-regulated state development banking goes

beyond standard market failure arguments. Late development plans or mission-oriented
governments are not simply ‘fixing market failures’. Rather, development involves the
processes of creating markets and firms that do not exist in the first place. Such ‘missing
markets’ are the result of inherent risks and uncertainty both in late development and in
innovation. In this sense, NDBs and their accompanying policies and organizations enable
and shape the creation of firms and markets, whether this involves infant industry pro
duction or innovation (Burlamaqui, 2012; Mazzucato and Wray, 2015).6
However, neo-liberal critics claim that state-controlled subsidized credit leads to several
problems: financial repression (negative interest rates reducing the incentives to save);
the crowding out of private-sector investment; relation-based governance which can gen
erate ‘insider privileges’; non-competitive markets; cronyism and corruption; and misallo
cation of resources towards over-ambitious capital-intensive projects in labour-surplus
economies (McKinnon and Shaw, 1973; Acemoğlu et al., 2004; Musacchio and Lazzarini,
2014a). Their policy advice is to liberalize the banking sector, attract foreign banking, let
private competitive markets determine the interest rate, avoid capital controls, maintain
fiscal and monetary discipline (balance budgets and target inflation), create independent
central banks insulated from political pressures, and adopt a rules-based system of fi
nancing to promote private stock and bond markets.
Neo-liberal policy advice has some important shortcomings. First, there is no historical
evidence that effective catch-up strategies have been based on following neo-liberal poli
cy advice (Gerschenkron, 1962; Rodrik, 2004; Amsden, 2001). Second, private financial
markets are subject to ‘manias, panics, and crashes’ (Kindleberger and Aliber, 2011) such
as the Great Depression and the 2008 global financial crisis, among many others in ad
vanced countries. Independent central banks, ‘light touch’ regulation, and monetarist
policies reigned supreme leading up to the 2008 financial crash. The private financial sec
tor seems subject to a massive ‘soft budget constraint’7—since (p. 342) large banks are
deemed ‘too big to fail’; in fact, taxpayers bail them out when they fail to the tune of tril
lions of dollars.
Moreover, according to Stiglitz and Uy (1996), assessing the success or failure of a direct
ed credit programme is difficult for three reasons. First, there is usually no way of know
ing whether growth would have been higher or lower in the absence of the programme.
Second, a good programme requires risk-taking, which means that failures are inevitable.
A programme with nothing but successes would necessarily have been too conservative.
Third, many of the returns may be long term, so current profitability may not provide an
adequate measure of success. For example, the measure of Korea’s chemical and heavy
industry programme, Japan’s car industry, or Brazil’s aerospace industry should not be
how those industries fared in the first decade of their existence, but what eventually be
came of their technological development, world market share, and so on thirty years after
Page 5 of 36
they began. Similarly, low profits may reflect cyclical conditions rather than long-run
prospects.
Despite the shortcomings of the neo-liberal view, it is still important to examine the fac
tors and conditions that have allowed a select set of NDBs to make positive contributions
to industrial policy and structural transformation, and to identify why many other coun
tries, even when they have well-run NDBs, fail to achieve similar outcomes.
13.3 Lessons from Effective Development

Banks
The aim of this section is not to recount the historical experiences of development banks.
Indeed, there is a substantial literature that covers cases such as Japan, Korea, Taiwan,
China, Germany, and Brazil, among others (see e.g. Griffith-Jones and Ocampo, 2018;
UNCTAD, 2016; Naqvi et al., 2018; Ferraz and Coutinho, 2019). Rather, it is to draw out
some of the general patterns and mechanisms underlying why and how development
banking supports effective industrial policy.
First, NDBs, unlike most commercial banks, have developed and mobilized the special
ized sectoral and financial skills required to deal with higher-risk, long-term investments.
As noted by Sayers (1957):
The logically sound basis for the presumption against long-term commitments is
that it is much more difficult to estimate a borrower’s creditworthiness twenty
years ahead than six months ahead. The factors relevant to creditworthiness are
substantially different over the longer period and the capacity and experience re
quired in (p. 343) the bank manager are of an altogether different order, an order it
is not reasonable generally to expect unless he has specialized expert staff.
(Sayers, 1957)
There have been many forms of long-run investment financing that have served the func
tion of socializing the risks inherent in late development. The French experience in the
nineteenth century, where significant developments in long-term state-sponsored finance
occurred, provides the pioneering example. The creation in 1848–52 of institutions such
as the Crédit Foncier, the Comptoir d’Escompte, and the Crédit Mobilier was particularly
important.
The involvement of the Crédit Mobilier in Continental European railway investment was
notable. The bank acquired substantial expertise in long-term finance as a result of rail
way investments. This expertise was then disseminated to other Continental European
banks in which the Crédit Mobilier held shares. Of even greater importance than the out
come of the operations of the Crédit Mobilier were intangible benefits such as the imitat
ed skills of the engineers and technicians which it sent abroad, the efficiency of its admin
Page 6 of 36
istrators, and its organizational banking techniques, which were so widely copied
(Cameron, 1953).
Indeed, one of the main reasons why private banking in LDCs cannot easily assume the
risks of long-run financing is precisely their lack of sectoral expertise in assessing and
monitoring risk. There is substantial evidence that the subsequent success of develop
ment banks in Japan, South Korea, Germany, France, and Brazil owes much to the well-
trained staff with engineering and sectoral skills able to monitor loans and form teams
that contribute to and influence planning and coordinate government policies (Amsden,
2001; UNCTAD, 2016; Ferraz and Coutinho, 2019; Griffith-Jones and Ocampo, 2018;
Naqvi et al., 2018).
In explaining why and how the German state development bank staff expertise is relevant
to enhancing the effectiveness of industrial policy, Moslener et al. (2018) note:
Apart from the financial expertise that KfW staff has, it additionally employs ex
perts with specific—often engineering-type—sector knowledge in agriculture, en
ergy, transport, water, natural resources, and civil engineering, to name but a few.
This substantively distinguishes KfW from the commercial banking sector. This al
lows KfW to base its investment decisions on a broader set of criteria from inter
nal employees rather than relying on the market generally or external actors, such
as consulting firms. This deeper understanding of sectors and the related markets
is also essential not only to identify market imperfections, but also to anticipate
the consequences of the respective interventions and programmes…Such knowl
edge further increases the likelihood that a particular project will be more suc
cessful from a socio-economic as well as a commercial perspective. Technical ex
pertise further allows KfW to serve as an important conduit between its invest
ments in the private sector and government policy, adding to its information ad
vantage. Finally, KfW’s stamp of approval can effectively signal to other private in
vestors that the project is viable.
(Moslener et al., 2018: 71; italics added)
Moreover, the financial expertise of KfW staff allowed them to monitor industrial loans ef
fectively, especially during downturns in the business cycles or during financial crises.
(p. 344) In particular, staff were better able than simple government agencies to separate
insolvent companies from healthy ones that were merely suffering from the reduced lend
ing capacity of the private sector due to fallouts from the financial crisis (Moslener et al.,
2018: 80‒1).
Second, NDBs have increasingly taken on a macroeconomic role of countercyclical financ

ing to combat increasingly unstable and volatile financial markets (Griffith-Jones and
Ocampo, 2008). This is important to help maintain long-term investment, including in in
frastructure, ensuring the continuity of existing projects and helping new ones start, valu
able both for short-term growth and long-term development. The multilateral develop
ment banks (MDBs) collectively increased their lending commitments to emerging and
Page 7 of 36
developing economies by 72 per cent between 2008 and 2009, the year when private cap
ital flows to these countries fell most sharply as a result of the crisis (Griffith-Jones and
Gottschalk, 2012). This countercyclical lending by multilateral and regional development
banks was complemented by that of NDBs in emerging and developed countries.8
Third, successful state development lending has complemented and ‘crowded in’ private-
sector lending rather than displacing it. The high private savings and investment rates in
East Asian countries in the 1960s and 1970s, for example, would not have been possible if
this were not the case. Although directed credit amounts to as much as 75 per cent of the
loans of financial institutions in some countries (such as Brazil), in the period 1960 to
1980, even in Korea (which used directed credit most aggressively), directed credit
amounted to only about 40 per cent of total credit, and in Japan it never exceeded 15 per
cent (Stiglitz and Uy, 1996).
There are several reasons why development banks have been more influential than their
small share in lending might suggest. Because of their close ties to the government, their
lending provided information to entrepreneurs and other banks on the areas that the gov
ernment was promoting. In addition, other financial institutions valued information on the
development banks’ choice of clients (as distinct from sectors). This signalling effect only
works, of course, if development banks have sound institutional reputations, which they
did in Japan, Singapore, Taiwan (China), and, by and large, Korea (Stiglitz and Uy, 1996).
A fourth factor that enhanced effectiveness was substantial amounts of directed credit to
industry, based on broad functional criteria (such as whether the firm produced exports),
typically using objective performance measures (Amsden, 1989, 2001; Stiglitz and Uy,
1996). This was particularly important in the rapid growth cases of North East Asia (Amd
sen, 2001) but has also been important in the export manufacturing drive in Germany
(Naqvi et al., 2018). This dynamic export performance results from domestic processes of
learning, effort, and contestation among competing firms (Woetzel et al., 2018). Of
course, it is also necessary for state competition policy to prevent oligopolistic collusion
and capture of state subsidies, since internal (p. 345) competition among the selected few
subsidy recipients is central to inducing the maximum efforts of firms learning to become
as efficient as possible (Studwell, 2013).
Fifth, effective export performance has been the result of explicit NDB strategies to pro
mote large firms and/or industrial conglomerates. The literature on ‘strategic trade’ (see
Krugman, 1987) finds that state subsidization matters to export competitiveness in the
context of increasing returns and imperfect competition. It is well known that develop
ment finance has been central to national firm formation and ‘national champions’ such
as Toyota, POSCO steel, or Hyundai in East Asia (Amsden, 2001). In Germany, in the
post-2000 period, the KfW aimed much of its very selective export financing at promoting
the export of industrial plant and machinery by a relatively small group of well-known
firms such as Siemens, Krupp, Ferrostaal, Ihde, Fritz Werner, and AEG to the major newly
independent developing economies, while specific export finance programmes targeted
shipping and aircraft exports, including companies producing for Airbus (Naqvi et al.,
Page 8 of 36
2018). In Brazil, BNDES finance was central to the export success of national champions
in agri-business, aerospace (Embraer, the world’s third-largest passenger jet company),
mining, oil and gas, among others.
Sixth, macroeconomic policies need to enable the generation of high levels of savings and
investment, and public and private development banking. Indeed, there is considerable
debate about what range of inflation and fiscal deficits is compatible with rapid growth
(Rodrik, 2016: 8). However, especially in less developed countries that do not have well
developed capital markets for government bonds, the greater the use of the inflation tax
(by printing money), the greater the prospects that hyper-inflationary pressure will
emerge.
Hyper-inflationary pressures result in increasing uncertainty and an increase in transac

tion costs, both of which are likely to slow down private-sector investment. Second, they
limit the credibility of government bonds which reduces the borrowing capacity of the
state to fund investments. It also tends to restrict the development of private banking
credit, which limits both the level and diversity of private credit expansion through the
banking system, which, in turn, limits the financing of investment (Calomiris and Haber,
2014). Finally, strong inflationary pressure tends to generate cycles of overvalued ex
change rates which reduces the competitiveness of infant industry exporters, reducing in
turn the extent to which export markets can become an opportunity to either expand mar
ket share or provide a realistic focal point for domestic firms to compete abroad (see Ro
drik, 2008, on exchange rate policy and industrial policy).
Avoiding balance-of-payments crises is a large part of maintaining a viable macro envi

ronment (Thirlwall, 2011, 2019).9 Because ‘big push’ industrialization drives (p. 346) gen
erate widespread ‘investment hunger’, the macroeconomic challenges of industrial policy
are formidable. The danger of overexpansion in a ‘big push’ strategy is typically the build-
up of unsustainable and destabilizing external debt positions that will occur if the invest
ment does not lead to productivity improvements, increasing competitiveness and net ex
port earnings. The debt overhang that could result also leads to increases in the share of
interest payments on the external debt, which in turn will reduce the state resource avail
able for investment in infrastructure, which is a central input into growth in this strategy.
As Amsden (2001) notes:
That conditions of ‘lateness’ are inherently conducive to overexpansion is suggest

ed by the fact that when a debt crisis occurs, it almost always occurs in a latecom
er country. This is because diversification in the presence of already well-estab
lished global industries involves moving from labour-intensive to capital-intensive
sectors characterized by economies of scale.
(Amsden, 2001: 252)
The scale of ‘big push’ investments was, indeed, instrumental in the debt crises in Latin
America in 1982 and in East Asia in 1997, as both were preceded by a surge in invest
ment (Amsden, 2001: 253). Thus, mobilizing national savings, both public and private, be
Page 9 of 36
comes central to sustaining a high investment effort without incurring potentially destabi
lizing levels of external debt. The higher savings and investment level in East Asia is
probably the main reason why its growth rates recovered after its macro crises, unlike
Latin America, where savings and investment rates are considerably lower. It is well
known that high levels of national public savings have supported robust investment rates
(Stiglitz and Uy, 1996; Krieckhaus, 2002).10
Thus, it is not surprising that NDB disbursements tend to grow and crowd in private in
vestment the most in East Asia and the OECD, economies where macroeconomic condi
tions generally allowed governments to develop deep capital markets, especially in the is
suance of long-run bonds. An often neglected and paradoxical trend is that the asset size
of the two largest NDBs in the world, CDB in China and KfW in Germany, have grown
fastest since the late 1990s, an era characterized by financial deregulation processes of
‘financialization’. The triple A credit ratings in Germany have allowed the KfW to become
one of the first national development banks to follow in the footsteps of the supranational
World Bank and European Investment Bank in tapping into international capital markets
(Naqvi et al., 2018). The dramatic rise in assets of the CDB have all been driven by its
ability to issue primarily domestic (but also international) bonds. Indeed, it has played a
central role in creating the domestic bond market (Xu, 2018). On the other hand, BNDES
in Brazil has grown substantially on the more (p. 347) precarious and politically fragile ba
sis of relying on government transfers of earmarked taxes and treasury transfers (Musac
chio et al., 2017). This is because the legacy of hyperinflation in the 1980s and 1990s gen
erated macro stabilization policies anchored in high real interest rates to accommodate
investors’ demand for short-term returns and liquidity (see section 13.4).
13.4 Diversification of Development Financing

Matters for Industrial Policy
While most national development banks are majority or wholly state owned, it is impor
tant to note that the vast majority of total investment, even where NDBs have been piv
otal to effective industrial policy and structural transformation (East Asia, Germany,
France, Brazil) is private, particularly for middle-income and high-income economies.
This draws attention to why and how NDBs can crowd in private investment and thus
leverage their benefits to the greatest possible extent.
The early experiences of development banking marked the role of government in en
abling the development of private development banking. In France and Germany from the
nineteenth century until the Second World War, government support for private emerging
development banks took the form of share capital provision, loans at lower than market
interest rates, the provision of state guarantees underwriting these institutions’ bond is
sues, or a combination of the three. From the early 1900s, the Japanese government
bought a substantial share of the bonds issued by private long-term credit banks and was
a catalyst ensuring that other private banks and financial institutions subscribed to these
bonds. It allowed development banks to issue long-term bonds or debentures, whose mar
Page 10 of 36
ket the government helped create. This privilege helped redress the mismatch between
the maturity structure of the banks’ assets and their liabilities, a problem that had
plagued commercial banks (Stiglitz and Uy, 1996). Limiting competition also enabled the
long-term credit banks to obtain funds more cheaply than they otherwise could have
(Stiglitz and Uy, 1996). Only in the post-Second World War period do we find that almost
all national development banks are state-owned enterprises (SOEs), mostly financed
through retained profits and domestic taxation (particularly a portion of personal income
taxes [PAYE], but also through domestic government bonds, international capital markets,
and concessional finance) (on the varied sources of NDB finances, ownership and lending
patterns, see UNCTAD, 2016; and Griffith-Jones and Ocampo, 2018).11
The current association of national development banking as a state-led process

(p. 348)
distracts attention from the wide variation in the importance of private long-run financing
across countries, particularly at low- and middle-income levels. Indeed, it is possible to
identify two basic patterns of relation-based sources of late-development and innovation-
oriented financing. The dominant pattern in most LDCs since the Second World War has
been the single/limited long-run financing model. This generally involves reliance on one
dominant national development bank financing public and private industrial firms. Per
haps the most well-known example is BNDES in Brazil (for a discussion of BNDES and
how it has worked, see Amsden, 2001; Musacchio and Lazzarini, 2014a, 2014b; Colby,
2012; Tavares de Araujo, 2013). In single long-run financing situations, such economies
lack depth but also diversity in long-run financing mechanisms. This may, as we shall see
in the case of Brazil, be the result of macro instability and chronically high inflation and
uncertainty due to balance-of-payments crises that inhibit long-run government and cor
porate bond markets, and/or explicit government policies limiting cross-shareholdings be
tween banks and industrial firms. The latter would mean that while business conglomer
ates may develop, such groups grow outside manufacturing.
The second basic model, and one more likely to sustain the investment and innovative ca
pacity of the economy, is a diversified/multiple financing model. This would include a na
tional development bank co-financing projects with state/regional development banks and
bank-based industrial conglomerates, as in Japan, South Korea, Taiwan, and Germany. In
China, the model is based around a national development bank, policy banks in agricul
ture and trade, and a decentralized system of state government (town and village enter
prises [TVEs,] and later local government financing vehicles), which constitute multiple
poles of long-run financing as well as competing with each other in the large internal
market (Qian and Weingast, 1997; Ang, 2016). In the United States, it is based on the
large retained earnings of corporations, public banks, and a vibrant investment and ven
ture-capital sector.
Indeed, national development banking has been effective in crowding in productive in
vestment when it has been able to draw on and nurture the construction of bank-based in
dustrial business groups. Historical evidence suggests that bank-based industrial busi
ness groups have been central to the catch-up processes in Germany, France, Italy, Swe
den, Japan, South Korea, Taiwan, and Israel (Khanna and Yafeh, 2007). Such relation-
Page 11 of 36
based financiers are able to take a longer-run view of catch-up investments (and there
fore become more ‘patient’ investors) than either stock markets or independent commer
cial banks, both of which operate under impersonal, rules-based systems. There is evi
dence that, when a dominant national development bank intervenes, co-financing
arrangements and/or co-ownership with private financial institutions can enhance the re
turns to capital of state-led long-run financing (De Aghion, 1999).
The multi-financing model has several advantages for crowding in private investment in
productive ways. First, it reduces the extent to which rent-seeking in search of long-run
funds becomes a political game centred on one or a small number of state (p. 349) banks.
While there may be considerable legal and illegal rent-seeking involved in trying to obtain
access to selective long-run state finance, the existence of several bank-based industrial
conglomerates will likely increase the competition and contestation over such funds. Such
contestation, even in countries with substantial cronyism and corruption, may provide in
centives to generate more viable and dynamic production empires, since economic
prowess will improve the prospects of winning these rent-seeking contests.
Second, it enhances the prospects of exchange of information, expertise, and collabora

tion between the public and private development banks. Business conglomerates are able
to pool scarce managerial talent and provide internal financing to undertake risky ven
tures that are difficult to otherwise finance in the context of underdeveloped capital mar
kets (Leff, 1978; Amsden, 2001; Khanna and Yafeh, 2007). The so-called transaction banks
in such conglomerates (particularly in Germany, Japan, Taiwan, and South Korea) comple
ment the financial, engineering, and other technical expertise to monitor sectoral invest
ments. Such conglomerates are important not only because in many catch-up processes,
large and concentrated firm size is a key to competitiveness (where scale economies are
relevant), but also because they provide an important complement to state planning and
coordination (Mahmood and Rufin, 2005: 352‒3).
Third, it increases the incentives for private business groups to monitor the quality of in
vestments, to increase their technology learning efforts, and to become more innovative,
because they have a stake in long-gestating and complex ventures. Indeed, one of the rea
sons behind the successes of NDBs in East Asia in developing dynamic firms and sectors
has been the requirement on the part of governments to demand substantial collateral in
return for loans as a way to incentivize effort (Stiglitz and Uy, 1996; Xu, 2018).
Finally, the diversity of long-run financing enhances innovation because it enables a

greater variety of experimentation. Apart from the need to generate a high volume of
long-run investment funds, the diversity of these funds becomes as important. This is be
cause a greater diversity of venture capital ‘bets’ increases the likelihood of success. As
Mokyr (1994) notes: ‘Diversity is the root of creativity, and much of what we call techno
logical creativity is the ability to absorb, assimilate and apply ideas borrowed from oth
ers’ (1994: 48). Indeed, the diversity of experimentation is also central to why the process
of market competition and rivalry produces discovery and innovation, a process neoclassi
cal economics has been ill-equipped to explain (Hayek, 1945). It is not a coincidence that
Page 12 of 36
the highest rate of patent applications and approvals (an albeit imperfect proxy of innova
tion) come from OECD countries but also emerging economies with diverse sources of
long-run financing (e.g. South Korea, China).
Recent evidence from the McKinsey Global Institute on the characteristics of countries
with the fastest long-run growth rates (so-called ‘outperformers’) suggests a greater pres
ence and performance of large firms (defined as firms with at least US$500 million in
sales). Moreover, this success is accompanied by greater competition or ‘creative destruc
tion’ among large firms within these economies: 55 per cent of the top quintile are re
placed within a decade compared to less than 40 per cent among (p. 350) slower-growing
economies (Woetzel et al., 2018). The report further argues that, rather than just picking
winning firms or sectors, the focus was on boosting productivity and enabling competi
tion. Performance criteria and the limited time frame of subsidies, and exposure to export
markets, all increased firm effort and disciplined lagging performance. Large firm devel
opment also received government support in the form of subsidies for infant industries,
including low-cost loans, preferential exchange rates, low tax rates, and R&D subsidies.
In sum, scale, and diversity of long-run financing are two central features of industrial
policy that delivers dynamic growth and structural transformation. As Amsden (2001)
first observed, the structure and size of business organizations matters for achieving the
scale and scope necessary to achieve dynamic infant industries and exporters in latecom
ers. The above discussion suggests it matters in other ways, such as providing a diversity
of financing options to support learning and experimentation, as well as providing the
prospect of internal competition among competing business groups. All the cases of long-
run growth identified as ‘outperformers’ have been countries with access to multiple
sources of long-run financing. For economies without this diversity in sources of long-run
financing, an important policy issue is how to diversify these sources. Several options are
being pursued today. First, setting up several, more specialized policy banks could diver
sify organizational effort. In fact, apart from the main development bank, CDB, China has
large policy banks such as the Export–Import Bank of China and the Agricultural Develop
ment Bank of China. Second, attracting foreign direct investment (FDI) diversifies long-
run finance since multinationals have retained earnings and access to global capital mar
kets. Third, reaching out to a large diaspora is an option for some countries such as Is
rael, India, Bangladesh, and Ethiopia. The latter, for instance, raised substantial capital
through diaspora bonds to fund the Renaissance Dam project. Finally, and most important
for low-income countries, international development banks have enhanced the diversity of
funding. The World Bank Group has done this in two ways. First, it has undertaken local
currency bond issuance through their signalling and demonstration effects, which
strengthens confidence in a country’s domestic bond markets attracting foreign issuers
and investors. Second, investment funds, such as the International Finance Corporation
(IFC), have invested in both established and emerging private-sector companies.
Page 13 of 36
13.5 Case Study: The Political Economy of De

velopment Banking in Brazil
The Brazilian case is useful for examining the broader political economy in which national
development banking takes place. While Brazil’s main national development bank, BN
DES, has been central to one of the most remarkable stories of growth and structural
transformation since the 1950s, its policies have been insufficient to prevent (p. 351) a
dramatic slowdown in manufacturing performance in the post-1990 period. Even more in
triguing is that the latter period is subject to enormous variations in manufacturing per
formance: a great industrial divide involving the rise of national champions and the fall of
the rest.
13.5.1 The Positive Side of Development Banking in Brazil
On the one hand, Brazil has had one of the best-performing and most successful national
development banks in the world. The national development bank, founded in 1952, initial
ly focused on financing infrastructure as part of the country’s drive towards moderniza
tion and industrialization. Later in the decade, its focus broadened to support the
country’s capital goods industry and then, in the 1960s and 1970s, other industrial sec
tors (Armijo, 2013). The state also introduced important ‘forced savings’ schemes which
were used to fund significant growth of the national development bank, BNDES, and the
spectacular growth of relatively well-run SOEs in heavy industry and mining, which con
tributed to public savings via their profitability and provided a focal point for technologi
cal development (Trebat, 1983).12 BNDES was central to the long-run financing that en
abled rapid growth and structural transformation in the period 1950 to 1980 during
which Brazil developed one of the most diversified and sophisticated industrial and manu
facturing sectors among middle-income countries.13
Since the 1990s, BNDES has been the focal point of industrial restructuring and ‘strate
gic’ large-scale privatizations (maintaining corporate control through ‘golden share’ mi
nority positions), and has supported exporting sectors (with a focus on capital goods and
engineering services) and, more recently, the internationalization of large national corpo
rations, the so-called national champions.14
The extensive privatization programme, in particular, transferred ownership on preferen

tial terms to established and emerging domestic business groups (Almeida, 2009; Massi,
2015). This process resulted in the emergence of firms in agri-business, steel, aerospace,
mining, oil and gas, paper and pulp, engineering and construction, and telecommunica
tions, many of whom have become innovative frontier firms.
The restructuring of Brazilian capitalism, or what Musacchio and Lazzarini

(p. 352)
(2014a) refer to as the ‘re-invention of state capitalism’, involved the construction of mul
tiple and diversified sources of long-run financing and a governance focal point around
sectoral restructuring for these targeted firms. This process took two basic forms. First,
under BNDES leadership, the government pursued a policy of purposefully constructing
Page 14 of 36
‘national champions’, largely through what were called ‘strategic privatizations’. The eq
uity of many of the viable, but financially strapped SOEs was transferred to private do
mestic business groups, but BNDES (through its investment banking arm, BNDESPar)
and state pension funds also maintained strategic minority shares. In the case of oil and
gas and some electricity firms, the state maintained a majority share but also received
vast amounts of BNDES loans.15 In both cases, export credits also increased substantially.
These state policies enhanced the value of rents by enhancing the profitability and techni
cal capacity of these firms to compete in export markets.
Second, the centrepiece of the construction of innovative ‘national champions’ was long-
run subsidized credit financing, most of which came from BNDES.16 The outcome of the
processes of privatization coordinated in the Cardoso administrations (1994‒2002) had
created a change in ownership patterns that was effectively a ‘coordinated’ process of re
structuring innovative firms close to the technological frontier. The basic pattern is de
scribed by Aldrighi and Postali (2010) as follows:
The complex arrays that have led BNDESPar, Banco do Brasil, Petrobras, and pen
sion funds of companies currently or formerly controlled by the government to
hold large equity stakes or even to take part in the controlling coalitions in some
business groups stem primarily from their participation in the privatization
process. Given its intent to maximize revenues from the privatization auctions, the
Federal Government urged BNDES and pension funds to act as financiers for the
bidders, notably Brazilian private groups.
(Aldrighi and Postali, 2010: 372)
The state created rents by assuming the socialization of risk in promoting innovative ac
tivity and increasing the scale of business groups and firms through a combination of
loans and taking equity positions. With the rise in BNDES funding in the post-2002 period
(reaching a peak of 4.3 per cent of GDP in 2010), these loans tended to target the largest
firms in the country: in the period 2002 to 2011, the ten largest loans made up 35 per
cent of all BNDES disbursements (Hochstetler and Montero, 2013: 1491, table 1). Other
mechanisms through which the state increased the diversity of long-run financing and
growth potential of emerging national champions were the increasing equity (p. 353) par
ticipation of state pension funds (Almeida, 2009), the provision of export subsidies and
credits,17 and the promotion of local content laws, through public procurement policies,
in sectors such as oil, shipbuilding, automobiles, and wind turbines (Almeida and Schnei
der, 2012: 17).
The strategy of creating frontier, innovative firms was not just about financing, but also
about increasing their scale, promoting their internationalization into emerging multina
tionals able to compete at the apex of global value chains, and promoting their innovative
capacity. The result of this process is a series of world-class Brazilian multinationals at
the cutting edge of frontier technologies18 in aerospace, high-tech agriculture, agri-busi
ness, steel, telecoms, mining, oil and gas, bio-fuels, automobiles, and bio-technology
(Brainard and Martinez-Diaz, 2009; Musacchio and Lazzarini, 2014a, 2014b; Perez-Ale
Page 15 of 36
man and Chaves-Alves, 2016; Ferraz and Coutinho, 2019).19 Many of Brazil’s ‘national
champions owe their existence to past capacities that were constructed and developed as
state firms in this period’ (Aldrighi and Postali, 2010).
By 2015, BNDES was the third-largest national development bank in the world, after the
China Development Bank and Germany’s KfW. In the context of a country where commer
cial banking has focused on short-term loans, BNDES, with its strong technical expertise
and record of promoting structural transformation, has been held up as a role model for
countries considering setting up development banks (UNCTAD, 2016). (p. 354)
13.5.2 The Limitations of BNDES and the Importance of Broad Politi

cal Economy Factors
Since 1990, despite the best efforts of BNDES to increase disbursements and fund world-
class national champions, overall GDP and manufacturing output and productivity growth
rates have collapsed and there was substantial de-industrialization from as early as
1980.20 The share of the manufacturing sector in total GDP fell from a peak of 31.3 per
cent in 1980 to less than 14 per cent in 2015. While BNDES investment may have ‘crowd
ed in’ private-sector investments, it has done so within a range that is far too low: the pri
vate investment rate maintained a rate in the 14‒15 per cent range throughout the peri
od, which is well below East Asian counterparts. Whatever the particular design of indus
trial policy, this low level of investment ultimately generates fewer experiments, risks,
and learning processes.
The Brazilian experience suggests that macroeconomic policy and context as well as poli
cy decisions on banking laws have contributed to reliance on a single/limited long-run fi
nancing model. As a result, both scale and diversity of long-run financing mechanisms
have been limited, even during periods of rapid growth. The subsequent slowdown in
growth after 1980 suggests that the nature of development patterns matters more for
avoiding growth collapses that are common features of LDCs. Let us examine the histori
cal evolution of the processes behind this problem.
First, a series of development banking laws and policies inhibited the development of pri
vate-bank-based industrial conglomerates, and therefore limited multiple sources of long-
run finances. Many of these policies, either initiated or enforced by BNDES, were meant
to promote internal competition. BNDES put conditions on its loans specifying low debt/
equity ratio ceilings that constrained the size of national firms (Amsden, 2001: 226). This
decision, along with a banking law in 1964 which restricted banks and industrial firms
from having cross-holdings (Armijo, 1993) was meant to both create internal competition
and prevent further concentration of industrial assets which historically were linked to
the ‘insider’ advantages that big businesses had with a banking system with limited ac
cess to credit (Calomiris and Haber, 2014: 390–414). In a country with historically very
high levels of income and asset distribution, the rationale for this policy is understand
able. However, the policy constrained the size of domestic manufacturing firms (Amsden,
2001: 227) which is one of the reasons why there were very few capital goods firms that
Page 16 of 36
exported. They simply did not have the scale and scope to compete. It was thus not sur
prising to find, among all less developed (that is, non-OECD) countries, there was no pri
vate Brazilian business group in (p. 355) manufacturing in the top 50 (ranked by sales) in
the world in 1993 (Amsden, 2001: 203, table 8.5).
The second factor that limited the development of private investment banking in the
twentieth century was the increasingly excessive use of the inflation tax (by printing mon
ey) through the mid-1990s. Inflation rates averaged over 20 per cent in the period 1946
to 1963, and were to reach 60‒70 per cent by 1962‒64 (Calomiris and Haber, 2014: 391).
This ultimately culminated in hyperinflation in the 1980s and early 1990s. Chronic and
high inflation resulted in increasing uncertainty and an increase in transaction costs, both
of which inhibited the emergence of a private investment banking sector and resulted in
the state becoming the main source of long-run development financing. Second, it limited
the credibility of government bonds, reducing the borrowing capacity of the state to fund
investments, and also tended to restrict the development of private banking credit. This
limited both the level and diversity of private credit expansion through the banking sys
tem, which, in turn, limited the financing of investment (Calomiris and Haber, 2014). Fi
nally, strong inflationary pressure tended to generate cycles of overvalued exchange rates
which reduced the competitiveness of domestic manufacturing firms and thus their ability
to increase exports.21
Third, the growing reliance on external debt made the economy vulnerable to balance-of-
payments crises. The main culprit behind the worsening economic performance of the
late 1970s to the mid-1980s was the reliance on a debt-led growth strategy, which result
ed in a growing balance-of-payments crisis, increasing external debt and rising inflation
(Fishlow, 1989; Bacha and Bonelli, 2013). Due to the unforeseen dramatic rise in world in
terest rates, the country was forced to declare a suspension of external debt payments at
the end of 1982.22 The inability of the central government to control and discipline state
financing contributed to a spiralling and unsustainable debt. Throughout the 1980s and
into the 1990s, governors used the state banks to finance expanding deficits, accumulat
ing staggering debts. State bank debts had reached US$70 billion by 1991 and amounted
to a massive US$96 billion in 1998 (Kingstone, 1999: 177). State and local governments
were responsible for about two-thirds of Brazil’s US$148 billion foreign debt (Mainwar
ing, 1999: 191). The climate of hyperinflation and the debt restructuring imposed led to a
series of policies that would both hamper state finances and negatively affect the invest
ment climate (Aldrighi and Postali, 2010: 360‒1).
Fourth, the macro policies after 1994 to handle the legacy of inflation also inhibited pri
vate investment banking. Macro stability, in particular sustaining low inflation, was a
(p. 356) main priority of each of the administrations in this period. The history of inflation
taxes and episodes of hyperinflation had eroded the political support of the poor (who
mostly paid for this tax) and asset owners, particularly within industry, who were unable
to undertake meaningful long-run strategic planning. This price stability was achieved,
however, by maintaining real interest rates that were among the highest in the world, as
well as an overvalued exchange rate, in the period 1999 to 2015 (Afonso et al., 2016).23
Page 17 of 36
The legacy of hyperinflation meant that Brazilian savers demanded two things: high inter
est rates and short-term returns to ensure liquidity (Orair and Gobetti, 2017: 14).
Among the consequences of this macro policy environment is that it reduces the effective
ness of BNDES action on industrial policy. First, in the period 1998 to 2014, interest pay
ments on debt absorbed more government spending than did infrastructure, education,
and health (De Magalhães and Costa, 2018: 18‒22). Second, high real interest rates re
duced the extent to which BNDES policies could ‘crowd in’ private investors who were
unwilling to undertake risks despite government efforts provide a substantial amount of
loans at subsidized interest rates (Ferraz and Coutinho, 2019).
As a result, the ‘national champions’ policies, while generally successful, appropriated a

significant proportion of long-run financing. This occurred in the context of an economy
that did not have alternative forms of long-run financing investment arrangements, such
as bank-based industrial conglomerates, or well-financed regional development banks.
While BNDES disbursements increased to 4.5 per cent of GDP after 2007, 60 per cent of
that funding essentially went to fund the formation of ‘national champions’. That left only
40 per cent (or a little over 1.5 per cent of GDP) available for long-run subsidized funding
for potentially innovative smaller and medium-sized firms—not enough to drive produc
tive investment. Moreover, policies to reduce the role of state development banks, while
understandable given their previous activities building up massive external debts in the
1980s, undermined the extent to which a more decentralized and diversified set of long-
run financing organizations would emerge across levels of government, as has occurred
in China, South Korea, or Germany. In 2000, fiscal governance was reformed by the adop
tion of hard-budget constraint legislation—the Fiscal Responsibility Law—which imposed
transparency requirements on states, and municipalities’ accounts and imposed a ceiling
on states’ spending.
13.5.3 Policy Debates: Assessing the Role and Effectiveness of BN

DES in Brazil
There are several specific criticisms of the ‘national champions’ model. The first set of ar
guments revolves around the criticism that BNDES was focusing too much of its (p. 357)
lending on sectors that were already internationally competitive, such as Petrobras and
the agri-business sector (Almeida, 2009; Almeida and Schneider, 2012; Tavares de Araujo,
2013; Musacchio and Lazzarini, 2014b). According to this line of argument, lending
should not have focused on firms that were at the frontier, since industrial policy should
focus on providing incentives to stimulate novel learning instead of reinforced specializa
tion in utilities and commodities such as mining, oil, steel, agri-business, and aerospace
(Almeida and Schneider, 2012; Musacchio and Lazzarini, 2014b: 31‒2).
It is true that BNDES followed a strategy of ‘going with winners’ or ‘betting on the
strong’ rather than ‘picking winners’. This is revealed in the fact that the default rates on
its loans in the period 2009 to 2011 are almost nil, at 0.16 per cent compared to the na
tional financial system rate of 3.8 per cent (Colby, 2012: 19). This is hardly the profile of a
Page 18 of 36
venture capitalist investment fund. One can also make the case that because BNDES
maintains these equity shares, they are not ‘letting go of winners’ by selling on these
shares to private equity firms.
The second, and related, criticism is that BNDES lending reinforced a pattern of so-called
‘low-technology-intensive’ sectors. Almeida and Schneider (2012) argue that ‘if industrial
policy is supposed to create new competitive advantages, then it has failed to do so: the
most competitive industrial sectors in 2011, measured by the trade surplus, are the same
as in 1996, despite the goals in successive industrial policies to promote technology-in
tensive sectors…Rather the bias towards conservatism has consolidated the specializa
tion of Brazilian exports around resource-based industries and commodities’ (2012: 18‒
20). Both criticisms add up to the diagnosis that there was a misplaced focus on competi
tive ‘insiders’ at the expense of potentially more innovative ‘outsiders’ who were not yet
competitive and thus the share of medium- and high-tech production suffered as a result.
A third criticism concerns the returns to capital investment that BNDES equity participa
tions, through its investment banking arm, BNDESPar, have yielded. Bruschi et al. (2013)
compare the individual performance of the shares held by BNDESPar with returns on the
Ibovespa index, the main stock market index in Brazil. For firms that BNDESPar acquired
after 2004, they find that 60 per cent of the shares performed worse than the stock mar
ket index in the same period, 2004 to 2012. Their conclusion is that BNDES will be un
likely to continue to generate the majority of their income from equity investments.
The criticism that BNDES subsidized already competitive firms is misleading. These lend
ing patterns need to be seen in historical and political context. First, the fragility of BN
DES financing in the 1980s has created an understandable conservative bias in its spend
ing since 1994. The legacy of macroeconomic instability can impair the ability of state de
velopment banks to finance risky new ventures in the high-tech and capital goods sectors,
since the very legitimacy of the bank’s mission needs to be reconstructed.
Second, it is not clear that many of these near-frontier technology firms would have been
competitive in the context of the well-known custo Brazil (Brazil Cost)—the relatively high
prices of many domestic goods and services, a reflection of the (p. 358) concentrated pro
duction structure, and the failure of government regulation to introduce competition laws
(which implicitly meant that producers in several sectors are effectively earning large
rents).24
Third, it is not at all clear that without long-run subsidized financing and state equity in
jections many of them would have become the innovative firms they did. There are not
many examples of foreign investment banks funding long-run R&D programmes of firms
in LDCs and there is little evidence that the private domestic banking system would have
done so either. Moreover, many of the export credits were essential to the export success
of these firms: Embraer, for instance, depended on them to keep its products competitive
(as KfW has done with Airbus).
Page 19 of 36
Fourth, since NDBs take risks, there will be inevitable failures. Equity positions will allow
them to reap some of the financial rewards in order to offset the inevitable failures and
ensure a stable funding source so it does not lose its appetite for risk (Mazzucato and
Penna, 2018).
Fifth, the idea that BNDES lending patterns have reinforced the pattern of ‘low-technolo
gy’ exports is misleading. The categories ‘low-tech’ and ‘high-tech’ direct attention away
from the point made by Baldwin (2013) that the integration of global value and supply
chains means sectors are less relevant than the stages of production. And what matters is
the capacity to innovate at the level of design and product development. In this respect,
most of the ‘national champion’ firms are innovative or based on state-supported R&D in
vestments. The fact that Brazil is competitive in agri-business (supposedly ‘low-tech’) as
well as in aerospace (supposedly ‘high-tech’) owes more to the capacity to design and de
velop new products than to the sector it is producing for. There are many countries in
South Asia and sub-Saharan Africa, including South Africa, and even in South East Asia,
that would more than welcome the innovative capacity of so many so-called ‘low-tech’
sectors in agri-business, steel, and oil and gas that Brazil has developed (see Chapter 1 of
this volume for a discussion of the problems of outdated categorization of what is ‘indus
trial’ or ‘low-tech’, which obscures dynamic, technologically sophisticated activities tak
ing place within resource-based industries, agriculture, and services).
13.5.4 The Politics of Development Banking in Brazil
This discussion of development banking in Brazil suggests that policy design and reform
need to be understood in terms of broader political economy dynamics. The economy in
the post-1990 period maintained a limited diversity of long-run financing (p. 359) organi
zations, in large part because the dominant large business groups had their core busi
nesses outside manufacturing, and also because commercial banks were deriving sub
stantial profits from purchasing high-yielding government assets (as a result of anchoring
inflation on high real interest rates and maintaining fiscal surpluses) and other financial
assets (particularly equities and derivatives). This made the opportunity cost of investing
in risky innovation-based start-ups and/or in experienced but financially strapped capital
goods industries and start-up ventures very high. In this context, the amount of subsi
dized long-run financing that was left over for ‘outsider’ but potentially innovative firms
was simply too small to provide the vast amount of long-run financing that innovation-
based strategies require. What is more, the BNDES leadership was well aware of the
need to promote innovation and there were indeed many smaller loans to a variety of in
novative firms (see Hochstetler and Montero, 2013). The ‘insider/outsider’ problem was
not due to a lack of technically competent, well-meaning, and informed technocrats; it is
generally accepted that Brazil does not have a skills shortage in its main development
and banking agencies. Rather, this problem, along with the insufficient savings and in
vestment rates, has much more to do with the historical political economy dynamics, and
in particular, the way economic and political order was maintained in this period.
Page 20 of 36
The political nature of development banking is evident in recent political crises in Brazil.
The discovery of the disappearance of nearly US$10 billion from the accounts of Petro
bras in late 2014 set off the largest corruption scandal in Latin American history. The La
va Jato probe, undertaken by an increasingly proactive and independent judiciary and
federal police in 2015, has uncovered a complex bribery and kickback scheme in which,
among other things, the biggest construction companies and other private industrial
groups, many of whom received massive BNDES financing, paid bribes to politicians’ per
sonal accounts and to political party campaign funds in exchange for padded government
contracts in construction, shipbuilding, and other Petrobras investments.
Large-scale corruption scandals are not new to Brazil, nor is there evidence that BNDES
knew of the corrupt practices of its clients, but the magnitude and political fall-out reveal
more profound tensions than the merely political. An important aspect of the discontent
concerns the alienation, for a number of reasons, of much of the middle class. First, there
was growing discontent around the poor quality of public services such as transport, edu
cation, and health, as well as increasingly expensive housing, fuelled by the boom in fi
nancial and commodity markets. Much of the frustration was compounded by the inability
of the government to address the growing levels of crime and urban violence. Second,
much of middle-class discontent stemmed from the fact that they have borne the brunt of
the increased tax burden to fund pro-poor expansions in the welfare state and have not
received proportionate benefits (Saad Filho and Morais, 2014). Third, middle-class groups
were losing much of their economic privileges relative to lower income groups. This has
occurred because of the stagnation of average real wages in the context of minimum
wages rising, and because there was a (p. 360) decline in the creation of well-paid employ
ment, most likely the result of the stagnation in manufacturing employment and the rise
of lower-paid service employment.25
A further source of discontent, not only among the middle class, but also among some big
business groups, was the perceived ‘insider bias’ in the massive loan portfolio of BNDES,
the main provider of long-run financing.26 The most infamous case, and a focal point of
protest, was the case of JBS, the meat-packing company, which borrowed over 8.1 billion
reais from BNDES to purchase meat-packing companies in the United States, Europe,
and Australia, and became the world’s largest meat packer. In the process, BNDES pur
chased a 21.3 per cent stake in JBS in 2007 (Leahy and Schipani, 2017). While big busi
ness groups in oil, shipbuilding, and construction, along with some food groups, com
prised an inner circle of support for the Workers Party (PT)-led administrations between
2004 and 2015, other groups, particularly the banking sector and media conglomerates,
were committed opponents of these strategies, which they viewed as promoting crony
capitalism and excessive state control over resources.
The discontent over such ‘corporate welfare’ became a focal point for middle-class
groups, finance, and the media in street protests against the government from 2014 on
wards and were a focal point of anti-corruption discourse against the PT and the broader
political party system. The corruption scandal became an instrument of political contesta
tion and protest (Leahy and Schipani, 2017). Given media and banking opposition to PT, it
Page 21 of 36
was not surprising that a focal point of the scandal was the link between Petrobras and
big insider supporters of PT in oil, shipbuilding, construction, and food (despite the fact
that all political parties were involved in Lava Jato; Leahy and Schipani, 2017). Within
Congress, the opposition to the PT ultimately led to the impeachment of President Dilma
Rousseff and the demise of the legitimacy of state-led development banking.
Of course, the corruption scandal itself has had a profoundly negative effect on the econ
omy because of increased political uncertainty. The companies involved in the scandals,
some of the largest in the country, have faced profound financial consequences. Petro
bras, which in 2014 accounted for approximately 13 per cent of Brazil’s gross domestic
product, was estimated to have lost at least US$88 billion by mid-2015. Petrobras and
Odebrecht have sold assets worth billions of dollars to pay their debts. Unemployment in
Brazil hit a new high in April 2017 at 13.6 per cent, more than (p. 361) double the rate it
was in late 2013, prior to the start of Lava Jato. Thus even a well-run organization like
BNDES has not been immune to the tempests of political contests and controversies. The
current administration under President Jair Bolsonaro has initiated a process of reducing
its role in the economy.
13.6 General Policy Considerations

First, the effectiveness of a national development bank depends upon the same factors
that make industrial policy effective more generally. It requires a clear set of ‘carrot and
sticks’, and the ability to ‘let go of losers’ (Amsden, 2001). From this perspective, the fo
cus on simple, objective performance criteria, such as export growth, has been helpful for
implementing reciprocal control mechanisms. When successful, national development
banks have been able to monitor projects and loans with sectoral experts, many of whom
hold PhDs in a wide range of areas. Relying on financial experts alone is insufficient.
Of course, the effective implementation of a reciprocal control mechanism implies that

the state has the power and authority to mobilize resources and provide a ‘carrot-and-
stick’ mechanism. Implicit in this is the political power to remove subsidies from powerful
economic elites when their infant industries perform poorly. Moreover, if a reciprocal con
trol mechanism is to be financed, the state requires the political power not only to mobi
lize resources but to ensure that a high proportion of loanable funds in the financial sec
tor are channelled towards long-gestating manufacturing or other high-value investments
as well as securing sources of funding for development banking (Woo, 1991; Kapadia,
2016).27
Second, NDBs do not necessarily need to finance all of the venture capital on projects.
The experience of East Asian and other economies (such as Germany, and in some in
stances, Brazil) suggests that policymakers should consider diversified/multiple financing
sources by getting endowment funds, donors, domestic banks, and conglomerate groups
and foreign firms to co-finance targeted projects. Even when the state privatizes SOEs in
strategic sectors, that does not mean it needs to lose control over industrial policy.
Page 22 of 36
Instead, the state can maintain ‘golden shares’ in priority projects to maintain state con
trol over strategic decisions, as has been practised in the privatization process overseen
by BNDES in Brazil. In fact, according to Musacchio and Lazzarini (2014a, 2014b),
(p. 362) the ‘state as minority shareholder’ model has several advantages. First, it limits
the use of SOEs as a conduit for clientelist practices and expensive social policies such as
(uneconomic) employment creation. Second, majority private shareholding enhances the
incentives for focusing on profitability, which should increase firm efficiency and growth.
Third, the state is still in a position, as a significant shareholder, to influence corporate
strategy and development policies in strategic sectors. Finally, state holdings provide ac
cess to long-run financing, which is underdeveloped and expensive in private Brazilian
capital markets. Indeed, in a survey on the re-emergence of state capitalism, The Econo
mist (2012b) refers to the ‘Leviathan as minority shareholder’ model as ‘one of the
sharpest new tools in the state-capitalist toolbox’ (2012b: 5).
Third, and related to the previous point, NDBs need to secure stable sources of funds so
that they can act like venture capitalists and ‘lenders of first resort’. That is, NDBs ‘must
be able to strike the right balance between risks and rewards, ensuring that investments
are structured across a risk-return spectrum so that lower risk investments help to cover
higher risk ones’ (Mazzucato and Penna, 2018). The ability to secure long-run bond fi
nancing is one main option along with maintaining equity stakes. The latter allows the
bank to benefit from the rare successes in order to offset the inevitable failures. Equity
shares also help the state more easily monitor the frequent tax evasion strategies of firms
who have benefited from NDB financing. There is evidence from mineral and fuel mining
that governments retain a higher share of the proceeds from mineral and fuel rents when
state equity shares in these sectors are higher (Lundstøl et al., 2013, 2018). This is be
cause state presence on boards of directors makes it harder for companies to evade tax.
Fourth, in order to ensure macroeconomic sustainability in its balance of payments, NDBs

should target sectors and projects that generate net foreign exchange earnings. This is
particularly the case for growth strategies that rely on the increasing use of foreign sav
ings to finance investment as in, say, contemporary Ethiopia. When plans for NDB financ
ing do not consider the importance of generating net foreign exchange, balance-of-pay
ments crises tend to follow, which can in turn precipitate a growth collapse, even in coun
tries where the NDB is sound in terms of technical capacity (e.g. Brazil, 1982‒94).
Fifth, the historical evidence suggests that national development banks have contributed
to effective industrial policy through the financing of ‘national champions’. This has been
achieved through the financing of infrastructure and agricultural research in a range of
sectors and heavy industrial projects, particularly in the steel, chemical, electronics, auto
motive, and mining sectors, aerospace, and capital goods. This financing has been orient
ed towards public enterprises, but also towards large-scale domestic private conglomer
ates. The emphasis on financing domestic firms stems from the evidence that the most dy
namic and innovative firms in LDCs are not generally subsidiaries of multinationals (Ams
den, 2008). This emphasis on vertical industrial policy, or national firm formation, is cen
tral to the development of sophisticated and complex production (Amsden, 2001), since
Page 23 of 36
firms form the focal point through which people with skills and ideas can collaborate and
eventually migrate to form new (p. 363) ventures. It is no accident that major industrial
clusters, such as Detroit in the 1930s or Silicon Valley in the 1980s, took off from the
spin-offs of workers from one lead firm in each case (Hausmann, 2018).
Sixth, effective NDB financing achieves sectoral and firm success when it involves ‘target
ed ambition’. NDB have unwritten complex missions in aerospace, high-tech agriculture,
capital goods, high-speed trains, wind and solar technology, bio-fuels and bio-technology,
among others. These missions have worked when they involved funding hundreds or even
thousands of PhDs to work in relevant research centres to help firms design and develop
the technology for innovative capacity, and when the NDB has collaborated with relevant
government agencies (e.g. see Perez-Aleman and Alves, 2016 on Brazilian bio-technology
efforts; Figueiredo, 2014 on Brazilian agriculture; Moslener et al., 2018 on Airbus). Albert
Hirschman long ago pointed out that ‘big push’ development projects can overwhelm gov
ernment capacity. However, he also argued that development is about taking on ambi
tious projects and developing the capacity to solve problems as they arise in the process
of learning and experimentation. NDBs provide a focal point to implement selective indus
trial policy by providing the patient capital and sectoral expertise to undertake these mis
sions.
Seventh, macroeconomic policy and conditions need to enable the development of private
capital markets and long-run government bond markets. While financial deregulation pos
es many challenges to state intervention and industrial policy, the dramatic increases in
the size of the Chinese and German development banks is the result of their credit rat
ings and ability to issue long-run bonds. In the case of China, the CDB has itself been in
strumental in creating bond markets. It is ironic that some of the most orthodox central
banks (e.g. Bundesbank in Germany) provide the context through which significant in
creases in state development banks (e.g. KfW) can grow.
References
Abramovitz, Moses (1986) ‘Catching up, Forging ahead and Falling behind’, Journal of
Economic History 46(2): 385–406.
Acemoğlu, Daron, Philippe Aghion, and Fabrizio Zilibotti (2004) ‘Distance to Frontier,
Selection, and Economic Growth’, Journal of the European Economic Association 4(1): 37–
74.
Afonso, José, Eliane Araújo, and Bernardo Guelber Fajardo (2016) ‘The Role of Fiscal and
Monetary Policies in the Brazilian Economy: Understanding Recent Institutional Reforms
and Economic Changes’, The Quarterly Review of Economics and Finance 62: 41–55.
Aldrighi, Dante and Renato P. Colistete (2013) ‘Industrial Growth and Structural Change:
Brazil in a Long-run Perspective’. USP Working Paper No. 3. Department of Economics-
FEA.
Page 24 of 36
Aldrighi, Dante and Fernando A. Postali (2010) ‘Business Groups in Brazil’, in Asli M. Col
pan, Takashi Hikino, and James R. Lincoln (eds) The Oxford Handbook of Business
Groups. Oxford: Oxford University Press.
Almeida, Mansueto (2009) ‘Desafios da real política industrial brasileira do século XXI’.
Discussion Paper No. 1452, Instituto de Pesquisa Econômica Aplicada (IPEA). (p. 364)
Almeida, Mansueto and Ben Ross Schneider (2012) ‘Globalization, Democratization, and
the Challenges of Industrial Policy in Brazil’. Korean Development Institute, July.
Amsden, Alice (1989) Asia’s Next Giant. Oxford: Oxford University Press.
Amsden, Alice (2001) The Rise of ‘the Rest’. Oxford: Oxford University Press.
Amsden, Alice (2008) ‘The Wild Ones: Industrial Policies in the Developing World’, in Nar
cís Serra and Joseph E. Stiglitz (eds) The Washington Consensus Reconsidered: Towards a
New Global Governance. Oxford: Oxford University Press, pp. 95–118.
Ang, Yuen Yuen (2016) How China Escaped the Poverty Trap. New York: Cornell Universi
ty Press.
Arias, Diego, Pedro A. Vieira, Elisio Contini, Barbara Farinelli, and Michael Morris (2017)
Agriculture Productivity Growth in Brazil: Recent Trends and Future Prospects. Washing
ton, DC: World Bank.
Arizala, Francisco, Jesus R. Gonzalez-Garcia, Charalambos G. Tsangarides, and Mustafa

Yenice (2017) Growth Breaks and Growth Spells in Sub-Saharan Africa. Washington, DC:
International Monetary Fund.
Armijo, Leslie (1993) ‘Brazilian Politics and Patterns of Financial Regulation, 1945–1991’,
in Sylvia Maxfield, Chung H. Lee, and Stephan Haggard (eds) The Politics of Finance in
Developing Countries. New York: Cornell University Press, pp. 259–90.
Armijo, Leslie (2013) ‘The Public Bank Trilemma: Brazil’s New Developmentalism and the
BNDES’, in Peter Kingstone and Timothy J. Power (eds) Democratic Brazil Ascendant.
Pittsburgh, PA: University of Pittsburgh Press, pp. 230–47.
Arrow, Kenneth J. (1962) ‘The Economic Implications of Learning by Doing’, Review of

Economic Studies 29(3): 155–73.
Bacha, Edmar L. and Regis Bonelli (2013) ‘Padrões de desenvolvimento industrial no

Brasil: passado e futuro’. Mimeo.
Bagnai, Alberto, Arsène Rieber, and Thi Anh-Dao Tran (2016) ‘Sub-Saharan Africa’s
Growth, South–South Trade and the Generalised Balance-of-payments Constraint’, Cam
Page 25 of 36
Baldwin, Richard (2013) ‘Trade and Industrialization after Globalization’s Second Un
bundling: How Building and Joining a Supply Chain are Different and Why it Matters’, in
Robert C. Feenstra and Alan M. Taylor (eds) Globalization in an Age of Crisis: Multilateral
Economic Cooperation in the Twenty-first Century. Chicago, IL: University of Chicago
Press, pp. 165–212.
Barros de Castro, Lavínia (2009) ‘Financiamento do desenvolvimento: experiência com

parada Brasil-Coreia (1950–1980)’. Available at https://web.bndes.gov.br/bib/jspui/
handle/1408/908.
Brainard, Lael and Leonardo Martinez-Diaz (eds) (2009) Brazil as an Economic Superpow
er? Understanding Brazil’s Changing Role in the Global Economy. Washington, DC:
Brookings Institution Press.
Bruschi, Claudia, Sergio G. Lazzarini, and Aldo Musacchio (2013) ‘Análise do retorno dos
investimentos do BNDESPAR por meio de variações nos preços das ações investidas’.
INSPER Research Report, São Paolo.
Bulmer-Thomas, Victor (1994) The Economic History of Latin America since Indepen
dence. Cambridge: Cambridge University Press.
Burlamaqui, Leonardo (2012) ‘Global Finance and Chinese Financial Development: The
Emergence of a State-led Financial Model of Globally Oriented Financial Governance’.
IPEA Working Paper, Brasília. (p. 365)
Calomiris, Charles W. and Stephen Haber (2014) Fragile by Design: Banking Crises,
Scarce Credit, and Political Bargains. Princeton, NJ: Princeton University Press.
Cameron, Rondo E. (1953) ‘The Credit Mobilier and the Economic Development of Eu
rope’, Journal of Political Economy 61: 461.
Colby, Seth (2012) ‘Explaining the BNDES: What It Is, What It Does and How It Works’.
CEBRI Working Paper Artigos Vol. 3, Ano VII, Rio de Janeiro.
Colby, Seth (2013) ‘Searching for Institutional Solutions to Industrial Policy Challenges: A
Case Study of the Brazilian Development Bank’. PhD dissertation, Johns Hopkins Univer
sity.
De Aghion, Beatriz Armendariz (1999) ‘Development Banking’, Journal of Development

Economics 58(1): 83–100.
De Luna-Martinez, José and Carlos C. Vicente (2012) ‘Global Survey of Development

Banks’. Policy Research Working Paper No. 5969. Washington, DC: World Bank.
De Luna-Martinez, José, Carlos C. Vicente, Ashraf bin Arshad, Radu Tatucu, and Jiyoung
Song (2017) Survey of National Development Banks. Washington, DC: World Bank.
Page 26 of 36
De Magalhães, Luis C. and Carla Rodrigues Costa (2018) ‘Arranjos institucionais, custo da
dívida pública e equilíbrio fiscal: a despesa “ausente” e os limites do ajuste estrutural’.
IPEA Discussion Paper 2013, Brasilia.
Dewatripont, Mathias and Eric Maskin (1995) ‘Credit and Efficiency in Centralized and
Decentralized Economies’, The Review of Economic Studies 62(4): 541–55.
Elstrodt, Heinz-Peter, James Manyika, Jaana Remes, Patricia Ellen, and César Martins
(2014) ‘Connecting Brazil to the World: A Path to Inclusive Growth’. McKinsey Global In
stitute, May.
Ferraz, João C. and Luciano Coutinho (2019) ‘Investment Policies, Development Finance
and Economic Transformation: Lessons from BNDES’, Structural Change and Economic
Dynamics 48: 86–102.
Figueiredo, Paulo N. (2014) ‘Technological Catch-up and Indigenous Institutional Infras

tructures in Latecomer Natural Resource-related Industries: An Exploration of the Role of
EMBRAPA in Brazil’s Soybeans and Forestry-based Pulp and Paper Industries’. IRIBA
Working Paper No. 3. London: Department for International Development (DFID).
Fishlow, Albert (1989) ‘Latin American Failure against the Backdrop of Asian Success’,
The Annals of the American Academy of Political and Social Science 505(1): 117–28.
Frieden, Jeffry A. (1991) Debt, Development, and Democracy: Modern Political Economy
and Latin America, 1965–1985. Princeton, NJ: Princeton University Press.
Gallagher, Kevin P. and Sarah Sklar (2016) ‘Nationalizing Development Finance: Putting
National Development Banks in Context’. Boston University Global Economic Governance
Initiative Working Paper. (Cited in Rogerio Studart and Kevin P. Gallagher (2017) ‘Infras
tructure for Sustainable Development: The Role of National Development Banks’. GEGI
Policy Brief No. 7. Global Economic Governance Initiative, Boston University.)

Griffith-Jones, Stephany and Ricardo Gottschalk (2012) ‘Exogenous Shocks—Dealing with

the Only Certainty: Uncertainty’. Paper presented at Commonwealth Secretariat Finance
Ministers Meeting Tokyo, 10 October.
Griffith-Jones, Stephany and José Antonio Ocampo (2008) ‘A Counter-cyclical Framework

for a Development-friendly IFA’, in José Maria Fanelli (ed.) Macroeconomic Volatility, Insti
tutions, and Financial Architectures: The Developing World Experience. New York: Pal
grave Macmillan, pp. 25–44. (p. 366)
Griffith-Jones, Stephany and José Antonio Ocampo (eds) (2018) The Future of National
Development Banks. Oxford: Oxford University Press.
Page 27 of 36
Griffith-Jones, Stephany, José Antonio Ocampo, Felipe C. Rezende, Alfredo Schclarek, and
Michael Brei (2018) ‘The Future of National Development Banks’, in Stephany Griffith-
Jones and José Antonio Ocampo (eds) The Future of National Development Banks. Oxford:
Hamilton, Alexander (1781) Report on Manufactures.
Hausmann, Ricardo (2018) ‘Development Collective Know-how and Us’. Lecture at Lon
don School of Economics, January. Available at http://www.lse.ac.uk/Events/Events-As
sets/PDF/2018/01-LT/20180129-Ricardo-Hausmann.pdf.
Hayek, Frederick A. (1945) ‘The Use of Knowledge in Society’, American Economic Re
view 35(4): 519–30.
Hochstetler, Kathryn, and Alfred P. Montero (2013) ‘The Renewed Developmental State:
The National Development Bank and the Brazil model’, Journal of Development Studies
49(11): 1484–99.
Kapadia, Anush (2016) ‘The Structure of State Borrowing: Towards a Political Theory of
Control Mechanisms’, Cambridge Journal of Regions, Economy and Society 10(1): 189–
204.
Khanna, Tarun and Yishay Yafeh (2007) ‘Business Groups in Emerging Markets: Paragons
or Parasites?’, Journal of Economic Literature 45(2): 331–72.
Kindleberger, Charles P. and Robert Z. Aliber (2011) Manias, Panics and Crashes: A Histo
ry of Financial Crises. Basingstoke: Palgrave Macmillan.
Kingstone, Peter (1999) Crafting Coalitions for Reform: Business Preferences, Political In
stitutions, and Neoliberal Reform in Brazil. University Park, PA: Pennsylvania State Uni
versity Press.
Kornai, Janos (1979) Economics of Shortage. Amsterdam: North Holland.
Krieckhaus, Jonathan (2002) ‘Reconceptualizing the Developmental State: Public Savings

and Economic Growth’, World Development 30(10): 1697–712.
Krugman, Paul (1987) ‘Is Free Trade Passé?’, Journal of Economic Perspectives 1(2): 131–
44.
Leahy, Joe and Andres Schipani (2017) ‘Culture of Corruption Engulfs Brazilian Elite’, Fi
nancial Times, 25 May.
Leff, Nathaniel H. (1978) ‘Industrial Organization and Entrepreneurship in the Develop

ing Countries: The Economic Groups’, Economic Development and Cultural Change 26(4):
661–75.
Page 28 of 36
Lundstøl, Olav, Gaël Raballand, and Fuvya Nyirongo (2013) ‘Low Government Revenue
from the Mining Sector in Zambia and Tanzania: Fiscal Design, Technical Capacity or Po
litical Will?’. ICTD Working Paper No. 9.
Lundstøl, Olav, Gaël Raballand, and Fuvya Nyirongo (2018) ‘Revenue Sharing in Mining
in Africa: Empirical Proxies and Determinants of Government Take’. ICTD Working Paper
No. 81.
Mahmood, Ishtiaq P. and Carlos Rufin (2005) ‘Government’s Dilemma: The Role of Gov
ernment in Imitation and Innovation’, Academy of Management Review 30(2): 338–60.
Mainwaring, Scott (1999) Rethinking Party Systems in the Third Wave of Democratiza
tion: The Case of Brazil. Stanford, CA: Stanford University Press.
Malan, Pedro and Regis Bonelli (1992) ‘The Success of Growth Policies in Brazil’, in Si
mon Teitel (ed.) Towards a New Development Strategy for Latin America: Pathways from
Hirschman’s Thought. Washington, DC: IDB/Johns Hopkins University Press, pp. 47–101.
Maskin, Eric S. (1996) ‘Theories of the Soft Budget-constraint’, Japan and the World Econ
omy 8(2): 125–33. (p. 367)
Massi, Eliza (2015) ‘The Political Economy of Development Finance: The BNDES and
Brazilian Industrialisation’. PhD dissertation, SOAS, University of London.
tor Myths, Vol. 1. London: Anthem Press.
Mazzucato, Mariana and L. Randall Wray (2015) ‘Financing the Capital Development of
the Economy: A Keynes–Schumpeter–Minsky Synthesis’. Levy Economics Institute of Bard
College Working Paper No. 837.
Mazzucato, Mariana and Caetano C. Penna (2016) ‘Beyond Market Failures: The Market
Creating and Shaping Roles of State Investment Banks’, Journal of Economic Policy Re
form 19(4): 305–26.
Mazzucato, Mariana and Caetano C. Penna (2018) ‘National Development Banks and Mis
sion-Oriented Finance for Innovation’, in Stephany Griffith-Jones and José Antonio Ocam
po (eds) The Future of National Development Banks. Oxford: Oxford University Press, pp.
253–77.
McKinnon, Ronald and Edward Shaw (1973) ‘Financial Deepening in Economic Develop
ment’. Washington, DC: Brookings Institution.
Mokyr, Joel (1994) ‘Technological Change, 1700–1830’, in Roderick Floud and Deirdre
McCloskey (eds) The Economic History of Britain since 1700, Vol. 1: 1700–1860. Cam
bridge: Cambridge University Press, pp. 12–43.
Page 29 of 36
Moslener, Ulf, Matthias Thiemann, and Peter Volberding (2018) ‘National Development
Banks as Active Financiers: The Case of KfW’, in Stephany Griffith-Jones and José Antonio
Ocampo (eds) The Future of National Development Banks. Oxford: Oxford University
Press, pp. 63–85.
Musacchio, Aldo and Sergio Lazzarini (2014a) Reinventing State Capitalism. Cambridge,
MA: Harvard University Press.
Musacchio, Aldo and Sergio Lazzarini (2014b) Workshop on State-Owned Enterprises in

the Development Process: The Case of Brazil. Paris: OECD.
Musacchio, Aldo, Sergio Lazzarini, Pedro Makhoul, and Emily Simmons (2017) ‘The Role
and Impact of Development Banks’. Mimeo. Available at http://people.brandeis.edu/
~aldom/papers/
The%20Role%20and%20Impact%20of%20Development%20Banks%20-
%203-9-2017.pdf.
Naqvi, Natalya, Anne Henow, and Ha-Joon Chang (2018) ‘Kicking away the Financial Lad
der? German Development Banking under Economic Globalisation’, Review of Interna
tional Political Economy 25(5): 672–98.
Nassar, André M. (2009) ‘Brazil as an Agricultural and Agroenergy Superpower’, in Lael

Brainard and Leonardo Martinez-Diaz (eds) Brazil as an Economic Superpower? Under
standing Brazil’s Changing Role in the Global Economy. Washington, DC: Brookings Insti
tution Press.
North, Douglass (1990) Institutions, Institutional Change and Economic Performance.

Orair, Rodrigo O. and Sergio W. Gobetti (2017) ‘Brazilian Fiscal Policy in Perspective:
From Expansion to Austerity’. Working Paper No. 160. International Policy Centre for In
clusive Growth, UNDP and Government of Brazil.
Perez-Aleman, Paola and Flavia Chaves Alves (2016) ‘Reinventing Industrial Policy at the
Frontier: Catalysing Learning and Innovation in Brazil’, Cambridge Journal of Regions,
Economy and Society 10(1): 151–71.
Qian, Yingyi and Barry R. Weingast (1997) ‘Federalism as a Commitment to Reserving

Market Incentives’, Journal of Economic Perspectives 11(4): 83–92. (p. 368)
Rodrik, Dani (2004) ‘Rethinking Growth Strategies’, WIDER Annual Lecture 8. Helsinki:
UN-WIDER.
on Economic Activity 2008(2): 365–412.
Rodrik, Dani (2016) ‘An African Growth Miracle?’, Journal of African Economies 27(1): 10–
27.
Page 30 of 36
Saad-Filho, Alfredo and Lecio Morais (2014) ‘Mass Protests: Brazilian Spring or Brazilian
Malaise?’, Socialist Register 50: 227–46.
Sayers, Richard S. (1957) Central Banking after Bagehot. Oxford: Oxford University
Press.
Stiglitz, Joseph E. and Marilou Uy (1996) ‘Financial Markets, Public Policy, and the East
Asian Miracle’, The World Bank Research Observer 11(2): 249–76.
Stiglitz, Joseph E. and Andrew Weiss (1981) ‘Credit Rationing in Markets with Imperfect
Information’, American Economic Review 71(3): 393–410.
Storm, Servaas (2015) ‘Structural Change’, Development and Change 46(4): 666–99.
Studwell, Joe (2013) How Asia Works: Success and Failure in the World’s Most Dynamic
Region. London: Open Road+ Grove/Atlantic.
Tavares de Araujo Jr, José (2013) ‘The BNDES as an Instrument of Long-run Economic
Policy in Brazil’. CINDES Working Paper, Rio de Janeiro.
The Economist (2012a) ‘Economic Policy in Brazil: Sparking Recovery’, 15 September.
The Economist (2012b) ‘A Special Report on State Capitalism’, 21 January.
The Economist (2013) ‘Grounded: Special Report on Brazil’, 28 September.
Thirlwall, Anthony P. (2011) ‘Balance-of-payments-constrained Growth Models: History

and Overview’, PSL Quarterly Review 64(259): 307–51.
Thirlwall, Anthony P. (2019) ‘Thoughts on Balance-of-payments-constrained Growth after

40 Years’, Review of Keynesian Economics 7(4): 554–67.
Torres, Ernani and Rodrigo Zeidan (2016) ‘The Life-cycle of National Development Banks:
The Experience of Brazil’s BNDES’, The Quarterly Review of Economics and Finance 62:
97–104.
Trebat, Thomas J. (1983) Brazil’s State-owned Enterprises: A Case Study of the State as
Entrepreneur. Cambridge: Cambridge University Press.
United Nations Centre for Trade and Development (UNCTAD) (2016) The Role of Devel
opment Banks in Promoting Growth and Sustainable Development in the South. Geneva:
UNCTAD.
Woetzel, Jonathan, Anu Madgavkar, Jeongmin Seong, James Manyika, Kevin Sneader, Oliv
er Tonby, Andrés Cadena, Rajat Gupta, Acha Leke, Hayoung Kim, and Shishir Gupta
(2018) ‘Outperformers: High-growth Emerging Economies and Companies that Propel
Them’, McKinsey Global Institute (MGI).
Page 31 of 36
Woo, Jung-en (1991) Race to the Swift: State and Finance in Korean Industrialization.
New York: Columbia University Press.
Woo-Cummings, Meredith (ed.) (1999) The Developmental State. New York: Cornell Uni
versity Press.
Xu, Jiajun, Xiaomeng Ren, and Xinyue Wu (2019) ‘Mapping Development Finance Institu
tions Worldwide: Definitions, Rationales, and Varieties’, Institute of New Structural Eco
nomics. Peking University.
Xu, Qiyuan (2018) ‘China Development Bank: Born Bankrupt, Born Shaper’, in Stephany
Griffith-Jones and José Antonio Ocampo (eds) The Future of National Development Banks.
Oxford: Oxford University Press, pp. 36–62.
Notes:
(1) A World Bank survey defines a development bank as ‘a bank or financial institution
with at least 30 per cent state-owned equity that has been given an explicit legal mandate
to reach socio-economic goals in a region, sector, or particular market segment’ (De Lu
na-Martinez and Vicente, 2012: 4). A definition that does not require state ownership to
be a necessary condition defines ‘development finance institutions’ (DFIs) as ‘legally inde
pendent and government-supported financial institutions with explicit official missions to
promote public policy objectives’ (Xu et al., 2019: 14‒19). See also Xu et al. (2019: 64, ta
ble A4) for competing definitions of NDBs.
(2) According to Gallagher and Sklar (2016), the level of total assets of national develop
ment reached approximately US$5 trillion in 2015, which is far larger than the level of
loans of the multilateral development banks (around US$1 trillion in the same year).
(3) KfW is ‘Credit Institute for Reconstruction’.
(4) By 2015, the five largest development banks in terms of assets (in billions of US dol
lars) were: (i) China Development Bank ($1,665); (ii) European Investment Bank ($624);
(iii) German Development Bank (Kfw) ($539); (iv) World Bank ($276); (v) Brazilian Devel
opment Bank (BNDES) ($238) (Ferraz and Coutinho, 2019: 910, table 1).
(5) Griffith-Jones et al. (2018) identify five crucial roles for NDBs: (i) counteracting the
procyclical behaviour of private financing; (ii) promoting innovation and structural trans
formation; (iii) enhancing financial inclusion; (iv) supporting the financing of infrastruc
ture investment; and (v) supporting the provision of public goods, including promoting
‘green growth’.
(6) Indeed, from advances such as the Internet and microchips to biotechnology and nan
otechnology, many major technological breakthroughs—in both basic research and down
stream commercialization—were only made possible by direct public investment willing
and able to take risks before the private sector was willing to (Mazzucato, 2015).
Page 32 of 36
(7) The soft budget constraint (1979) holds wherever a funding source (such as a bank or
government) is unable to keep an enterprise to a fixed budget, that is, whenever the en
terprise can extract ex post a bigger subsidy or loan than would have been considered ef
ficient ex ante (Maskin, 1996: 25). Kornai (1979) analysed how the centralized, socialist
economies of Eastern Europe and the Soviet Union were plagued by soft-budget con
straints. The term has been applied to explain problems of SOEs in capitalist economies
and used as a justification for advocating privatization.
(8) De Luna-Martinez and Vicente (2012) provide evidence that these banks increased
their lending from US$1.16 trillion to US$1.58 trillion between 2007 and 2009.
(9) Growth reversals or collapses in developing economies have frequently been preceded
by foreign exchange shortage, marked by urgent pressure to seek increased aid flows to
finance immediate imported input requirements and the rising level of external debt
(Arizala et al., 2017; Bagnai et al., 2016).
(10) The East Asian economies largely achieved high savings rates through the coercive
power of the state, which was deployed through various forms of ‘forced savings’. Among
the coercive elements were restrictions on consumer credit, financial restraint, mandato
ry provident pension contributions (used in Singapore and Malaysia), and encouragement
of postal savings (Stiglitz and Uy, 1996). Other factors that contributed to high levels of
private savings were banking and industrial policies (including entry restrictions) that
guaranteed high levels of profitability, capital accumulation, and income growth (Stiglitz
and Uy, 1996; Storm, 2015).
(11) According to a World Bank survey (De Luna-Martinez and Vicente, 2012: 4), almost
three-quarters of NDBs surveyed are 100 per cent state owned, 21 per cent have between
50 and 90 per cent state ownership, and in only 5 per cent do governments have minority
ownership.
(12) On the growth of SOEs in this period, see Musacchio and Lazzarini (2014b). The ex
pansion of SOEs meant that the SOE share of total gross fixed capital formation in
creased from 13 per cent in 1965 to 29 per cent in 1979 (Frieden, 1991: 107).
(13) It is well documented that BNDES played a dominant role in deploying subsidized
long-run credit in this period (see Barros de Castro, 2009; Tavares de Araujo, Jr., 2013;
Colby, 2013). In the period 1952 to 1961, between 80 and 90 per cent of all lending went
to the public sector. In 1962 to 1971, the private sector received an average of 66 per
cent of all lending, rising to an average of approximately 75 per cent of all loans going to
the private sector in the period 1972 to 1977 (Colby, 2012: 164), mostly in intermediate
and capital goods.
(14) In 2007–14, loan disbursements increased from R$96 billion to R$188 billion at con
stant prices, a growth of 96 per cent in real terms (UNCTAD, 2016: 12).
Page 33 of 36
(15) In the latter case, and in particular the case of Petrobras, the state has anchored
much of its industrial policy around developing linkages from large offshore oil finds into
shipbuilding, construction, and deep-sea oil exploration equipment. During this period,
Petrobras set in motion a US$237 billion investment programme for these projects.
(16) The value of loans disbursed by BNDES in 2010 was more than three times the total
amount provided by the World Bank in 2010 (Musacchio and Lazzarini, 2014b: 30) and
still twice that amount in 2014 (Torres and Zeidan, 2016: 101). Between 2009 and 2016,
subsidies from the treasury to BNDES loans totalled US$48 billion (The Economist, 2 De
cember 2017).
(17) Interview with João Carlos Ferraz, vice-president of BNDES (August 2011).
(18) In historical perspective, there are several features that unite the stories behind the
emergence of these innovative firms. First, all of them have their origins as domestic
business enterprises. Second, most of these firms come from sectors that have built up
substantial productive and managerial capabilities over long periods of time, going back
in some cases to the early twentieth century, and have benefited from long periods of pro
tectionism (whether as a result of explicit state strategies or as unintended conse
quences). Third, all these firms have benefited from state policies, whether in the form of
access to long-run, subsidized financing, protectionist policies, state-financed research
and development, or public procurement. Fourth, exposure to international markets, and
particularly growing through exports, even during the protectionist era, allowed these
firms to gather information on the demands of foreign consumers and compelled them to
seek out productivity-enhancing innovations to compete in world markets. Fifth, the mili
tary played a key role from the Vargas era in the 1930s through the 1970s in creating a
discourse around the development of ‘strategic’ sectors, beginning with the steel indus
try, but also in oil and gas, and aerospace. This discourse not only helped governance fo
cal points emerge around such sectors but also provided continuity of funding even in
times of macroeconomic crisis and distress as well as providing research clusters and
long-run financing facilities and using SOEs as focal points to train and attract scientists
and engineers. This is most evident in the case of Embraer (the world’s third-largest aero
space company).
(19) By 2009, agricultural production in Brazil ranked first in the world in coffee, orange
juice, and beef; it ranked second in soybeans; third in chickens; fourth in corn and pork;
and fifth in cotton (Nassar, 2009). By 2012, the yields for Brazil’s main crops (sugarcane,
corn, cotton, soybeans, and wheat) reached levels similar to those of developed
economies (Elstrodt et al., 2014: 49‒50;) agricultural productivity growth has been
among the fastest in the world since 2000 (Arias et al., 2017: 4‒8). Brazil is also a world
leader in mining (e.g. Vale is the world’s second-largest mining company and Petrobras is
a leading oil company and a producer of sophisticated deep-sea oil exploration equip
ment) (Brainard and Martinez-Diaz, 2009).
Page 34 of 36
(20) Manufacturing growth in the period 1950 to 1980 averaged nearly 8 per cent per
year but averaged around 2 per cent per year in the post-1990 period. According to
Aldrighi and Colistete (2013), while average annual manufacturing labour productivity
grew at 5.8 per cent in the period 1945 to 1980, it declined to 2.1 per cent in the period
1980 to 1990 and then collapsed to minus 0.5 per cent in the period 1995 to 2009.
(21) Despite its rapid industrial catch-up, Brazil’s share in world market exports was to
decline from 2.9 per cent in 1950 to 1.1 per cent in 1960, and would remain at that share
through 1970 (Bulmer-Thomas, 1994: 271).
(22) The ‘big push’ strategy of PNDII development plan did not generate the net foreign
exchange to finance debt payments in the 1980s, something East Asian economies were
able to accomplish. The country’s foreign debt (in constant 1982 dollars) grew from
US$9.2 billion in 1967 to US$27.8 billion in 1973 and further increased to US$43.3 bil
lion in 1979, which represented two and a half times the value of the country’s exports
(Malan and Bonelli, 1992: 85).
(23) In the new model, the guarantee of government credibility became a requirement of
the international market, and the liberalization that saw the opening up of the economy
commercially and financially served, together with privatization, to guarantee the Brazil
ian government a supply of liquidity (Afonso et al., 2016).
(24) There are numerous examples (The Economist, 2013: 4). In 2012, ‘cost of doing busi
ness’ surveys indicate that exporting costs US$2,215 per container in Brazil compared
with an average of US$1,197 in the rest of Latin America, and US$1,029 in the United
States. To take another example, the cost of electricity for industrial users in Brazil is the
third highest in the world—about twice those of China and Korea, and two and a half
times that of the United States (The Economist, 2012a).
(25) The real wage (deflated by domestic prices), while rising slightly in the period 1995
to 1998, has generally stagnated in the period 1998 to 2009 (Aldrighi and Colistete, 2013:
37, figure 11).
(26) In the period 2008 to 2010, about US$16 billion was channelled to the food industry,
and US$30 billion to Petrobras (Tavares de Araujo, Jr., 2013: 11). By 2009 Petrobras be
came by far the largest borrower, with almost 40 per cent of total loans held by listed cor
porations (Musacchio and Lazzarini, 2014b: 31‒2). From 2007 to 2016, BNDES subsi
dized loans increased five-fold, reaching R$867 billion (US$262 billion) (Leahy and
Schipani, 2017). The main beneficiaries included Odebrecht (the country’s largest engi
neering and construction firm); Petrobras (the state oil company which accounts for 10
per cent of GDP); Embraer (the world’s third-largest commercial jet builder); Ambev (the
Brazilian arm of the world’s largest brewer, ABInBev); and JBS (the world’s largest meat
packer) (Leahy and Schipani, 2017).
(27) As Woo-Cummings (1999: 10), notes: ‘finance is the tie that binds the state to the in
dustrialists in the developmental state.’ Theda Skocpol (as quoted in Woo-Cummings,
Page 35 of 36
1999: 11) further adds that ‘[t]he answers [to questions about financial resources] pro
vide the best possible general insight into the direct or indirect leverage a state is likely
to have for realizing any sort of goal it may pursue. For a state’s means of raising and de
ploying financial resources tell us more than could any other single factor about its exist
ing (and its immediately potential) capacities to create or strengthen state organizations,
to employ personnel, to co-opt political support, to subsidize economic enterprises, and to
fund social programs.’
Jonathan Di John
Jonathan Di John is a senior lecturer in the political economy of development at

SOAS, University of London. His main areas of expertise are development economics,
institutional economics, and the political economy of growth and development in
Latin America, especially in Brazil and Venezuela. His research focuses on the politi
cal economy of industrial strategy, taxation, development strategies in oil economies,
and social service delivery in violent contexts. He has undertaken consultancy work
for the World Bank, United Nations Development Programme (UNDP), the UK De
partment for International Development (DFID), the Organisation of Economic Coop
eration and Development (OECD) and the International Monetary Fund (IMF).
Page 36 of 36
Technical Change, the Shifting ‘Terrain of the Industrial’, and Digital In
dustrial Policy
Technical Change, the Shifting ‘Terrain of the Industri

al’, and Digital Industrial Policy
Antonio Andreoni
Technical change is a major driver of structural transformation and industrial mutations

within and across sectors of the economy. We show how, by deploying different concepts
of sector—commodity/product, production/technology, or location-based taxonomies—we
can better capture the heterogeneity of production activities, shifting sectoral bound
aries, industrial mutations, sources of technical change, and non-linear patterns of struc
tural change. These are important dimensions for industrial policy targeting. We analyse
these technological dynamics with an industrial ecosystem framework structured around
several sectoral value chains underpinned by different technology platforms. Within this
framework, we highlight the role of digital technologies alongside other key enabling
technologies and discuss technological change trajectories and cross-sectoral diversifica
tion patterns. Against this background, we discuss the specific challenges of deploying
digital technologies effectively faced by developing countries. To address these chal
lenges and capture windows of digital opportunity, industrial policy must be articulated
along both sectoral and cross-sectoral mission-oriented strategies.
Keywords: technical change, production, sectors, heterogeneity, industrial mutations, industrial ecosystem, digi
talization, developing countries, industrial policy
14.1 Introduction
INDUSTRIALIZATION is a complex process involving interdependent changes across mul
tiple domains and structures of the economy. Since the First Industrial Revolution, techni
cal change has been increasingly recognized as one of the most fundamental drivers of
these structural transformations and one of the most distinctive features of modern eco
nomic growth (Schumpeter, 1939; Kuznets, 1973; Rosenberg, 1976; Lazonick, 1990;
Abramovitz, 1989). Technical change makes it possible to improve products, diversify
them, and even introduce completely new ones. It is also responsible for increasing pro
ductivity within sectors and, in turn, for changes in the way in which enterprises organize
production processes and engage with other producers along the value chain—that is, in
Page 1 of 27
dustrial Policy
dustry organization (Richardson, 1972). Technical change is thus intertwined with several
innovation processes and can lead to the transformation of the ‘organizational
economy’ (Simon, 1991) and the creation of new markets (Schumpeter, 1942). Over the
long run, technical change and related innovation finally trigger broader changes in the
fabric of society, its institutional and political configuration, and the distribution of power
among organizations and even nations (Hirschman, 1958; Dosi, 1982; Amsden, 1991;
Chang, 2002; Pérez, 2002; Reinert, 2008; Andreoni and Chang, 2017, 2019).
In some cases, technical change can be so transformative—that is, it can change

(p. 370)
products and production so much—that the nature and boundary of traditionally defined
economic sectors also change. For example, what used to be called an agricultural prod
uct or sector becomes more like a manufacturing product or industry, and a manufactur
ing industry can become more like a service one (and vice versa). These sectoral shifts
make it difficult to draw the boundaries of the traditionally defined ‘sectoral terrains’ of
agriculture, industry, and services (and sub-sectors within them). From a dynamic point of
view, these sectoral shifts also challenge a simplistic ‘linear’ idea of structural change ac
cording to which countries develop by moving from the primary to the secondary sector
and from the secondary to the tertiary sector.
Pre-classical (Antonio Serra, Giovanni Botero, and Pietro Verri) and classical political
economists (Adam Smith, Karl Marx, and David Ricardo) relied on the concept of division
of labour and rents to account for technical change and scarce resources in economic de
velopment (Scazzieri, 1993; Andreoni and Scazzieri, 2014; Reinert, Chapter 19, this vol
ume). Within the classical economic framework, these dynamics were linked to and
framed within different sectors of the economy—that is, primary and secondary. The idea
of sector was also critical among development economists of the twentieth century in the
study of structural change (for a review including circular and cumulative causation theo
ries, see Toner, 1999) and in accounting for the input‒output relationships (and linkages)
constituting the production matrix of an economy (Leontief, 1941; Hirschman, 1977). Sev
eral other scholars (Kaldor, 1967 in particular; see also Andreoni and Chang, 2017) have
used the same concept of sector to highlight the special properties of manufacturing,
with respect to technical change and productivity dynamics. More recently, with the re
vival in structural economic analysis (Lin, 2010; Rodrik et al., 2017), there has been an in
creasing focus on sectoral productivity and countries’ normal patterns of structural
change, and the ways in which these analyses can inform ‘new’ industrial policy thinking
(see Andreoni and Chang, 2019 for a critical review).
The conventional way of defining a ‘sector’ is by reference to a class of goods to be con

sumed and traded in the market (see Weiss, Chapter 5, this volume, for a discussion on in
dustrial policy from a neoclassical perspective). This standard concept of sector is useful
in many respects: for example, it helps in capturing sectoral and sub-sectoral differences;
linking a cluster of production activities to specific demand dynamics and markets; and fi
nally, from a policy point of view, identifying and coordinating sets of organizations and
institutions around specific sectoral industrial strategies.
Page 2 of 27
dustrial Policy
While widely used in empirical and policy work, the concept of sector has, however, be
come increasingly problematic in analysing technical change and informing industrial pol
icymaking in both advanced economies and among late industrializers. This is particular
ly the case as technical change has been increasingly shifting the ‘terrain of the industri
al’ and, in doing so, has made conventional approaches unable to capture important sec
toral dynamics and ‘industrial mutations’ (Schumpeter, 1942). These dynamics and muta
tions include changes in the innovation mode and productivity scope of each sector, sec
toral boundaries, and structural change dynamics.
In this chapter we address these limitations in three main steps. First, we review
(p. 371)
different concepts of sector and assess their usefulness (and limitations) with respect to
different analytical questions and policy objectives. We present five reasons why a stan
dard concept of sector alone is problematic for understanding technical change and the
shifting terrain of the industrial. Next, we move to the analysis of how technical change is
driving the shifting terrain of the industrial. Drawing on Andreoni (2018), we advance an
‘industrial ecosystem framework’ made up of several sectoral value chains underpinned
by different technology platforms. Within this framework, we highlight the role of digital
technologies alongside other key enabling technologies and discuss technological change
trajectories and patterns of sectoral diversification. Example cases representing several
forms of ‘industrial mutations’ are introduced, that is, cases in which technical changes
have redesigned the boundaries and technical content of traditionally defined sectors. We
examine these developments, identifying both sector-specific and industrial ecosystem
targets for industrial policies, and in particular focus on the new ‘digital terrain’ emerg
ing from digital technical change. We conclude by arguing that digitalization makes the
rethinking of sectors particularly critical and that industrial policies are becoming even
more important in filling the digital capability gap across developing countries.
14.2 Technical Change and the Concepts of

Sector: ‘The What’, ‘The How’, and ‘The Where’
Over the last century, since the provocative work of Clapham (1922) denouncing the use
of the ‘empty economic boxes’ in the analysis of sectors, economists have widely debated
different conceptualizations of sector, in particular, what accounts for the ‘terrain of the
industrial’, that is, the industrial sector. The terms of this debate were set by classical po
litical economists in the nineteenth century. Their definition of ‘primary’ and ‘secondary’
sectors stemmed from a set of holistic considerations, including different human needs
(primary and secondary); different social classes; the materiality (vis-à-vis intangible na
ture) of produce; and specific technical properties of production, that is, its return dy
namics. Specifically, division of labour as a form of technical change was associated with
increasing returns, while rents and non-reproduceable scarce resources were associated
with decreasing returns.
Page 3 of 27
dustrial Policy
Building on these classical analytical categories, several contributions have advanced dif
ferent conceptualizations of sector, often in relation to technical properties of production
or produce, and their changes. Indeed, in his seminal work, Piero Sraffa (1926) chal
lenged the foundations of supply curve economics by unpacking the unfolding of increas
ing and diminishing returns in different sectors of the economy (p. 372) (for a develop
ment, see Andreoni and Scazzieri, 2014). In doing so, Sraffa had to confront the problem
of defining a sector, its boundaries and its relationship with other sectors. Sraffa recog
nized two potentially alternative concepts of sector: one according to which sectors are
defined by ‘all the undertakings which employ a given factor of production, as, for exam
ple, agriculture or the iron industry’; and another more restrictive concept of sector in
cluding ‘only those undertakings which produce a given type of consumable commodity
as, for example, fruit or nail’ (Sraffa, 1926: 538).
While the latter commodity/product-based concept of sector has acquired significant trac
tion among economists since the 1950s, as observed by Becattini (1962), it does shift the
problem of defining the sector to one of defining the commodity/product. Commodities
can be conceptualized as a pure ‘means of exchange’ or as a ‘means of satisfying specific
human needs’. The first definition was adopted in the study of market competition and de
veloped along the idea of different degrees of commodity/product differentiation (Robin
son, 1933). The second, commodity/product-based definition of sector sees it as a collec
tion of goods that satisfy the same function or class of human needs (Lancaster, 1966).
Following Lancaster’s idea that consumers do not consume goods but aggregates of char
acteristics, we could claim that a sector is not made up of enterprises producing a certain
type of good, but rather of a group of enterprises which produce a certain ‘characteris
tic’, for example, mobility, communication. The challenges in such a conceptualization of
sectors are thus shifted from the level of market exchange to the level of the structure of
needs and means at a certain historical moment. Interestingly, the concept of ‘tertiary
production’ was first advanced by Fisher (1939) to capture those industries (beyond
transport and commerce) directly satisfying consumer wants. By using the consumer’s
perspective, Fisher identified and ranked sectors in a sort of descending order of con
sumption urgency—from primary to secondary and tertiary.
From a more empirical perspective, by adopting a relatively simple and common-sensical

commodity/product taxonomy and classes, economists have been able to link sectoral-
commodity/product dynamics to specific properties of demand/markets, for example dif
ferences in price elasticity. Others have also attempted to develop taxonomies of products
based on their technological qualities and content. For example, building on the classifi
cation proposed by the OECD (1994) and other indicators of technological activity and
rankings in manufacturing sectors, Lall (1998) developed a technological classification of
tradable products distinguishing ‘resource-based manufactures’, ‘low-technology manu
factures’, ‘medium-technology manufactures’ and ‘high-technology manufactures’. Each
of these product clusters were also associated with traditionally defined economic sectors
in an empirical analysis of industrialization and structural change. However, while Lall’s
widely used classification is aimed at distinguishing different products (and associated
Page 4 of 27
dustrial Policy
sectors), their ‘technological ranking’ is intuitively associated with both technical proper
ties of commodities and the technical properties of the production processes underpin
ning them (e.g. the technology or R&D intensity of these products/sectors).
The alternative conceptualization of sectors is indeed based on ‘factors of produc

(p. 373)
tion’ more than commodities and their properties, as already pointed out by Sraffa
(1926). This production-based conceptualization of sectors found some developments in
the 1950s and 1960s, especially among applied economists. For example, Wolfe (1955:
406) proposed to ‘define the primary sector as that group of industries in which an in
crease in productivity, in terms of goods and services per man-hour, is limited by natural
growth factors, the secondary by mechanical factors, and the tertiary by relatively unaid
ed human skills’. Scholars using input‒output analysis similarly recognized a sector as
defined by firms using the same production function.
Building on the Marshallian definition of a sector proposed in Industry and Trade (1920),
other scholars defined a sector as the set of firms which are capable—at least potentially
—of performing each other’s production processes within certain parameters of efficien
cy. This definition of sector recognizes how the boundaries of the sector are not necessar
ily set but can change depending on the number of firms that have the incentives and ca
pabilities to engage in certain production activities. A company might be potentially in
multiple sectors, even if it is operating in only one commodity/product-defined sector. In
deed, as we shall see, diversification and lateral migration across sectors are typical dy
namics of the growth of the firm (Penrose, 1959; Lazonick, 1991; Best, 1990; Andreoni,
2018; Chapter 6, this volume).
Within this family of production-based (more than commodity/product-based) definitions

of sector, other scholars have advanced a conceptualization of sector based on their ‘tech
nical progress’ and distinctive ‘patterns of innovation’. With respect to the first one, for
example, Fourastie (1951: 124) proposed calling ‘primary and secondary, activities that
have enjoyed great technical progress, reserving the term primary for agricultural activi
ties and the term secondary for nonagricultural activities’. According to him, tertiary ac
tivities are those which have enjoyed ‘only slight technical progress’. While this type of
definition remains unsatisfactory, it introduces a further dynamic criterion to distinguish
sectors, that is, their scope for technical change.
More recently, Pavitt (1984) introduced an innovative taxonomy to study sectoral patterns
of technical change and innovation modes. Pavitt’s taxonomy consists of four categories
of industrial firms:
(a) Supplier-dominated, including firms mainly operating within traditional manufac

turing sectors such as textiles and agriculture which tend to rely on technical innova
tions external to the firm;
(b) Scale-intensive, including large firms producing basic materials like steel and
plastics and consumer durables such as automotive. For firms operating in these in
dustries, innovation may be both internal and external to the firm with a medium lev
el of appropriability;
Page 5 of 27
dustrial Policy
(c) Specialized suppliers, including smaller firms and specialist contractors whose
main activity is to produce high-tech machineries, tooling, and instruments. (p. 374)
Given their specialization and customization capability, these firms have a high level
of technology appropriability;
(d) Science-based firms, including high-tech R&D-intensive enterprises relying on
both internal and external sources of innovations (e.g. public laboratories, university
research), typically operating in industries like pharmaceuticals and electronics. Giv
en their continuous focus on product and process innovation, these are firms with a
high degree of appropriability from tacit knowledge and formalized, though protect
ed, knowledge, that is, patents.
As pointed out by Castellacci (2008): ‘Pavitt’s model of the linkages between science-
based, specialized suppliers, scale-intensive and supplier-dominated industries provides a
stylized and powerful description of the core set of industrial sectors that sustained the
growth of advanced economies during the Fordist age.’ Unlike the commodity/product-
based definition of sectors, the Pavitt taxonomy enables sectors to be defined according
to the ways in which technical change and innovation occur within different types of com
panies within these sectors and, thus, how organizations become competitive (Lazonick,
1990).
Competitiveness conditions—and, in particular, profitability—are also determined by the

degree of barriers to entry and concurrent imperfect competition. Reinert (1994) uses
these two parameters to establish a dynamic taxonomy of economic activities and mea
sures the extent to which profitability results (or not) in the ability of firms to raise
wages. In this framework, how fast technical change becomes subject to ‘perfect competi
tion’—the ‘gravity’ in the system—will be a key determinant of the ability of an economic
activity to raise living standards. The dynamic imperfect competition that leads to high
barriers to entry—high profits and the possibility of high wages and high taxation—histor
ically was mainly found in manufacturing industry, although new sectors have been wit
nessing similar patterns.
Alongside the two main conceptualizations of sector—commodity/product-based or pro

duction/technology-based—briefly reviewed so far, there is one last definition which is as
sociated with a more sociological and spatial way of clustering economic activities and
firms. Building on Alfred Marshall, Giacomo Becattini (1979, 2002) resurrected the con
cept of ‘industrial district’ as a way of defining clusters of firms based on their social ties,
geographical proximity, technological interdependencies, and mixed relationships of co
operation and competition. This neo-Marshallian idea of identifying firms according to
their ‘sense of belonging’ points to the fact that the relationships with a certain location
of production (and market) define the firm better than its produce or production technolo
gies. The idea of industrial district points to the existence of a unit of production beyond
the individual firm, that is, a cluster of ‘locally embedded firms’ whose characteristic fea
tures result from their embeddedness rather than referring to a simplistic product- or
production-based classification of sector. Two firms in the same sector are going to be
very different, despite producing the same product (or using the same factors of produc
Page 6 of 27
dustrial Policy
tion) if one of them is embedded in a certain industrial district, while the other is not (see
also Andreoni, 2018; Andreoni and Lazonick, 2020).
This brief excursus into the history of economic analysis reveals how sectors (and
(p. 375)
firms as their components) can be defined according to the ‘what’ they produce—com
modity/product-based classification—, the ‘how’ they produce—production/technology-
based classification, and the ‘where’ they produce—location-based classification. Each of
these definitions tends to reveal certain issues, while concealing others. And according to
our epistemological stance in the conceptualization of sector, we might end up having a
more or less clear picture of the ‘sectoral terrain’ we are considering, its boundaries and
its role in driving structural change. In this respect, and with a focus on goods, Oscar
Lange writes: ‘the classification of goods cannot be made on a purely arbitrary basis, be
cause the laws of economics would be then dependent on the particular classification
adopted. This would restrict the significance of the propositions of economics to a degree
that would make them valueless. Each proposition can be changed in its opposite by a
mere reclassification of goods’ (Lange, 1953: 103).
Let’s then consider issues that arise from different classifications of sectors and how they
impact our understanding of technical change.
To start with, a commodity/product-based classification of sectors tends to become quick

ly outdated because of changes in products and their differentiation. The fact that since
the ICT revolution material products might become platforms for accessing intangible
contents or services has accelerated this disconnect between traditionally defined sectors
and their differentiated products (Andreoni, Chang and Labrunie, 2021). Thirty years ago,
Becattini (2002: 485) had already posed the following question: ‘What is the significance
today—amid the growing differentiation and personalization of products, intense intra-
sectoral fragmentation, and the proliferation of “cross-sectors”—of the traditional “indus
tries” and the old, clearly-defined markets for substantially homogeneous items?’
Moreover, as we have seen, when we adopt a definition of sector simply based on ‘what’
is produced, there is a tendency to define primary sectors like agriculture or fishing as
‘low’-technology sectors. Their produce is perceived as relatively basic, with unfavourable
terms of trade and targeting relatively inelastic demand. However, if we instead concep
tualize a sector according to the ‘how’ and, potentially, ‘where’ of production, we realize a
number of important factors associated with technical change which are also relevant for
industrial policy.
First, the issue of heterogeneity in the how of production. If we look at the ‘how’ of pro
duction we discover that products within the same product groups can be obtained from
extremely different production processes involving different types, combinations, and lev
els of technology (Andreoni, 2014). This means that product-based definitions of sectors
might obfuscate high degrees of structural heterogeneity within sectors and might induce
the wrong type of comparative assessments across sectors (Andreoni and Chang, 2017;
see also Andreoni and Tregenna, 2019 on problems of structural heterogeneity in the
analysis of premature de-industrialization). For example, let’s look at the food and bever
Page 7 of 27
dustrial Policy
age industry. Food can be (and, indeed, still is) produced using traditional techniques and
relatively basic machinery (especially in developing countries); (p. 376) however, in other
countries food production has reached extremely high levels of industrial sophistication.
Food products are obtained from ‘industry-type’ controlled processes, involving complex
and highly automated technologies to benefit from scale economies and meet quality
standards. In these cases, food production also relies on just-in-time global supply chains
—as in the automotive industry—and the distribution of the produce involves sophisticat
ed cold and logistics chains for timely market delivery. All this suggests that food and bev
erages can be a medium/high-tech industry, depending on the ‘how’ of production.
Second, as already noted, this heterogeneity in the ‘how’ of production is often associated
with heterogeneity in the ‘where’ of production. Becattini (2002) distinguishes firms with
in the same sectoral classification—for example, garments—according to the external
economies they can benefit from. A garment firm might be ‘encamped’ in its location and
receive very little help; thus, this firm might easily move from that location without a re
duction in productivity or innovativeness. In contrast, another garment firm might be
‘rooted’ in its location and benefit from externalities arising from its productive surround
ings and relationships with other firms or markets. These collective capabilities are major
drivers of innovation, technical change, and diversification (Penrose, 1959; Freeman,
1974; Andreoni, 2018). Even if these two firms are both sectorally defined as garment
firms, the encamped firm is distinctively different from the rooted firm when it comes to
their production, technological, and organizational capabilities. In this case, the ‘where’
of production, that is, being rooted in a certain industrial district or not, becomes the
main unit of analysis for understanding heterogeneity.
Third, the issue of industrial mutations. In order to reach ever-rising product standards,
even a sector traditionally perceived as low tech must transform itself into a higher-tech
sector. This means it will have to adopt complex processes and technologies and rely on
quite sophisticated supply chains. For example, producing and selling fresh fruit might re
quire more sophisticated technologies than producing a T-shirt, if to retain freshness
firms in the sector must engage with sophisticated cold-chain and logistic solutions
(Cramer and Sender, 2015). An agrobusiness firm involved in this sector and operating in
markets that value freshness will undergo an innovative process of ‘industrial mutation’
making it more like what would be traditionally considered a manufacturing company.
Fourth, the issue of the sources of technical change. By changing the ‘how’ of production,
that is, changing the technologies and processes required to obtain a certain product or
reach a certain market, even traditional sectors can become major drivers of technical
change and innovation. Indeed, by responding to location specific challenges and oppor
tunities in the production of agro-commodities, firms in the agricultural sector or mining
have triggered and steered innovation in a number of machineries, tooling, and instru
mentation industries. Again, if we simply limit our attention to the commodity/product it
self without looking at the production process behind it, we might miss these important
technical changes. For a long time, the agricultural sector has been associated with de
Page 8 of 27
dustrial Policy
creasing returns and other stylized (p. 377) features which undermine the scope for ‘man
ufacturing agrarian change’ (Andreoni, 2011 and 2014).
Fifth, understanding structural change as a non-linear process. Grading sectors from pri
mary to secondary and tertiary according to a simple focus on the ‘what’ produced or a
superficial understanding of the ‘how’ of production, has often resulted in a ‘linear’ view
of structural change. According to this view and traditional sectoral categorizations,
countries are naturally supposed to move from agriculture to manufacturing and finally to
services. This linear view of structural change is, however, problematic with respect to
the analysis of the relationships between agriculture and manufacturing, between manu
facturing and services, and between services and agriculture. The reason is that in the
process of structural change, traditionally defined sectors continuously cross their bound
aries and thus shift their sectoral terrains. Technical change and technological interde
pendencies across sectors make for a large part of these sectoral crossings and shifts
along the process of structural change.
Kay (2009) pointed out how the relationship between agriculture and manufacturing is
symbiotic in economic development, as the development of each sector creates the condi
tions for the development and productivity increases of the other in a circular and cumu
lative fashion, more than in linear process. Increasing agricultural productivity results
from industrializing agricultural processes through technical change stemming mainly
from manufacturing (Andreoni, 2011). These intersectoral linkages are destined to
change and ‘vary according to the particular phase of the development process and as
structural conditions and international circumstances change’ (Kay, 2009: 116). For exam
ple, it has been observed that, with the increase of productivity in agriculture, linkages
between agriculture and services have also been expanding in magnitude and quality. Ex
amples include post-harvest facilities such as transport, communications, information ser
vices for production control in agriculture, market services. Some of these activities sit at
the boundaries of traditionally defined agricultural and service sectors.
Similarly, since the 1980s, several scholars have pointed out how the boundaries between
manufacturing and services have become blurred as a result of two parallel processes of
change. These are, first, the outsourcing of several activities like R&D from manufactur
ing companies to service companies and second, the ‘servitization’ of manufacturing com
panies, the process whereby companies in manufacturing sectors increasingly add value
to their products by adding post-sales services (for a review see Baines et al., 2017). Be
cause of the outsourcing of business- and production-related activities from traditionally
defined manufacturing companies, we have witnessed the expansion of the service sector
(also from a national accounting point of view).
If we look at product-based classifications of sector, companies engaging in R&D or tech

nology scaling-up are in the ‘service sector’, as they provide business services. However,
if we look at the ‘how’ they produce these services we realize that this happens by engag
ing in technology activities which were traditionally classified within the manufacturing
sector. Similarly, many companies included in manufacturing sectors are increasingly of
Page 9 of 27
dustrial Policy
fering a wide array of product‒services combinations. For (p. 378) example, traditional
downstream services are built into the product—‘embedded services’, such as in-flight
monitoring of aero engine systems. In other cases, such as the provision of network-infra
structure solutions in information and communication technology (ICT), a company can
provide an ‘integrated product‒service solution’ satisfying several consumer needs. Be
yond case study research, Park and Chan (1989: 211) were among the first to show em
pirically how ‘the evolution of the intersectoral relationship between services and manu
facturing in the course of development is symbiotic, in the sense that…structural change
of the former is bound to affect that of the latter’.
To summarize, going beyond the commonly used commodity/product-based classification

of sectors, in this section we have shown how the ‘what’, the ‘how’, and the ‘where’ of
production matter. Each conceptualization of sector offers different analytical lenses for
navigating the traditionally defined sectoral terrains, and their heterogeneity and evolv
ing boundaries in modern industrial economies. Technical change is the main driver of
these heterogeneous and evolving structural transformations. We have pointed out how,
with changing production technologies, technical change constantly reshapes the nature
of traditionally defined sectors as it changes the production activities that firms are con
ducting within them. For example, firms in the traditionally defined agricultural sector
might start using advanced technologies to guarantee quality standards of certain prod
ucts for the international market. Similarly, manufacturing companies in mechanical in
dustries might start using advanced digital technologies to perform activities which were
traditionally associated with electronics industries. Finally, manufacturing companies
might become service companies, and vice versa (for a detailed discussion on ‘technology
fusion’, see Andreoni et al., 2021).
In section 14.3 we will look at how technical change underpins changes across different
and blurred sectoral terrains and how, in doing so, it opens space for diversification and
innovation in the what, how and where of production. In some cases, even completely
‘new’ sectors can emerge at the interfaces of traditionally defined sectors. This more sys
temic/cross-sectoral focus does not mean abandoning traditional sectoral distinctions or
undermining the existence of sectoral differences across economic activities. As we have
seen in this section, different types of conceptualization of sectors serve different analyti
cal purposes and can lead to different industrial policy considerations and different sec
toral targeting decisions.
However, we suggest looking at traditionally defined sectors from different perspectives

and within a systemic framework that fully acknowledges interdependencies across sec
tors and cross-cutting technical change dynamics. This is particularly important in the
study of technical change. As stressed by Nathan Rosenberg (1963):
It is necessary to discard the familiar Marshallian approach, involving as it does

the definition of an industry as a collection of firms producing a homogenous prod
uct—or at least products involving some sufficiently high cross-elasticity of de
mand. For many analytical purposes it is necessary to group firms together on the
Page 10 of 27
dustrial Policy
basis of some features of the commodity or final product; but we cannot properly
appraise important aspects of technological developments in the nineteenth centu
ry (p. 379) until we give up the Marshallian concept of an industry as the focal
point of our attention and analysis. These developments [rapid technical change in
the American production of machine tools] may be understood more effectively in
terms of certain functional processes which cut entirely across industrial lines in
the Marshallian sense. (Rosenberg, 1963: 422; italics added)
14.3 ‘Shifting Terrains’: Technical Change and

Digitalization within Industrial Ecosystems
Modern industrial economies consist of complex and dynamic interdependencies span
ning along and across various traditionally defined ‘sectoral terrains’ and technology plat
forms. As eloquently documented by Tassey (2010: 6): ‘Most modern technologies are sys
tems, which means interdependencies exist among a set of industries that contribute ad
vanced materials, various components, subsystems, manufacturing systems, and eventu
ally service systems based on sets of manufactured hardware and software. The modern
global economy is therefore constructed around supply chains, whose tiers (industries) in
teract in complex ways.’
In fact, while some technologies (including organizational techniques) are sector specific,
several others are cross-cutting traditionally defined sectoral boundaries, that is, they are
deployed (or deployable) transversally. These are not simply traditionally defined general
purpose technologies (GPTs),1 but also clusters of technological and organizational capa
bilities which firms deploy in producing different products, reaching certain product stan
dards and functionalities (Penrose, 1959; Richardson, 1972; Andreoni, 2014). Because of
these cross-sectoral technological and organizational interdependencies, technical
change like the introduction of a new chemical process, industrial material, machine,
tooling, sensor, or embedded software can stem from one sector and then ‘reverberate’
across many sectors and several companies constituting them. Sector-specific adapta
tions of these new technologies are always present, although in the applications of these
technologies companies from different sectors rely on the same technology platforms (An
dreoni, 2018).
These interdependencies involve different types of business organizations and supply

chains both competing and cooperating along sectoral chains of production tasks and ac
tivities. Back in the 1980s, Luigi Pasinetti (1981) stressed how the term ‘industrial sector’
fails to describe the structure of economic sectors. It was for this reason that Pasinetti in
troduced the concept of ‘vertically integrated sector’ to analyse the chain structure of
tasks and activities of economic sectors. Along these lines, more recently, (p. 380) the lit
erature on global value chains/global production networks has focused on the global gov
ernance and distribution of production functions, activities, and tasks among different
networked production units located in different countries (a multi-tiered organizational
Page 11 of 27
dustrial Policy
structure). Within this framework, the dynamics of value creation and capture have been
centre stage (Milberg and Winkler, 2013; Gereffi, 2018; Ponte et al., 2019). These studies
tend to go beyond the sector as the main unit of analysis, replacing it with ‘task’ and
‘chain/network’. A production task is linked to a certain functional stage of the production
process or the production of a finite intermediate product component. The network is
multi-country (often more regionally confined than truly global) and composed of multiple
productive organizations involved in different stages and potentially operating in multiple
sectors (for a critical review see Andreoni, 2019a).
A number of ecosystem frameworks have been introduced with a view to capturing both
these systemic dynamics of technical and organizational change, thus capturing the new
terrain of the industrial. Among them, Andreoni (2018) developed a model of industrial
ecosystem consisting of a matrix of distinct sectoral value chains underpinned by differ
ent technology platforms (also called clusters of capability domains in specific industrial
locations). The idea of sectoral value chain is based on a product-based conceptualization
(and empirical operationalization) of sector, the ‘what’ of production. However, it also in
tegrates other insights into sector-specific supply-chain features, different degrees of
‘vertical integration’, and spatial distribution of production activities in different loca
tions. Each sectoral value chain is underpinned by a sector-specific combination of tech
nologies; thus, each sectoral value chain can also be identified according to the ‘how’ of
production. These combinations of technologies draw from broader cross-sectoral tech
nology platforms constituting several types of key enabling technologies, including mod
ern digital ones. These enabling technologies can be potentially deployed across a range
of different sectoral value chains, bypassing the traditional idea that certain sectors are
doomed to be low tech or characterized by low scope for innovation and technical
change. Figure 14.1 illustrates the industrial (p. 381) ecosystem framework, with a focus
on a number of digital technology platforms (and other key enabling technologies) and a
list of representative sectoral value chains.
Figure 14.1 Industrial ecosystem framework
Source: Author.
Within this framework, technical change is a sector-specific process as it involves chang

ing sector/product-specific combinations of technologies whereby products are designed,
produced, commercialized, and serviced. However, technical change is also a cross-sec
toral process of transformation constantly redefining the boundaries across sectors and
the emergence of new ones. Emerging technologies tend to have a disruptive impact on
existing technology platforms, and require organizations to develop new sectoral value
Page 12 of 27
dustrial Policy
chains capable of producing and/or deploying and/or commercializing them for the mar
ket.
With the increasing transformation of the global manufacturing landscape and the digital
ization of the economy, cross-sectoral technical change and innovation dynamics have ac
quired increasing importance. Traditionally, industrial ecosystems structured around ver
tically integrated sectors would mainly experience internal processes of ‘structural
learning’ (Andreoni, 2014), whereby a new technical solution emerging in one sector
would be increasingly adopted by firms in the same sector to transform their business.
In modern industrial ecosystems characterized by the presence of multiple sectoral sup

ply chains involving multiple actors (at different levels/tiers) located in various locations,
the emergence of an innovative digitally networked solution can rapidly spread across
sectoral value chains. These technical change dynamics are ultimately the result of innov
ative processes occurring within business enterprises (and other organizations and insti
tutions) and their continuous attempt to strike a balance between their specialization and
diversification, given certain market opportunities/incentives and their capabilities. While
often presented as disruptive technological changes, many digital technologies have
evolved and emerged from the same engineering and organizational principles as in pre
vious industrial revolutions and have been incrementally diffused across sectoral value
chains since the 1990s (Andreoni and Anzolin, 2019; Andreoni et al., 2021).
These innovative processes often involve firms crossing sectoral boundaries and ventur
ing into different sectoral terrains. Interfaces between sectoral value chains become ar
eas for the development of cross-sectoral ‘symbiotic relationships’. For example, ad
vanced materials such as membranes and polymers for micro-tubing developed in the de
sign and manufacture of medical devices can be adopted to increase the performance of
components and technology systems in the automotive sector. The production of the same
advanced materials might also require fundamental innovations in the production tech
nology supply chains (e.g. co-injection technologies for micro-tubing), and a set of infra-
technologies to test and validate the new products or production technologies. These new
production technologies, in turn, can lead to advances and the removal of bottlenecks in
the production technologies underpinning other sectoral supply chains (Andreoni, 2018).
The industrial ecosystem framework points to the importance of tracking these trajecto
ries of technical change, cutting across sectoral boundaries and constantly reshaping
each sectoral terrain. For example, throughout the 1990s and 2000s, (p. 382) technical
change and innovations in electronics and mechatronics have transformed the automotive
and machine tool industries—today electronics is dominant even in traditional metal me
chanical sectors. Similarly, the increasing reliance on digital technology platforms in mod
ern production is altering the nature of agriculture and manufacturing production, while
blurring the boundaries between physical and digital production technologies and service
systems. Across all sectors we can see signs of these ‘industrial mutations’: from satellite
systems controlling autonomous tractors or data-driven feeding systems in agriculture to
advances in intelligent automated systems, robotization, and additive manufacturing, and
Page 13 of 27
dustrial Policy
related data analytics—IoT, digital platforms, and digital supply chains in industrial and
service sectors (Sturgeon, 2017).
Industrial mutations driven by technological change—in particular, the widespread diffu

sion of digital technologies—have started shifting the terrain of the industrial and led sev
eral scholars to claim that we are on the verge of a new industrial revolution. However,
these shifts are not occurring in the same ways across all production locations, regions,
and countries. The ‘where’ of production does matter when we move from a stylized dis
cussion of technical change across sectoral value chains and technology platforms to a
more specific analysis of technical change in context. Digital technical change opportuni
ties are not equally distributed as companies, regions, and countries face different chal
lenges. The effective adoption of these new technologies presupposes the existence of
productive organizations and supply chains around them endowed with basic and inter
mediate production capabilities and supported by enabling infrastructures such as reli
able electricity, standardization, and connectivity. If we look at a wide range of fast
emerging and developing countries throughout the first two decades of the twenty-first
century, these conditions are still largely missing for most developing countries, as well
as in many regions across mature industrial economies (Andreoni and Anzolin, 2019).
Among late industrializers like China and India, while the use of digital production tech
nologies is speeding up, these technologies are still concentrated in a few sectoral value
chains (and companies and supply chains within them) and the full automation of rou
tinized tasks is far from being as diffused as many observers seem to suggest. There are
multiple reasons for this, including the infrastructural preconditions and trade-offs posed
by digital production technologies, with respect to both the hardware and software of
these technologies and, even more critically, their production system integration. Having
said that, several production locations have started connecting and cross-sectoral value
chains using digital production technologies are gradually emerging, also thanks to indus
trial policies by governments such as those of China, Thailand, and Malaysia (Andreoni,
2019b).
In low- and middle-income countries, such as Brazil and South Africa, companies are fac
ing more fundamental problems related to access to these digital production technolo
gies, their integration and retrofitting into existing production systems, not to mention
the availability of basic production capabilities and enabling infrastructure. In these coun
tries, the lack of job creation is not so much the diffusion of digital technologies as, large
ly, the result of structural and political economy problems, including premature de-indus
trialization, lack of productive organizations in key (p. 383) GVC stages, and basic produc
tive capabilities (Andreoni and Tregenna, 2020). In countries like Thailand and Vietnam
where the political economy configuration—developmental state coalition—has led to
high levels of investment and increasing numbers of export-led competitive companies,
governments are pushing for digital technical change and diffusion, especially in sectors
like automotive and electronics.
Page 14 of 27
dustrial Policy
In least developed countries, especially in the African continent, the lack of competitive
productive organizations makes the deployment of these digital technologies in produc
tion even more rare. While there are suitable applications for some basic information and
communication technologies (ICTs), such as for financial transactions on online finance
platforms or sharing some basic agriculture price data, these are in fact more like appli
cations of the Third Industrial Revolution than actual digital production technologies. A
limited number of companies are experimenting with the use of digital production tech
nologies, to the extent they are involved in production activities. So, for example, we see
some experimental applications in high-value agricultural products, extractive processes,
and trade logistics. Unfortunately, given the limited amount of manufacturing industries
and competitive companies, these countries are still unable to capture the potential ‘digi
tal dividend’ promised by the big new wave of technical change.
To summarize, the diffusion and effective deployment of digital technologies across sever
al sectoral value chains will be determined by several factors. In particular, the extent to
which these new technologies are the most cost-effective way to produce a certain com
ponent or product (still not always the case), and the extent to which companies have a
sufficient bundle of capabilities to make the absorption and effective deployment of these
technologies possible. Without the development of basic and intermediate-level capabili
ties, that is, basic production capabilities, and technology absorption and retrofitting ca
pabilities, digital technical change will remain a technological mirage. Industrial policy
can, however, play an important role in shaping technical change and redesigning the
sectoral boundaries and terrain of the industrial.
The final section before the conclusions (section 14.4) looks at the implications of techni
cal change and the shifting terrain of the industrial for industrial policy design (what se
lectivity means), and implementation (what cross-sectoral institutions are required).
14.4 Industrial Policy: Challenges and Opportu

nities in the New Digital Terrain
Since the First Industrial Revolution, industrial policies have always shaped and driven
the transformation of the economy, particularly when engagement with new technologies
was uncertain and required coordination and the commitment of resources (p. 384)
(Chang, 2002; Andreoni and Chang, 2019). The theoretical perspectives advanced in this
contribution have pointed to the fact that the opportunities associated with technical
change are very heterogeneous, in some cases sector and process specific, in other cases
relying on cross-sectoral technology platforms underpinning several sectoral value
chains.
With the emergence of a new ‘digital terrain’, several manufacturing industries remain
the main target of industrial policy investment. However, while some of them—the ma
chine tool industry, for example—might play a key ‘technology push role’, others—high-
tech agriculture and production-related services in particular—will play a key ‘demand
Page 15 of 27
dustrial Policy
pull role’. For example, the application of manufacturing principles to agricultural pro
duction could deliver dramatic productivity gains and better international market access,
while creating the demand for an increasingly advanced and data-driven agricultural
equipment industry. The enormous potential of digitalization in mining could similarly be
a demand pull factor towards the development of world-leading mining equipment indus
tries in the African continent and several other countries in Latin America. By targeting
investments at the intersection of these demand pull and technology push dynamics, de
veloping countries can also manage to decrease their reliance on pre-made machinery
from advanced industrial economies—and the associated trade burden. Investing in these
intersections of emerging industrial ecosystems can be also a way of laterally entering
sectoral value chains and engaging with technologies which would be otherwise difficult
to access.
The specific challenges faced by developing countries in engaging with the Fourth Indus
trial Revolution can be clustered in five main groups.
14.4.1 Technology Absorption, Deployment, and Technical Change:

‘Capability Threshold’
First, the basic and more advanced production capabilities required for the absorption
and effective deployment of the new digital production technologies, and their diffusion
along the supply chains, are scarce and unevenly distributed. Moreover, digital produc
tion technologies have also raised the ‘capability threshold’ that companies are required
to achieve to make effective use of the new technologies. This is not because parts of
these technologies are completely new—for example, automated machinery goes back at
least as far as the Second Industrial Revolution (2IR), but mainly because the Fourth In
dustrial Revolution (4IR) is about the ‘fusion of existing and new technologies’ into com
plex integrated technology systems (Tassey, 2007; Andreoni, Chang, and Labrunie, 2021).
Managing complex integrated technology systems like a fully automated production line,
combining robots and IoT technologies, is an extremely demanding task for a productive
organization in a developing country.
(p. 385) 14.4.2 Production System: Retrofitting and Integration
Second, industrialization is about commitment of resources under conditions of uncer

tainty (Chang and Andreoni, 2020). Most of these commitments involve physical capital
that embodies certain technologies and cannot be remoulded in any significant way to
embody other technologies. Very often the commitments are organizational too, involving
individuals specializing in specific skills. These commitments are critical because they
raise productivity, for example by enabling economies of scale in production. However,
depending on the degree of reversibility of the investments, these commitments make fu
ture changes more difficult. This introduces a very specific challenge in developing coun
tries. Existing companies that can make technology investments have already committed
resources and have to see how they can retrofit and integrate new digital production
technologies into their existing production plants. Setting up brand-new plants is less
Page 16 of 27
dustrial Policy
common as it requires significant long-term investment and access to markets, and these
new plants may be difficult to run given the lack of basic and digital infrastructure.
14.4.3 Infrastructure: Basic and Digital
Third, digital production technologies are very demanding when it comes to the infra
structure required to enable their use in production. Developing countries face signifi
cant challenges providing affordable and reliable electricity, as well as good enough con
nectivity. In some cases, these infrastructural bottlenecks have been bypassed by off-grid
energy technologies and wireless connectivity systems. While these solutions work in cer
tain areas, they are not always able to provide the quality and reliable services needed to
run digital production technologies effectively. As a result, the productivity and quality
improvements that digital production technologies offer are offset by these infrastructur
al bottlenecks, which can make technology investments by individual companies too risky
and ultimately uneconomical.
14.4.4 Technology Diffusion: 4IR Islands and the Digital Capability

Gap
Fourth, despite the fact that in almost all developing countries it is possible to find ‘4IR
islands’, that is, a company which is engaging with some digital production technologies,
many of these technologies remain contained within the company and their production
cells. Occasionally, a few close suppliers who have the basic production capabilities to
make use of them might be linked to these 4IR islands. Moreover, the 4IR islands often
rely on enabling infrastructural capabilities (connectivity, electricity, (p. 386) etc.) built by
the same company at their own cost for their own plants. Without these infrastructures,
companies would not even be able to switch on digital production technologies. Around
these 4IR islands the large majority of companies and sectors are still operating fully
within the Third Industrial Revolution technology paradigm and are unable to operate at
the same standards as the island companies. In some other cases, especially among least
developed countries, companies have not even engaged with the Second Industrial Revo
lution yet. This means that it is extremely difficult for the leading companies—such as an
OEM—to link backward and nurture their supply chains. The ‘digital capability gap’ be
tween company islands and suppliers is so extreme and so costly to address (given the ex
isting endogenous asymmetries—see section 14.4.5) that the diffusion of 4IR technologies
remains very limited.
14.4.5 Technology Access and Affordability: Endogenous Asymmetries
Fifth, digital production technologies are complex and are controlled by a limited number
of advanced countries and their leading companies. Developing countries rely enormously
on the importation of these technologies from advanced economies, and in many cases
even when they are able to mobilize significant resources to access them, they are tied to
their buyers with respect to both hardware and software components. International buy
ers and OEMs control the source, type, and utilization of the digital production technolo
Page 17 of 27
dustrial Policy
gies by setting parameters of engagement for the suppliers. Those who cannot meet these
parameters are cut out. The importance of using common protocols and software plat
forms for the deployment of digital production technologies carries the risk of a vertical
ization and concentration of power. Also, many of these technologies are not ‘plug and
play’, that is, the acquisition of the hardware goes hand in hand with the need for expen
sive technology services and royalties for the use of the related software (Sturgeon,
2017).
To capture these opportunities and address these new challenges, the most recent inter
national industrial policy experiences point to the importance of adopting new industrial
policy frameworks and principles (Andreoni, 2016, 2019a; Andreoni and Chang, 2019; An
dreoni and Tregenna, 2020). These new frameworks should capture the changing reality
of production systems and products, and their underlying technology platforms and orga
nizational models, that is, the industrial ecosystem. New industrial policy principles
should also reflect the need for more strategic coordination among (and within) public
and private sectors; better targeting and policy alignment; and the introduction of feasi
ble policies whose implementation goes hand in hand with a change in the country’s polit
ical economy and the emergence of new productive alliances, the following sentence re
fer to all five points above 14.4.1–14.4.5.
Against this background, the following five industrial policy design principles must be
considered if the opportunities of the new digital terrain are to be captured:
(p. 387)
• Digital industrialization opportunities are about both incremental changes in produc

tion and disruptive technological innovations. Industrial policy must be appropriate,
that is, it must address those foundational—basic and intermediate—technical and or
ganizational capability requirements without which the digital innovation opportuni
ties cannot be captured, effectively deployed, and diffused across the industrial ecosys
tem.
• Digital industrialization is sustainable and inclusive only if it creates conditions for
more domestic value creation, capture, distribution, and retention. Digital technologies
provide new platforms for developing linkages in the ecosystem, including more and
new productive organizations and customers. Governing the technological, competi
tion, and regulatory processes through which these linkages are formed (or prevented)
is critical for sustainable value addition and distribution in the economy. Digitalization
introduces a new policy dimension in the strategic management and regulatory en
forcement of ‘glo-cal’ relationships.
• Systemic changes call for systemic policy frameworks. Digital industrialization is
about a systemic transformation of the economy, that is, the emergence of a new indus
trial ecosystem. This is why the effectiveness of a digital industrial policy will depend
on interventions able to reflect both the specificities of existing sectors and the new
opportunities emerging within and across them.
Page 18 of 27
dustrial Policy
• Strategic targeting, coordination, and policy alignment are critical for effective im
plementation. Systemic transformation cannot be achieved with generic policy frame
works. Strategic choices have to be made to unleash the potential that digital technolo
gies offer for the innovative industrial renewal of existing sectors, and the emergence
of new competitive players in the industrial ecosystem. For successful policy imple
mentation, these choices must be effectively coordinated and aligned, and budgetary
inconsistencies and scarce public resource dispersion and capture avoided.
• New coalitions for change forged around better aligned productive interests are nec
essary for the successful implementation of industrial policy. The design of industrial
policy must take into account the different interests, capabilities, and distribution of
power within and among different organizations and constituencies in the economy,
and must find ways in which policies can lead to better alignment of productive inter
ests. Without such alignment, policies tend to be fragmented, difficult to enforce, and
vulnerable to unproductive rent capture.
The digital industrial policy principle and targets highlighted above point to the need for
both sector-specific digital industrial policy and cross-sectoral interventions, as well as
new ways of factoring in an assessment of the political economy feasibility of industrial
policy interventions.
• Sector-specific measures will take into account the organization, needs and condi
tions of firms within a particular industry, in particular the specific types of: digital
skills; digital technology infrastructures and services; challenges and barriers to
(p. 388) linkages development; competition conditions and value capture; and sectoral
regulatory frameworks and incentives, including procurement and market regulation.

• Cross-sectoral interventions should focus on those opportunities and challenges that
are faced by different firms across sectors, especially those related to broader founda
tional capabilities (e.g. basic and transversal digital skills, digital and manufacturing
extension services), technology infrastructure (e.g. digital software licensing, connec
tivity, data quality, and affordability) and broader financing, investment, and regulatory
conditions in the country.
The effectiveness of sectoral and cross-sectoral interventions across key policy areas will
depend on the extent to which the government is able to reach strategic alignment be
tween sectoral measures and cross-sectoral interventions, and to develop a governance
framework beyond policy silos. Breaking out of policy silos is thus both a matter of what/
how policy interventions are designed as well as how resources are allocated and gov
erned along different structural cycles (Andreoni et al., 2017; Andreoni and Chang, 2019).
The industrial ecosystem (Andreoni, 2018) provides the framework to operationalize and
foster cross-sectoral innovations around mission-oriented industrial policies (Mazzucato,
2018; Mazzucato and Kattel, Chapter 12, this volume).
Mission-oriented industrial policy provides a vision for the creation of new productive
sectors and markets around emerging human needs and social challenges. These mis
Page 19 of 27
dustrial Policy
sions are not simply visions; they are also focal points for coordination within an industri
al ecosystem as they trigger transformations along and across technology platforms and
sectoral value chains. For example, the environmental sustainability grand challenge re
quires a concerted transformation in the patterns of production and consumption of sev
eral sectoral value chains. It also requires the introduction and cost-effective diffusion of
new sources of energy, new advanced materials, new industrial processes. All these tech
nical changes trigger a reconfiguration of the technology platforms on which modern in
dustrial ecosystems are built as well as a change in the industry organization.
While industrial policy design and governance frameworks are critical, the effective im
plementation and enforcement of any industrial policy will critically depend on the extent
to which the policy is able to promote the emergence of a new coalition of productive in
terests, or offer existing powerful groups alternative and more productive ways to oper
ate in the economy. In other words, especially in developing and middle-income countries
with a poorly developed capitalist class, industrial policies will have to be designed taking
into account the political economy feasibility of interventions. This assessment should not
discourage bold industrial policy measures; rather, it should help identify feasible and in
cremental pathways for transformation (Chang and Andreoni, 2020).
There are a number of sectors in which productive interests predominate, while capabili
ties and power are relatively evenly distributed. In these sectors, the political settlement
structure might allow for quick wins and the emergence or consolidation of (p. 389) new
coalitions for change. In other sectors, where interests have been more entrenched and
have even resulted in state capture, consolidated power structures are more difficult to
change and open up to new actors and competition. Here, competition policy alongside
other regulatory and incentive restructuring can force the emergence of new deals. These
deals can open the way to transition paths towards more productive and technologically
dynamic sectors.
Overall, digital industrialization will also produce potential trade-offs and new conflicts in
the economy, for example with respect to employment. Given the challenges faced by
SMEs, digital technologies, if not suitably governed, can exacerbate the existing divide
between big and small firms to the detriment of the much-needed re-industrialization of
the country. Digital industrial policy must therefore govern these processes and make
sure that the digital industrial dividend is incrementally distributed across different firms,
their employees, and society more broadly.
14.5 Conclusions
Economists have always struggled to grasp the complex structure and dynamics of pro
ductive transformation. Since the First Industrial Revolution, several concepts of sectors
have been used as compositional heuristics to group and differentiate economic activities,
markets, and firms. These heuristics and resulting taxonomies have been also used to
capture the way in which technical change constantly shifts the sectoral terrains of tradi
tionally defined agriculture, industry, and services. In this contribution we have provided
Page 20 of 27
dustrial Policy
an analytical discussion of the evolution of different concepts of sectors and demonstrat
ed how different concepts can serve different purposes.
With a focus on technical change in the digitalization era, we have relied on an industrial
ecosystem framework to discuss several processes of innovation, industrial mutation, and
shifting sectoral terrains. Indeed, traditional sectoral terrains have been increasingly
shifting and blurring as a result of changes in the technological platforms underpinning
sectoral value chains. Many of these platforms have been increasingly acquiring a data-
driven and networked character. Since 2010, digitalization has emerged as the dominant
form of technical change.
However, digital transformation in production and products is diffusing unevenly across

the world. Industrial policies are central to boosting technical change, and their effective
ness depends critically on the adoption of appropriate sectoral and cross-sectoral instru
ments. Industrial policy targeting requires a focus on key parts of modern industrial
ecosystems. New heuristics are critical to reveal opportunities and challenges within sec
toral value chains and those nested at the interstices of industrial ecosystems. Mission-
oriented initiatives can play a key coordinating role and support countries in developing
context-specific pathways to capture the new windows of digital opportunities.
(p. 390) Acknowledgements

Thanks to Ha-Joon Chang, Simon Roberts, and Timothy Sturgeon for discussions and Eric
Reinert and Eva Paus for comments to an early version of this chapter. Also, thanks to all
participants to the workshops held in preparation of the OUP Handbook, and in particular
the editors of the volume.
References
Abramovitz, Moses (1989) Thinking about Growth. Cambridge: Cambridge University
Press.
Amsden, Alice (1991) ‘Diffusion of Development’, American Economic Review 81: 282–6.
Andreoni, Antonio (2011) ‘Manufacturing Agrarian Change: Agricultural Production, In

ter-sectoral Learning and Technological Capabilities’. DRUID Working Paper Series No.
11/2013.
Andreoni, Antonio (2014) ‘Structural Learning: Embedding Discoveries and the Dynamics
of Production’, Structural Change and Economic Dynamics 29: 58–74.
Andreoni, Antonio (2016) ‘Varieties of Industrial Policy: Models, Packages and Transfor
mation Cycles’, in Akbar Noman and Joseph E. Stiglitz (eds) Efficiency, Finance, and Vari
eties of Industrial Policy. New York: Columbia University Press, pp. 245–305.
Page 21 of 27
dustrial Policy
Andreoni, Antonio (2018) ‘The Architecture and Dynamics of Industrial Ecosystems: Di
versification and Innovative Industrial Renewal in Emilia Romagna’, Cambridge Journal of
Economics, 42: 1613–42.
Andreoni, Antonio (2019a) ‘A Generalised Linkage Approach to Local Production Systems

Development in the Era of Global Value Chains, with Special Reference to Africa’, forth
coming in Ravi Kanbur, Akbar Noman, and Joseph E. Stiglitz (eds) Quality of Growth in
Africa. New York: Columbia University Press, pp. 264–94.
Andreoni, Antonio (2019b) ‘Robotising Regions: A Comparative Analysis of Industrial Poli

cy for Robotisation across China, Malaysia and Thailand’. South African Research Chair
in Industrial Development, SARChi Working Paper, University of Johannesburg.
Andreoni, Antonio and Guendalina Anzolin (2019) ‘A Revolution in the Making? Chal
lenges and Opportunities of Digital Production Technologies for Developing Countries’.
UNIDO Working Paper No. 7/2019. Vienna: UNIDO.
Andreoni, Antonio and Ha-Joon Chang (2017) ‘Bringing Production and Employment back
into Development’, Cambridge Journal of Regions, Economy and Society 10(1): 173–87.
Andreoni, Antonio and William Lazonick (2020) ‘Local Ecosystems and Social Conditions
of Innovative Enterprise’, in Arkebe Oqubay and Justin Yifu Lin (eds) The Oxford Hand
book of Industrial Hubs and Economic Development. Oxford: Oxford University Press.
Andreoni, Antonio and Roberto Scazzieri (2014) ‘Triggers of Change: Structural Trajecto
ries and Production Dynamics’, Cambridge Journal of Economics 38: 1391–408.
Andreoni, Antonio and Fiona Tregenna (2019) ‘Beyond the Inverted U: The Changing Na
ture and Structural Heterogeneity of Premature Deindustrialisation’. International Work
shop: The Future of Industrial Work: New Pathways and Policies of Structural Transfor
mation? UNIDO, Vienna, 19–20 September. (p. 391)
Andreoni, Antonio and Fiona Tregenna (2020) ‘Escaping the Middle-income Technology
Trap: A Comparative Analysis of Industrial Policies in China, Brazil and South Africa’,
Structural Change and Economic Dynamics, forthcoming.
Andreoni, Antonio, Ha-Joon Chang, and Matheus Labrunie (2021) ‘Natura non facit
saltus: Challenges and Opportunities for Digital Industrialisation across Developing Coun
tries’, European Journal of Development Research, forthcoming.
Andreoni, Antonio, Frederico Frattini, and Giorgio Prodi (2017) ‘Structural Cycles and In
dustrial Policy Alignment: The Private–Public Nexus in the Emilian Packaging Valley’,
Cambridge Journal of Economics 41: 881–904.
Page 22 of 27
dustrial Policy
Baines, Tim, Ali Ziaee Bigdeli, Oscar Bustinza, Victor Shi, James Baldwin, and Keith Ridg
way (2017) ‘Servitization: Revisiting the State-of-the-art and Research Priorities’, Interna
tional Journal of Operations & Production Management 37(2): 256–78.
Becattini, Giacomo (1962) Il concetto d’industria e la teoria del valore. Torino: Bollati Bor
inghieri.
Becattini, Giacomo (1979) ‘Dal “settore” industriale al “distretto” industriale. Alcune con
siderazioni sull’unità di indagine dell’economia industriale’, Rivista di Economia e Politica
Industriale 5(1): 7–21. (English translation: ‘Sectors and/or Districts: Some Remarks on
the Conceptual Foundations of Industrial Economics’, in E. Goodman, J. Bamford, and P.
Saynor (eds) Small Firms and Industrial Districts in Italy. London: Routledge, 1989, pp.
123–35.)
Becattini, Giacomo (2002) ‘Industrial Sectors and Industrial Districts: Tools for Industrial
Analysis’, European Planning Studies 10(4): 483–93.
Best, Michael H. (1990) The New Competition: Institutions of Industrial Restructuring.

Cambridge: Polity Press.
Brenahan, Timothy (2010) ‘General Purpose Technologies’, in Bronwyn H. Hall and

Nathan Rosenberg (eds) Handbook of Economics of Innovation. Amsterdam: Elsevier, pp.
761–91.
Castellacci, Fulvio (2008) ‘Technological Paradigms, Regimes and Trajectories: Manufac

turing and Service Industries in a New Taxonomy of Sectoral Patterns of Innovation’, Re
search Policy 37(6–7): 978–94.
Chang, Ha-Joon and Antonio Andreoni (2020) ‘Industrial Policy in the 21st Century’, De
velopment and Change 51(2): 1–38.
Clapham, John Harold (1922) ‘Of Empty Economic Boxes’, The Economic Journal 32(127):
560–3.
Cramer, Christopher and John Sender (2015) ‘Agro-processing, Wage Employment and Ex
port Revenue: Opportunities for Strategic Intervention’. Working Paper for the Depart
ment of Trade and Industry, Trade & Industry Policy Strategies (TIPS), Pretoria.
Dosi, Giovanni (1982) ‘Technological Paradigms and Technological Trajectories: A

Suggested Interpretation of the Determinants and Directions of Technical Change’, Re
search Policy 11(3): 147–62.
Fisher, Allan (1939) ‘Production in Primary, Secondary and Tertiary’, Economic Record
15(1): 24–38.
Page 23 of 27
dustrial Policy
Fourastie, Jean (1951) Machinisme et bien-etre. Paris: Les Editions des Minuit.
Freeman, Chris (1974) The Economics of Industrial Innovation. London: Pinter.
Gereffi, Gary (2018) Global Value Chains and Development: Redefining the Contours of
21st Century Capitalism. Cambridge: Cambridge University Press. (p. 392)
Hirschman, Albert Otto (1958) The Strategy of Economic Development. New Haven, CT:
Hirschman, Albert Otto (1977) ‘A Generalized Linkage Approach to Economic Develop

ment with Special Reference to Staples’, Economic Development and Cultural Change 25:
67–97.
Jovanovic, Boyan and Rousseau, Peter (2005) ‘General Purpose Technologies’, in Philippe
Aghion and Steven Durlauf (eds) Handbook of Economic Growth, Vol. 1B, Amsterdam: El
sevier, pp. 1181–224.
Kaldor, Nicholas (1967) Strategic Factors in Economic Development. Ithaca, NY: Cornell
University Press.
Kay, Cristobal (2009) ‘Development Strategies and Rural Development: Exploring Syner
gies, Eradicating Poverty’, Journal of Peasant Studies 36(1): 103–37.
Kuznets, Simon (1973) ‘Modern Economic Growth’, American Economic Review 63: 247–
58.
Lall, Sanjaya (1998) ‘The Technological Structure and Performance of Developing Coun
try Manufactured Exports, 1985–98’, Oxford Development Studies 28(3): 337–69.
Lancaster, Kelvin John (1966) ‘A New Approach to Consumer Theory’, Journal of Political
Economy 74: 132–57.
Lange, Oscar (1953) Price Flexibility and Employment. Chicago, IL: Cowles.
Lazonick, William (1990) Competitive Advantage on the Shop Floor. Cambridge, MA: Har
vard University Press.
Lazonick, William (1991) Business Organisation and the Myth of the Market Economy.
Leontief, Wassily (1941) The Structure of American Economy, 1919–1929. Cambridge,

Lin, Justin Yifu (2010) New Structural Economics: A Framework for Rethinking Develop
ment and Policy. Washington, DC: World Bank.
Page 24 of 27
dustrial Policy
Marshall, Alfred (1920) Industry and Trade: A Study of Industrial Technique and Business
Organization, and of their Influences on the Conditions of Various Classes and Nations.
London: Macmillan.
Mazzucato, Mariana (2018) ‘Mission-oriented Innovation Policies: Challenges and Oppor

tunities’, Industrial and Corporate Change 27(5): 803–15.
Milberg, William and Deborah Winkler (2013) Outsourcing Economics. Cambridge: Cam
bridge University Press.
OECD (1994) Globalisation and Competitiveness: Relevant Indicators. Paris: OECD Direc
torate for Science, Technology and Industry, DSTI/EAS/IND/WP9(94)19.
Park, Se-Hark and Kenneth Chan (1989) ‘A Cross-country Input–Output Analysis of Inter
sectoral Relationships between Manufacturing and Services and their Employment Impli
cations’, World Development 17(2): 199–212.
Pasinetti, Luigi Ludovico (1981) Structural Change and Economic Growth. A Theoretical
Essay on the Dynamics of the Wealth of Nations. Cambridge: Cambridge University Press.
Pavitt, Keith (1984) ‘Sectoral Patterns of Technical Change: Toward a Taxonomy and a
Theory’, Research Policy 6: 343–73.
Press.
Pérez, Carlota (2002) Technological Revolutions and Financial Capital: The Dynamics of
Bubbles and Golden Ages. Northampton, MA: Edward Elgar.
Ponte, Stefano, Gary Gereffi, and Gale Raj-Rechert (2019) Handbook of Global Value
Chains. Cheltenham: Edward Elgar. (p. 393)
Reinert, Erik S. (1994) ‘Catching-up from Way Behind: A Third World Perspective on First
World History’, in Jan Fagerberg, Bart Verspagen, and Nick von Tunzelmann (eds) The
Dynamics of Technology, Trade, and Growth. Aldershot: Edward Elgar, pp. 168–97.
Reinert, Erik S. (2008) How Rich Countries Got Rich and Why Poor Countries Stay Poor.
New York: Public Affairs.
Richardson, George (1972) ‘The Organisation of Industry’, The Economic Journal 82(327):
883–96.
Robinson, Joan (1933) The Economics of Imperfect Competition. Basingstoke: Macmillan.
Rodrik, Dani, Margaret McMillan, and Claudia Sepulveda (eds) (2017) Structural Change,
Fundamentals, and Growth: A Framework and Case Studies. Washington, DC: Interna
tional Food Policy Research Institute.
Page 25 of 27
dustrial Policy
Rosenberg, Nathan (1963) ‘Technological Change in the Machine Tool Industry, 1840–
1910’, Journal of Economic History 23(4): 414–43.
Rosenberg, Nathan (1976) Perspectives on Technology. Cambridge: Cambridge Universi

ty Press.
Scazzieri, Roberto (1993) A Theory of Production: Tasks, Processes, and Technical Prac
tices. Oxford: Clarendon Press.
Schumpeter, Joseph (1939) Business Cycles: A Theoretical, Historical, and Statistical

Analysis of the Capitalist Process. New York: McGraw-Hill Book Company.
Schumpeter, Joseph (1942) Capitalism, Socialism, and Democracy. New York: Harper &
Brothers.
Simon, Herbert A. (1991) ‘Organisations and Markets’, Journal of Economic Perspectives

5(2): 25–44.
Sraffa, Piero (1926) ‘The Laws of Returns under Competitive Conditions’, The Economic
Journal 144(36): 535–50.
Sturgeon, Timothy (2017) ‘The “New” Digital Economy and Development’. UNCTAD Tech
nical Note on ICT for Development. Geneva: UNCTAD.
Tassey, Gregory (2007) The Technology Imperative. Cheltenham: Edward Elgar.
Tassey, Gregory (2010) ‘Rationales and Mechanisms for Revitalising US Manufacturing

R&D Strategies’, Journal of Technology Transfer 35: 283–333.
Toner, Phillip (1999) Main Currents in Cumulative Causation. Basingstoke: Macmillan.
Wolfe, Martin (1955) ‘The Concept of Economic Sectors’, The Quarterly Journal of Eco
nomics 69(3): 402–20.
Notes:
(1) GPTs have been studied especially with reference to the emergence of new technology
paradigms and their broader impact on the economy (for a review see Jovanovic and
Rousseau, 2005; Brenahan, 2010).
Antonio Andreoni
Antonio Andreoni is associate professor in industrial economics at University College

London (UCL), and research head of UCL Institute for Innovation and Public Pur
pose. He is also visiting associate professor SARChI Industrial Development, Univer
sity of Johannesburg. He has acted as an adviser for the World Bank, OECD, DFID,
UNIDO, UNCTAD, ILO, UNDP, UNECA, UNU-WIDER, EU Growth and Innovation, and
to governments in South Africa, Tanzania, and the United Kingdom. His work on
Page 26 of 27
dustrial Policy
structural change, production, technology, innovation, financialization, corruption,
political economy, and industrial policy has appeared in Cambridge Journal of Eco
nomics, Development and Change, Cambridge Journal of Regions, Economy and Soci
ety, Structural Change and Economic Dynamics, Oxford Review of Economic Policy,
European Journal of Development Research, and Energy Policy.
Page 27 of 27
An Industrial Policy Framework to Advance a Global Green New Deal
An Industrial Policy Framework to Advance a Global

Green New Deal
Robert Pollin
This chapter argues that, as an analytic proposition and industrial policy challenge, it is
realistic that global CO2 emissions can be driven to net zero by 2050, the goal set in 2018
by the IPCC. This will require an average level of investment spending of about 2.5 per
cent of global GDP per year, focused in two areas: (1) dramatically improving energy-effi
ciency standards in the stock of buildings, automobiles and public transportation sys
tems, and industrial production processes; and (2) equally dramatically expanding the
supply of clean renewable energy sources—primarily solar and wind power—available at
competitive prices to all sectors and in all regions of the globe. The chapter examines the
industrial and financial policy measures that will be needed to support this global clean
energy investment project, focusing on approaches that have been implemented through
out the world to varying degrees.
Keywords: global clean energy investments, industrial/financial policies, net-zero goal, energy efficiency, solar and
wind power
15.1 Introduction
IN October 2018, the Intergovernmental Panel on Climate Change (IPCC), the most au
thoritative global organization advancing climate-change research, issued an alarming re
port titled Global Warming of 1.5°. This report emphasized the imperative of limiting the
increase in global mean temperatures to 1.5 degrees above pre-industrial levels as op
posed to the previous consensus target 2.0 degrees. The IPCC concluded that limiting the
global mean temperature increase to 1.5 rather than 2.0 degrees by 2100 will dramatical
ly lower the likely negative consequences of climate change. These include the risks of
heat extremes, heavy precipitation, droughts, sea level rise, biodiversity losses, and cor
responding impacts on health, livelihoods, food security, water supply, and human securi
ty.
Page 1 of 44
The IPCC estimates that to achieve the 1.5 degrees maximum global mean temperature
increase target as of 2100, global net CO2 emissions will have to fall by about 45 per cent
as of 2030 and reach net-zero emissions by 2050. I focus in this chapter on what it will
take for the global economy to reach net-zero CO2 emissions by 2050, and specifically, in
terms of the industrial and financing policies that will be needed for this project to suc
ceed. In the interests of space, I do not delve into the additional specific challenges
around also hitting the IPCC’s intermediate target of a 45 per cent CO2 emissions reduc
tion by 2030, though important additional challenges do emerge with achieving this 2030
goal.
In fact, purely as an analytic proposition and policy challenge—independent of the

(p. 395)
myriad of political and economic forces arrayed around these matters—it is entirely real
istic to allow that global CO2 emissions can be driven to net zero by 2050. By my higher-
end estimate, it will require an average level of investment spending throughout the glob
al economy of about 2.5 per cent of global GDP per year, focused in two areas: 1) dramati
cally improving energy-efficiency standards in the stock of buildings, automobiles and
public transportation systems, and industrial production processes; and 2) equally dra
matically expanding the supply of clean renewable energy sources—primarily solar and
wind power—available at competitive prices to all sectors and in all regions of the globe.
This level of clean energy investment spending would amount to about $2.6 trillion1 in the
first year of the programme, which I will set as 2024 and rise to an average of about $4.5
trillion per year between 2024 and 2050. This assumes that the project actually begins in
2021, but that the full-scale expansion in clean energy investments requires a three-year
lead time. Thus the start date for the full-scale programme becomes 2024. Total clean en
ergy investment spending for the full-scale twenty-seven-year investment cycle 2024–50
would amount to about $120 trillion.
These figures are for overall investment spending, including from both the public and pri
vate sectors. Establishing the right mix between public and private investment will be a
major consideration within the industrial and financing policies framework. As an initial
rough approximation, I assume that the breakdown should be divided equally—that is, 50
per cent public and private investment. For the first year of full-scale investment activity
in 2024, this would break down to $1.3 trillion in both public and private investments. A
major part of the policy challenge will be to determine how to leverage the public money
most effectively to create strong incentives for private investors.
It is important to emphasize at the outset that this clean energy investment project will
pay for itself in full over time. More specifically, it will deliver lower energy costs for ener
gy consumers in all regions of the world. This results because raising energy-efficiency
standards means that, by definition, consumers will spend less for a given amount of en
ergy services, such as being able to travel 100 miles on a gallon of petrol with a high-effi
ciency hybrid plug-in vehicle as opposed to 25 miles per gallon with the average car on
US roads today. Moreover, the costs of supplying energy through solar and wind power,
as well as geothermal and hydro, are now, on average, roughly equal to or lower than
Page 2 of 44
those for fossil fuels and nuclear energy. As such, the initial upfront investment outlays
can be repaid over time through the cost savings that will be forthcoming.
For 2018, global clean energy investments levels, including both energy efficiency and
clean renewable investments, were at about $570 billion, equal to about 0.7 per cent of
global GDP. Thus, the increase in clean energy investments will need to be in the range of
1.8 per cent of global GDP—that is, about $1.5 trillion at the current global GDP level of
about $86 trillion, then rising in step with global GDP growth thereafter (p. 396) until
2050. The consumption of oil, coal, and natural gas will also need to fall to zero over this
same thirty-year period. The rate of decline can begin at a relatively modest 3.5 per cent
in the initial years of the transition programme, but will then increase every year in per
centage terms, as the base level of fossil-fuel supply contracts to zero as of 2050.
Of course, both the privately owned fossil-fuel companies, such as Exxon-Mobil and
Chevron, and equally, the publicly owned companies such as Saudi Aramco and Gazprom
in Russia, have massive self-interests at stake in preventing reductions in fossil-fuel con
sumption as well as enormous political power. These powerful vested interests will simply
have to be defeated. How exactly this is accomplished is beyond the scope of this chapter,
other than to recognize that a critical foundation for advancing a successful global Green
New Deal will be to establish a viable set of industrial and financing policies to support
the project.
The structure of this chapter is as follows. Section 15.2 asks the first critical question for
designing a global clean energy investment project, which is: what is clean energy? I re
view evidence on natural gas, nuclear energy, and various forms of geoengineering as
providing clean energy alternatives to fossil fuels. But I conclude that all these approach
es present major problems. This conclusion then becomes the basis for recognizing that
building a global clean energy economy should rely mostly on dramatically expanding in
vestments in energy efficiency and clean renewable energy sources.
In considering the prospects for achieving major gains in energy efficiency, I introduce
the concept of ‘energy intensity ratios’ and review evidence on this ratio for the global
economy as well as for seven representative national economies, that is, China, the Unit
ed States, Brazil, Germany, Indonesia, South Africa, and South Korea. I will also focus on
Brazil, Germany, Indonesia, South Africa, and South Korea later in this study.
With respect to ‘clean renewable energy sources’, as I use the term, it excludes many
forms of bioenergy, such as ethanol from corn or sugarcane using conventional refining
methods. This is because, considered over a thirty-year life cycle, the emissions generat
ed from these energy sources are comparable to those from fossil fuels.2 The clean re
newable sources on which I focus in section 15.2 and throughout the study are solar and
wind power, as well as, to a more modest extent, geothermal and hydro power, as well as
bioenergy generated through low-emissions technologies.
Page 3 of 44
Section 15.3 presents a simple model through which I calculate the investment require
ments for creating a global net-zero-emissions economy as of 2050. I show through this
model that investments in energy efficiency and clean renewable energy at an average
2.5 per cent share of global GDP per year will be sufficient for achieving this end. The
model builds from the assumption of the most recent global energy model of the Interna
tional Energy Agency (IEA), which assumes that global GDP grows at an average annual
rate of 3.4 per cent per year over 2021–40.
Section 15.4 then considers the industrial and financial policy measures that will be need
ed to support this global clean energy investment project. I examine a range of (p. 397)
policy approaches that have been implemented throughout the world to varying degrees.
I also propose specific sources of funding that are capable of bringing total clean energy
investments to $2.6 trillion as of 2024—that is, 2.5 per cent of global GDP in 2024—along
with the capacity to increase funding at a rate corresponding with global GDP through
2050.
In section 15.5, I consider the domestic resource capacities in various countries to sup
port its clean energy transformation, focusing, again, on Brazil, Germany, Indonesia,
South Africa, and South Korea. To the extent that a country runs up against domestic pro
ductive capacity constraints while expanding its investments in energy efficiency and
clean renewable energy, it then must either scale back the clean energy investment
project or rely increasingly on imports to maintain the ambitious investment agenda. For
the five representative economies, I show how this domestic resource constraint will be
manageable.
One factor that will be important in enabling the expansion of domestic production in
clean energy will be the fact that the fossil-fuel sectors in all countries will be correspond
ingly contracting. Thus, in section 15.6, I show how the freeing up of economic resources
out of the activities tied to the fossil-fuel sector will be substantial in all cases, including
countries such as Germany and South Korea, which are presently dependent on imports
as their source of fossil-fuel energy.
In the concluding section 15.7, I briefly summarize the full set of findings in sections 15.2
–15.6. These findings demonstrate how a global clean energy project—that is, a Global
Green New Deal, as I understand the term—does indeed provide a viable path for achiev
ing a net-zero-emissions global economy as of 2050. I also show that the industrial and fi
nancial policy tools needed to deliver on this project are well understood and have been
well tested in various parts of the world, under a range of circumstances. These policy
tools now need to be implemented on a scale appropriate to the magnitude of the chal
lenge we now face with climate change.
This chapter covers a large number of issues within a relatively brief amount of space. At
the same time, due to space limitations, it does not cover several topics that are also criti
cal for understanding the full scope of industrial policy requirements for implementing a
successful global Green New Deal. One such critical set of issues covers the employment
impacts of the global clean energy investment project, which, in turn, breaks down into
Page 4 of 44
two components. The first is assessing the large employment creation opportunities that
will be generated through investing 2.5 per cent of GDP in clean energy projects in all re
gions of the world. The second is recognizing the job losses that will result through the
contraction of the global fossil-fuel industry, and the imperative of establishing a set of
just transition policies for both the workers and communities that will be negatively im
pacted as a result. I have addressed these issues at length elsewhere and will continue to
do so in future research.3
Related to this is the large set of questions on the developmental impact of the
(p. 398)
clean energy transition on economies that are presently net fossil-fuel exporters. These
questions are linked to the broader literature around the so-called ‘resource curse’. These
issues, again, lie beyond the scope of this chapter, even while the relevant literature is
quite extensive.4
Another set of critical issues that I have not been able to address here are the land use
requirements for building a global clean energy infrastructure. The work of the physicist
Mara Prentiss demonstrates that, in fact, through well-designed policies, these land-use
requirements will be relatively modest. But the overall global Green New Deal project will
benefit through developing with greater specificity the framework that Prentiss has de
veloped.5
15.2 What is Clean Energy?

15.2.1 Natural Gas
There are large differences in the emissions levels resulting through burning oil, coal,
and natural gas, with natural gas generating about 40 per cent fewer emissions for a giv
en amount of energy produced than coal and 15 per cent less than oil. It is therefore
widely argued that natural gas can be a ‘bridge fuel’ to a clean energy future, through
switching from coal to natural gas to produce electricity.6 Such claims do not withstand
scrutiny. At best, an implausibly large 50 per cent global fuel switch to natural gas would
reduce CO2 emissions by only 8 per cent. But even this calculation does not take account
of the leakage of methane gas into the atmosphere that results through extracting natural
gas through fracking. Recent research finds that when more than about 5 per cent of the
gas extracted leaks into the atmosphere through fracking, the impact eliminates any envi
ronmental benefit from burning natural gas relative to coal. Various studies have report
ed a wide range of estimates as to what leakage rates have actually been in the United
States, at fracking operations have grown rapidly. A recent survey paper puts that range
at between 0.18 and 11.7 per cent for different specific sites in North Dakota, Utah, Col
orado, Louisiana, Texas, Arkansas, and Pennsylvania.
It would be reasonable to assume that if fracking expands on a large scale in regions out
side the United States, it is likely that leakage rates will fall closer to the higher-end fig
ures of 12 per cent, at least until serious controls could be established. This then (p. 399)
Page 5 of 44
would greatly diminish, if not eliminate altogether, any emission-reduction benefits from a
coal-to-natural-gas fuel switch.7
15.2.2 Nuclear Energy
As of 2018, nuclear power provided 28 quadrillion British Thermal Units (Q-BTUs) of en
ergy throughout the global economy, amounting to 5 per cent of total global supply. Near
ly 90 per cent of global nuclear power supply is generated in North America, Europe, Chi
na, and India. In terms of the world reaching a net-zero CO2 emissions target by 2050, nu
clear power provides the important benefit that it does not generate CO2 emissions or air
pollution of any kind while operating. At the same time, the processes for mining and re
fining uranium ore, making reactor fuel, and building nuclear power plants all require
large amounts of energy.
But even if we put aside the emissions that result from building and operating nuclear
power plants, we still need to recognize the long-standing environment and public safety
issues associated with nuclear energy. These include:
• Radioactive wastes. These wastes include uranium mill tailings, spent reactor fuel,
and other wastes, which according to the US Energy Information Agency (EIA) ‘can re
main radioactive and dangerous to human health for thousands of years’ (EIA, 2012:
1).
• Storage of spent reactor fuel and power plant decommissioning. Spent reactor fuel
assemblies are highly radioactive and must be stored in specially designed pools or
specially designed storage containers. When a nuclear power plant stops operating,
the decommissioning process involves safely removing the plant from service and re
ducing radioactivity to a level that permits other uses of the property.
• Political security. Nuclear energy can obviously be used to produce deadly weapons
as well as electricity. Thus, the proliferation of nuclear energy production capacity cre
ates dangers of this capacity being acquired by organizations—governments or other
wise—which would use that energy as instruments of war or terror.
• Nuclear reactor meltdowns. An uncontrolled nuclear reaction at a nuclear plant can
result in widespread contamination of air and water with radioactivity for hundreds of
miles around a reactor.
Even while recognizing these problems with nuclear energy, it is still the case, as noted
above, that nuclear power presently supplies over 5 per cent of global energy supply. For
decades, the prevalent view throughout the world was that these risks associated with
nuclear power were relatively small and manageable, when balanced against its benefits.
However, this view was upended in the aftermath of the March 2011 (p. 400) nuclear melt
down at the Fukushima Daiichi power plant in Japan, which resulted from the massive 9.0
Tohuku earthquake and tsunami. The full effects of the Fukushima meltdown cannot pos
sibly be known for some time. Still, these safety considerations with nuclear energy must
be accorded significant weight. This is especially the case, given the high probability that
Page 6 of 44
the necessary tight standards for regulating nuclear power plants will become compro
mised if the number of such plants were to expand significantly on a global scale. As
such, nuclear energy cannot be seen as providing a major reliable long-term source of
non-carbon-emitting energy supplies.
15.2.3 Geoengineering
This includes a broad category of measures whose purpose is either to remove existing
CO2 or to inject cooling forces into the atmosphere to counteract the warming effects of
CO2 and other greenhouse gases. One broad category of removal technologies is carbon
capture and sequestration (CCS). A category of cooling technologies is stratospheric
aerosol injections (SAI).
CCS technologies aim to capture emitted carbon and transport it, usually through
pipelines, to subsurface geological formations, where it would be stored permanently.
One straightforward and natural variation on CCS is afforestation. This involves increas
ing forest cover or density in previously non-forested or deforested areas, with ‘reforesta
tion’—the more commonly used term—as one component.
The general class of CCS technologies has not been proven at a commercial scale, despite
decades of efforts to accomplish this. A major problem with most CCS technologies is the
prospect for carbon leakages that would result under flawed transportation and storage
systems. These dangers will only increase to the extent that CCS technologies are com
mercialized and operating under an incentive structure in which maintaining safety stan
dards will reduce profits. By contrast, afforestation is, of course, a natural and proven
carbon removal technology. At the same time, most deforestation projects throughout the
globe were undertaken to make space for raising crops and livestock. Relying heavily on
afforestation as a climate change strategy would therefore likely present serious land-use
competition problems.
The idea of stratospheric aerosol injections builds from the results that followed from the
volcanic eruption of Mount Pinatobo in the Philippines in 1991. The eruption led to a mas
sive injection of ash and gas, which produced sulphate particles, or aerosols, which then
rose into the stratosphere. The impact was to cool the earth’s average temperature by
about 0.6°C for fifteen months.8 The technologies being researched now aim to artificially
replicate the impact of the Mount Pinatubo eruption through deliberately injecting sul
phate particles into the stratosphere. Some researchers (p. 401) contend that to do so
would be a cost-effective method of counteracting the warming effects of greenhouse gas
es.
Lawrence et al. (2018) published an extensive review on the range of climate geoengi
neering technologies, including 201 literature references. Their overall conclusion from
this review is that none of these technologies are presently at a point at which they can
make a significant difference in reversing global warming. They conclude:
Page 7 of 44
Proposed climate geoengineering techniques cannot be relied on to be able to

make significant contributions…towards counteracting climate change in the con
text of the Paris Agreement. Even if climate geoengineering techniques were ac
tively pursued, and eventually worked as envisioned on global scales, they would
very unlikely be implementable prior to the second half of the century…This would
very likely be too late to sufficiently counteract the warming due to increasing lev
els of CO2 and other climate forcers to stay within the 1.50C temperature limit—
and probably even the 2°C limit—especially if mitigation efforts after 2030 do not
substantially exceed the planned efforts of the next decade (2018: 13–14).9
15.2.4 Energy Efficiency and Clean Renewable Energy
Given these major problems with natural gas as a ‘bridge fuel’, nuclear energy, and geo
engineering, it follows that we focus instead on the most cautious clean energy transition
programme, that is, investing in technologies that are well understood, already operating
at large scale, and, without question, safe. In short, we focus on investments that can dra
matically raise energy-efficiency standards and equally dramatically expand the supply of
clean renewable energy sources.
15.2.5 Energy Efficiency
Energy efficiency entails using less energy to achieve the same, or even higher, levels of
energy services from the adoption of improved technologies and practices. Examples in
clude insulating buildings much more effectively to stabilize indoor temperatures; driving
more fuel-efficient cars or, better yet, relying increasingly on well-functioning public
transportation systems; and reducing the amount of energy that is wasted both through
generating and transmitting electricity and through operating industrial machinery.
Expanding energy-efficiency investments support rising living standards because raising

energy-efficiency standards, by definition, saves money for energy consumers. (p. 402) A
major 2010 study by the US National Academy of Sciences (NAS) found, for the US econ
omy, that ‘energy-efficient technologies…exist today, or are expected to be developed in
the normal course of business, that could potentially save 30 per cent of the energy used
in the US economy while also saving money’. Similarly, a 2010 McKinsey & Company
study, focused on developing countries, found that, using existing technologies only, ener
gy-efficiency investments could generate savings in energy costs in the range of 10 per
cent of total GDP for all low- and middle-income countries.
In her 2015 book, Energy Revolution: The Physics and Promise of Efficient Technology,
Mara Prentiss argues, further, that such estimates understate the realistic savings poten
tial of energy-efficiency investments. This is because, in generating energy by burning
fossil fuels, about two-thirds of the total energy available is wasted while only one-third is
available for powering machines. By switching to renewable energy sources, the share of
wasted energy falls by 50 per cent. This is what Prentiss terms the ‘burning bonus’.
Page 8 of 44
After taking account of the burning bonus as well as the efficiency gains available in the
operations of buildings, transportation systems, and industrial equipment, Prentiss con
cludes, with respect to the US economy specifically, that economic growth could proceed
at a normal rate while total energy consumption could remain constant or even decline in
absolute terms. Prentiss’s conclusions regarding the US economy are consistent with the
most recent projections for global energy demand by the International Energy Agency
(IEA, 2019). As I discuss further in section 15.3.1, the IEA assumes that the global econo
my will grow at a 3.4 per cent average annual rate between 2018 and 2040. Neverthe
less, under their most conservative Current Policies Scenario, the IEA assumes that glob
al energy consumption will grow at a much slower 1.3 per cent per year. Under their
more ambitious Sustainable Development Scenario, they assume that global energy con
sumption will actually fall at an average rate of –0.3 per cent per year, while economic
growth still proceeds at a 3.4 per cent average rate.10
A useful way to measure the relationship between the level of economic activity and the
energy resources consumed to support that activity is the energy intensity ratio. The en
ergy intensity ratio is, straightforwardly, the level of total energy resources consumed in
any given economy divided by the economy’s GDP. I report in Table A15.1 (in the Appen
dix) below the most recent energy intensity figures for the world economy as well as for
seven representative large economies—China, the United States, Brazil, Germany, In
donesia, South Africa, and South Korea.
In section 15.2.6, I will focus on this ratio for the world economy as a key variable for es
timating the costs of reaching a zero CO2 emissions global economy by 2050. For now, it
will be useful to consider the patterns for the global economy and the respective national
economies. The units in which I measure the ratio are Q-BTUs of energy consumed/tril
lion dollars of GDP. As the table shows, the intensity ratio, as of 2018, was 6.6 Q-BTUs for
every $1 trillion of global GDP. With the individual country (p. 403) figures, we see that the
intensity ratios vary widely by country. Germany is the most efficienct economy, with the
lowest 3.5 intensity ratio. The United States is next, at with a 5.3 intensity ratio, following
by Brazil at 7.2. South Korea and Indonesia are at similar efficiency levels, with 9.3 and
9.8 intensity ratios respectively. China is operating at a relatively low efficiency level,
with a 12.7 intensity ratio. South Africa is the least energy efficient of the countries in our
sample, with an intensity ratio of 20.3.11
15.2.6 Estimating Costs of Efficiency Gains
This range of energy intensity figures notwithstanding, the aim of the clean energy in
vestment project will be to achieve dramatic improvements in efficiency in all national
economies across the global economy. The question we therefore need to address is: how
much will it cost to achieve such large-scale efficiency gains?
In fact, estimates as to the investment costs for achieving energy-efficiency gains vary
widely. In Table A15.2 in the Appendix, I show summary estimates from three sets of stud
ies. As we see, the 2008 World Bank study by Taylor et al. puts average costs at $1.9 bil
Page 9 of 44
lion per Q-BTU of energy savings, based on a study of 455 projects in both industrial and
developing economies. A 2010 study by McKinsey estimates costs for a wide range of
non-OECD economies at $11 billion per Q-BTU of energy savings. Focusing just on the US
economy, NAS estimated average costs for energy-efficiency savings in the buildings and
industrial sectors at about $29 billion per Q-BTU.12
It is not surprising that average costs to raise energy-efficiency standards would be sig
nificantly higher in industrialized economies. A high proportion of overall energy-efficien
cy investments are labour costs, especially projects to retrofit buildings and industrial
equipment. However, these wide differences in cost estimates shown in Table A15.2 do
not simply result from variations in labour and other input costs by regions and levels of
development.
Thus, the World Bank estimate of $1.9 billion per Q-BTU includes both industrialized and
developing countries, while the McKinsey $11 billion per Q-BTU estimate—nearly six
times greater than the World Bank figure—is primarily coming from developing-country
projects. These alternative studies do not provide sufficiently (p. 404) detailed method
ological discussions that would enable us to identify the main factors generating these
major differences in cost estimates. But it is at least reasonable to conclude from these
figures that, with on-the-ground real-world projects, there are likely to be large variations
in costs down to the project-by-project level. Thus, the costs for energy-efficiency invest
ments that will apply in any given situation will necessarily be specific to that situation,
and must always be analysed on a case-by-case basis. At the same time, for our present
purposes, we need to proceed with some general rules of thumb for estimating the level
of savings that is attainable through a typical set of efficiency projects in various regions
of the world, and more precisely an aggregated estimate for the global economy.
A conservative approach will be to allow that, relative to the World Bank and US National
Academy of Sciences figures, the mid-range cost estimate provided by McKinsey at $11
billion per Q-BTU of savings, is appropriate for low- and middle-income economies, such
as Brazil, Indonesia, and South Africa. Along the same lines, we could assume that the
cost figure for Germany will be equivalent to what the NAS study estimated for the Unit
ed States, at around $30 billion per Q-BTU of savings. The South Korean economy would
then be an approximate midpoint between those two other figures, at around $20 billion
per Q-BTU. As a working approximation for the global economy, this same midpoint figure
of $20 billion per Q-BTU of savings should be a credible high-end estimate, especially
while recognizing that the World Bank estimate for 455 projects in both developing and
advanced economies is ten times higher, at about $2 billion per Q-BTU of savings.
15.2.7 Rebound Effects
Raising energy-efficiency levels will generate ‘rebound effects’—i.e. energy consumption

increases resulting from lower energy costs. But such rebound effects are likely to be
modest within the current context of a global project focused on reducing CO2 emissions
and stabilizing the climate. Among other factors, energy consumption levels in advanced
Page 10 of 44
economies are close to saturation points in the use of home appliances and lighting—that
is, we are not likely to clean dishes much more frequently because we have a more effi
cient dishwasher. The evidence shows that consumers in advanced economies are likely
to heat and cool their homes as well as drive their cars more when they have access to
more efficient equipment. But these increased consumption levels are usually modest. Av
erage rebound effects are likely to be significantly larger in developing economies.13
But it is critical that all energy-efficiency gains will be accompanied by complimentary

policies (as discussed in section 15.4.1), including setting a price on carbon emissions to
discourage fossil-fuel consumption. Most significantly, expanding the (p. 405) supply of
clean renewable energy will allow for higher levels of energy consumption without lead
ing to increases in CO2 emissions. It is important to recognize, finally, that different coun
tries presently operate at widely varying levels of energy efficiency. For example, as we
saw in Table A15.1, Germany presently operates at an efficiency level roughly 50 per cent
higher than that of the United States. Brazil is at an efficiency level that is nearly three
times that of South Africa. There is no evidence that large rebound effects have emerged
as a result of these high efficiency standards in Germany and Brazil relative to those of
the United States and South Africa.
15.2.8 Renewable Energy
A critical point for building a net-zero global economy by 2050 is the fact that, on aver
age, the costs of generating electricity with clean renewable energy sources are now at
parity or lower than those for fossil-fuel-based electricity. Table A15.3 in the Appendix
shows the most recent figures reported by the International Renewable Energy Agency
(IRENA), for 2010 and 2017, on the ‘levellized costs’ of supplying electricity through al
ternative energy sources. Levellized costs takes account of all costs of producing and de
livering a kilowatt of electricity to a final consumer. The cost calculations begin with the
upfront capital expenditures needed to build the generating capacity, continue through to
the transmission and delivery of electricity, and include the costs of energy that is lost
during the electricity-generation process.
As we see in Table A15.3, the levellized costs for fossil-fuel-generated electricity range be
tween 4.5 and 14 cents per kilowatt hour as of 2017. The average figures for the four
clean renewable sources are all within this range for fossil fuels as of 2017, with hydro at
5 cents, onshore wind at 6 cents, geothermal at 7 cents and solar PV at 10 cents. The
costs of geothermal and hydro did not fall, and actually rose modestly, between 2010 and
2017. However, the costs of onshore wind fell by 25 per cent, from 8 to 6 cents. The most
impressive result though is with solar PV, in which levellized costs fell by 72 per cent
from 2010 to 2017, from 36 cents to 10 cents per kilowatt hour. These average cost fig
ures for solar and wind should continue to decline by significant amounts as advances in
technology and economies of scale proceed along with the rapid global expansion of these
sectors.
Page 11 of 44
We emphasize that these cost figures from the IRENA are simple averages. They do not
show differences in costs due to regional or seasonally specific factors.14 In particular, so
lar and wind energy costs will vary significantly by region and season. Moreover, both
wind and solar energy are intermittent sources—that is, they only generate energy, re
spectively, when the sun is shining or the wind is blowing. Of course, the central role of
energy storage systems to address these matters will need to be fully (p. 406) accounted
for when clean renewable energy systems are designed to provide a major share of an
economy’s overall energy load.15
Keeping all such considerations in mind, we can still roughly conclude from these figures
that, for the most part, clean renewable energy sources are rapidly emerging into a posi
tion at which they can produce electricity at comparable or lower costs than non-renew
able sources. As such, assuming that wind, solar, and geothermal energy production can
be scaled up to meet virtually all global demand by 2050, then the costs to consumers of
purchasing this energy should not be significantly different from what these consumers
would have paid for non-renewable energy. Indeed, overall, the costs to consumers of pur
chasing electricity from clean renewable sources, including hydro as well as wind, solar,
and geothermal power, are likely to be lower than what they would be from fossil-fuel
sources. It is critical to also emphasize that this is without factoring in the environmental
costs of burning oil, coal, and natural gas.
15.2.9 Costs of Expanding Renewable Capacity
By a substantial amount, the largest share of overall costs in generating electricity from
renewable sources are capital costs—that is, the costs of producing new productive
equipment, as opposed to the costs of operating that productive equipment once it has
been built and is generating energy. These capital costs are at about 65 per cent of total
costs for geothermal, 73 per cent for onshore wind, and 81 per cent for solar PV.16 From
these figures on levellized costs, we can also estimate the capital costs of installing re
newable energy capacity as a lump sum—that is, how much investors need to spend up
front to put this capital equipment into place and in running order.
I produce estimates of these lump sum capital costs in Table A15.4 in the Appendix.
Specifically, these figures represent the present values of total lump-sum capital expendi
tures needed to produce one Q-BTU of electricity from onshore wind, solar PV, and geot
hermal energy.17 As we see, the average lump-sum costs range from $112 billion per Q-
BTU for geothermal, $160 billion for onshore wind, and $190 billion for solar.
If we assume that, roughly speaking, the global expansion of clean renewable energy ca
pacity will consist of 45 per cent from wind and solar PV technologies, and 10 per cent
from geothermal energy, this would place the average costs of producing one Q-BTU of
overall renewable energy equipment at about $169 billion, which we can round up to
$170 billion per Q-BTU of clean renewable capacity. This $170 billion figure can therefore
serve as a benchmark for estimating the average costs of expanding the supply of clean
renewable energy on a global scale. At the same time, (p. 407) as with our cost estimate
Page 12 of 44
for investments in energy efficiency, we will want to err, if anything, on the side of overes
timating, rather than underestimating, the costs of expanding clean renewable energy.
Moreover, with the expansion of the globe’s clean energy supply proceeding rapidly over
2024–50, the average costs are likely to rise as production bottlenecks emerge. We there
fore will assume that the average costs of expanding the clean energy supply will be $200
billion per Q-BTU, that is, about 18 per cent higher than the $170 billion average figure
we have derived from the levellized costs data.
We can now work with our two rough high-end estimates of the overall costs of both rais
ing energy-efficiency standards and building new clean renewable energy capacity—$20
billion per Q-BTU for efficiency gains and $200 billion per Q-BTU for expanding renew
able capacity—to generate an estimate of the total costs of achieving a net-zero global
economy by 2050.
15.3 Economic Growth and Emissions Reduc

tions
In this section, I present a simple model to illustrate how the global economy can achieve
net-zero CO2 emissions by 2050 through investing about 2.5 per cent of global GDP per
year to raise energy-efficiency standards and to expand the supply of clean renewable en
ergy sources. The model works from the following assumptions:
1. Average costs for increasing energy efficiency and expanding clean renewable
production. As discussed, I assume that the average costs to increase energy efficien
cy by 1 Q-BTU will be $20 billion. I also assume that the average costs to expand pro
ductive capacity of clean renewable energy by 1 Q-BTU will be $200 billion.
2. Global GDP growth trend. The IEA’s forecast assumes an average global GDP
growth rate of 3.4 per cent between 2018 and 2040 (2019: 753). My model incorpo
rates this figure. To date, the IEA has not published a global GDP growth forecast
that extends beyond 2040. For the purposes of the current exercise, I assume that
the 3.4 per cent average global GDP growth rate will extend to 2050.
3. Clean renewable energy sources supply 100 per cent of global energy demand. As
discussed, there may be a case for relying to a limited extent on nuclear energy and
some types of carbon capture technologies beyond afforestation as a supplement to
clean renewable sources. But this model demonstrates how, as of 2050, it will be cost
effective as well as technically feasible to deliver 100 per cent of global energy sup
ply through clean renewables.
4. Three-year delay in bringing the project to scale. This is a thirty-year investment
project. But given that the current level of clean energy investments is in the range
(p. 408) of 0.7 per cent of global GDP,18 we must realistically allow for some incuba
tion time to pass before we can expect investments to rise by 1.8 percentage points
as a share of GDP, to a 2.5 per cent of annual GDP level. To reflect this consideration,
I assume, as noted above, that it will require three years of major initiatives within
the realms of industrial policy and financing to raise global clean energy investments
Page 13 of 44
by roughly 1.8 per cent of GDP relative to current investment levels. We therefore as
sume that the 2.5 per cent of GDP per year level of clean energy investments will oc
cur over twenty-seven years within the full thirty-year investment cycle, that is, be
tween 2024 and 2050. The initial three years of the model, 2021–3, will be needed to
develop an adequate industrial policy and financing environment to sustain clean en
ergy investments at this level.
The results of this model are presented in Tables 15.1–15.4.
15.3.1 Global Model Framework and Calculations
In Table 15.1, we start with the actual global GDP figure in 2018 of $86 trillion.19 We then
work with the IEA’s assumption of average global GDP growth over the subsequent thirty
years at 3.4 per cent per year. From our initial 2018 GDP figure of $86 trillion and our as
sumption of 3.4 per cent average annual growth, we can then estimate the level of GDP
every year through 2050. Under these assumptions, global GDP will be $260 trillion in
2050. The model projects global GDP in 2050 at $260 trillion. We can also then calculate
the ‘midpoint’ GDP figure over the 2021–50 thirty-year investment cycle. I define this
midpoint figure as being equal to the average of the estimates of GDP in 2021 and 2050,
assuming average annual GDP growth at 3.4 per cent. This midpoint figure, as we see, is
$178 trillion. From this midpoint figure, we can then readily calculate a ratio for average
annual clean renewable investments.
Page 14 of 44
Table 15.1 Thirty-year global GDP growth trajectory, 2021–50
2018 global GDP $86 tril

lion
Projected average annual GDP growth rate through 3.4%

2040 (from IEA, 2019: 753)
Projected 2021 GDP (with 3.4% average annual GDP $95 tril
growth) lion
Projected 2024 GDP (First year of 27-year investment $104 tril

cycle; with 3.4% average annual GDP growth) lion
Projected 2050 GDP (with 3.4% average annual GDP $260 tril
growth) lion
Midpoint GDP value for investment spending estimates $178 tril

(= (2021 GDP + 2050 GDP)/2) lion
Sources: World Development Indicators; International Energy Agency,

World Energy Outlook, 2019.
In Table 15.1, I estimate the level of clean energy investments necessary to bring the av
erage global energy intensity ratio down from its current level of 6.6 (Q-BTUs/ (p. 409) tril
lion dollars of GDP) to 2.0, a 70 per cent improvement in average global energy efficien
cy. The 2.0 energy intensity ratio is the figure projected by the IEA in its 2019 Sustain
able Development Scenario. This will be while, according to the IEA model, average glob
al GDP is growing at 3.4 per cent per year.
Page 15 of 44
Table 15.2 Global energy demand and energy-efficiency cost projec

tions for 2050
A) Total energy demand through alternative scenarios
2050 En Average annu Average

ergy con al energy de 2050 global
sumption mand growth energy inten
rate sity ratio
2050 energy 1,716 Q- 3.4% 6.6

demand, with BTUs
constant ener
gy intensity ra
tio
2050 energy 512 Q- –0.3% 2.0

demand BTUs
through IEA
Sustainable De
velopment Sce
nario
B) Cost of achieving energy savings through the IEA sustain

able development scenario
1. Total energy savings through the IEA 1,204 Q-BTUs (= 1,716

Sustainable Development Scenario Q-BTUs – 512 Q-BTUs)
2. Average cost of energy savings $20 billion/Q-BTU

through efficiency investments
3. Total cost of energy savings through $24.1 trillion (= rows 2 x

efficiency investments 3)
4. Average annual cost of energy sav $891 billion ($24.1 tril

ings through efficiency investments lion/27 years)
Notes: Actual 2018 global energy consumption = 568 Q-BTUs; global

energy intensity ratio = 6.6.
Source: IEA (2019: 678).
Page 16 of 44
As section A of Table 15.2 shows, if the global economy continues to operate at its current
6.6 energy intensity ratio through 2050, global energy consumption will be at 1,716 Q-
BTUs in 2050. By contrast, if the global economy does succeed in driving down the ener
gy intensity ratio to 2.0 through efficiency investments, it follows that global energy con
sumption will be at 512 Q-BTUs as of 2050.
As we then see in section B of Table 15.2, total energy savings achieved through operat
ing the global economy at a 2.0 rather than a 6.6 average intensity ratio will be 1,204 Q-
BTUs. Since, as our high-end figure, we assume that the average global cost of achieving
efficiency gains is $20 billion per Q-BTU, this means that achieving 1,204 Q-BTUs in glob
al efficiency gains will cost a total of $24.1 trillion. As Panel B of Table 15.2 shows, the av
erage annual investment level of the twenty-seven-year investment period is therefore
$891 billion.
In Table 15.3, we work with the 2050 global energy consumption figure of 512 Q-BTUs
from Table 15.2 to calculate the investment requirements for meeting this level of total
energy demand through clean renewable sources. As the table shows, as of the most re
cent IEA figures, global supply of clean renewables is 26 Q-BTUs. This means that the ex
pansion of supply as of 2050 will need to be 486 Q-BTUs. It also means that the average
growth rate for expanding the global supply of clean renewable energy will need to be at
around 10 per cent per year for the full 2024–50 investment cycle.
Table 15.3 Global clean renewable energy expansion and cost projec
tion for 2050
1. Total 2050 energy consumption through the 512 Q-BTUs

Sustainable Development Scenario
2. 2018 clean renewable energy supply (from IEA, 26 Q-BTUs

2019: 678)
3. Net expansion of clean renewables as of 2050 486 Q-BTUs (=

row 1 – row 2)
4. Average cost of expanding clean renewable $200 billion/Q-

supply BTU
5. Total cost of expanding global clean renewable $97.2 trillion (=

supply by 486 Q-BTUs as of 2050 row 3 x row 4)
Average annual cost of expanding global clean re $3.6 trillion (=

newable supply by 286 Q-BTUs as of 2050 row 5/27)
Source: IEA (2019).
Page 17 of 44
In terms of estimating the costs of this investment project, I then, again, assume a high-
end average cost figure for expanding global clean energy capacity, at $200 billion per Q-
BTU. Working from this figure, it follows, as shown in Table 15.3, that the total costs of
expanding global clean energy supply by 486 Q-BTUs as of 2050 will be $97.2 trillion. The
average annual costs over the twenty-seven-year investment cycle will therefore be $3.6
trillion.
In Table 15.4, I then summarize the figures for total and annual average costs for achiev
ing a net-zero global economy strictly on the basis of large-scale investments in (p. 410)
energy efficiency and clean renewables. As we see, total costs come to $121.3 trillion.
Over the twenty-seven-year investment cycle, this amounts to an average of $4.5 trillion
per year. Working from the estimates presented in Table 15.1, our figure for midpoint
global GDP between 2021 and 2050 is $178 trillion. This is how, finally, we are able to es
timate that the overall investment requirement for reaching a net-zero-emissions global
economy as of 2050 will amount, on average, to 2.5 per cent of global GDP per year.
Page 18 of 44
Table 15.4 Costs of thirty-year clean energy investment project as

share of average GDP, 2021–50
Total costs of clean energy investments
1. Energy efficiency $24.1 tril

lion
2. Clean renewable energy $97.2 tril

lion
3. TOTAL (= rows 1 + 2) $121.3 tril

lion
2024 costs of clean energy investments (year 1 of 27-

year investment cycle)
4. Energy efficiency $500 billion
5. Clean renewable energy $2.1 trillion
6. TOTAL (= rows 5 + 6) $2.6 trillion
Average annual costs of clean energy investments

(27-year cycle)
7. Energy efficiency $891 billion/

year
8. Clean renewable energy $3.6 trillion/

year
9. TOTAL (= rows 7 + 8) $4.5 trillion/

year
Total costs of clean energy investments as share of

midpoint GDP
10. Midpoint GDP (from Table 15.1) $178 trillion
11. Clean energy investments as share of midpoint 2.5%

GDP (= row 6/row 7)
Page 19 of 44
Source: IEA (2019).
15.4 Industrial and Financial Policies

15.4.1 Industrial Policies
Depending on specific conditions within each country, industrial policies will be needed to
promote technical innovations and, even more broadly, adaptations of existing clean ener
gy technologies. Again depending on circumstances, governments will need to deploy a
combination of industrial policy instruments, including research and development sup
port, preferential tax treatment for clean energy investments, and government procure
ment policies. Clean energy industrial policies will also need to include regulations of
both fossil-fuel and clean energy prices as well as emission standards.
One major policy intervention that can facilitate the creation of a vibrant clean energy
market will be for governments to themselves become both large-scale investors in ener
gy efficiency and purchasers of clean renewable energy. An important comparable histori
cal experience was the development of the Internet within the US military, beginning in
the 1940s. In the process of bringing the Internet to commercial scale, the US military
provided a guaranteed market for thirty-five years, which enabled (p. 412) the technology
to incubate while private investors gradually developed effective commercialization
strategies.20
But guaranteeing stable prices with the private-sector purchases of clean renewables is
also critical here. Such policies are termed feed-in tariffs. Specifically, these are contracts
that require utility companies to purchase electricity from private renewable energy gen
erators at prices fixed by long-term contracts. Feed-in tariffs were first implemented in
the United States in the 1970s, and a number of state and local programmes are current
ly operational in the United States today. However, the impact of feed-in tariffs has been
much more significant outside of the United States, especially in Germany, Italy, France,
Spain, and Canada. A 2009 study by the US Department of Energy found that these poli
cies in Europe ‘resulted in quick and substantial renewable energy capacity expansion’.21
This basic result has been affirmed through more recent research, including that by Mi
lanes-Montero et al. (2018), which showed how feed-in tariffs ‘have had a significant posi
tive influence on the economic profitability’ on solar PV companies in Europe. The key
factor in the success of these European programmes is straightforward: the guaranteed
prices for renewable energy were set to adequately reflect the costs of producing the en
ergy along with a profit for the energy provider. This then encouraged private renewable
energy investors by providing a stable long-term market environment.22
Feed-in tariffs have also had some successes as a policy tool in Africa. The African Devel
opment Bank reports as follows:
Page 20 of 44
Public investment is critical in bridging the gap between public demonstration of

new technologies and mature deployment. Feed-in tariffs are a prominent example
of such subsidies. These tariffs are a policy mechanism that offers compensation
to renewable energy producers, based on the difference between the cost of elec
tricity generation of each technology and the market price of electricity genera
tion that, in the case of RETs, is usually lower. In Kenya, for example, feed-in tar
iffs led to the high level of uptake of solar PV. As of 2011, 7 African countries used
feed-in tariff policies.
(African Development Bank Group, 2013)
Another important set of policies are those that aim to directly reduce fossil-fuel con
sumption. These include carbon caps and carbon taxes. In principle at least, a carbon cap
establishes a firm limit on the allowable level of emissions for major polluting entities,
such as utilities. Such measures will also raise the prices of oil, coal, and natural gas by
limiting their supply. A carbon tax, on the other hand, will directly raise fossil-fuel prices
to consumers, and aim to reduce fossil-fuel consumption through the resulting price sig
nals. Either approach can be effective as long as the cap is strict enough, or tax rate high
enough, to significantly reduce fossil-fuel consumption and as long exemptions are mini
mal to none. Raising the prices for fossil fuels will also, of (p. 413) course, create in
creased incentives for both energy-efficiency and clean renewable investments, as well as
a source of revenue to help finance these investments. We return to this point in section
15.4.
However, significant problems are also associated with both approaches. Establishing a
carbon cap or tax will have negative distributional consequences that will need to be ad
dressed in the policy design. All else equal, increasing the price of fossil fuels would af
fect lower-income households more than affluent households, since petrol, home-heating
fuels, and electricity absorb a higher share of lower-income households’ consumption. An
effective solution to this problem is to rebate to lower-income households a significant
share of the revenues generated either by the cap or tax to offset the increased costs of
fossil-fuel energy.23
Renewable energy portfolio standards for utilities, and energy-efficiency standards for
buildings and transportation vehicles, are similar in their intent to a carbon cap. That is,
renewable portfolio standards set a minimum standard that utilities must achieve in gen
erating electricity from renewable energy sources. Energy-efficiency standards for auto
mobiles set minimum miles-per-gallon levels (or comparable measures) that a given auto
fleet must achieve to be in compliance with the law. Comparable efficiency standards can
also be established for buildings in terms of allowable levels of energy consumption for a
given building size.
However, a major problem that has emerged with carbon caps as well as renewable and
efficiency standards has been with enforcement. As a major case in point, when these cap
programmes are combined with a carbon permit option—as in ‘cap-and-trade’ policies—
Page 21 of 44
the enforcement of a hard cap becomes difficult to sustain or even monitor, thereby weak
ening the impact of the policy.24
15.4.2 Providing Cheap and Accessible Financing
There are two separate, but interrelated policy considerations here. The first is: where
will the funding come from to support approximately $2.6 trillion in new clean energy in
vestments in 2024 and $3.6 trillion as an annual average over 2021–50? The second issue
is: how can these funds be most effectively channelled into the full range of specific
projects that will need to advance every year in order to build a net-zero global economy?
We consider these issues in turn.
(p. 414) 15.4.3 Sources of Aggregate Funding
In principle, it should not be especially challenging to solve this problem. To begin with,
as of 2018, Credit Suisse estimates that the total value of global financial assets was $317
trillion.25 The $2.4 trillion that I am proposing to channel into clean energy investments
as of 2021 amounts to 0.7 per cent of this total financial asset pool.
Still, it is important to anchor the discussion in specific proposals. Therefore, for purpos
es of illustration, I propose four large-scale funding sources to support public investments
in clean energy. Other approaches could also be viable. These four funding sources are:
1) a carbon tax, in which 75 per cent of revenues are rebated back to the public but 25
per cent are channelled into clean energy investment projects; 2) transferring funds out
of military budgets from all countries, but primarily the United States; 3) a Green Bond
lending programme, initiated by both the US Federal Reserve and the European Central
Bank; and 4) eliminating all existing fossil-fuel subsidies and channelling 25 per cent of
the funds into clean energy investments. Strong cases can be made for each of these
funding measures. But each proposal does also have vulnerabilities, including around po
litical feasibility. The most sensible approach is therefore to combine the measures into a
single package that minimizes their respective weaknesses as standalone measures. Ta
ble A15.5 in the Appendix presents this set of combined proposals in summary form.
1. Carbon tax with rebates. As noted above, carbon taxes have the merit of impacting cli
mate policy through two channels—they raise fossil-fuel prices and thereby discourage
consumption while also generating a new source of government revenue. At least part of
the carbon tax revenue can then be channelled into supporting the clean energy invest
ment project. But the carbon tax will hit low- and middle-income people disproportionate
ly, since they spend a larger fraction of their income on electricity, transportation, and
home-heating fuel. An equal-shares rebate, as proposed by Boyce (2019), is the simplest
way to ensure that the full impact of the tax will be equalizing across all population co
horts.
Consider, therefore, the following tax-and-rebate programme. Focusing, again, on 2024,

the first year of the full-scale investment programme, we begin with a tax at a low rate of
$20 per ton of carbon. Given current global CO2 emissions levels, that would generate
Page 22 of 44
about $625 billion in revenue. If we use only 25 per cent of this revenue to finance clean
energy investments, that amounts to roughly $160 billion for investment projects. The 75
per cent of the total revenue that is rebated to the public in equal shares would then
amount to $465 billion. This amounts to about $60 for every person on the planet, or
$240 for a family of four.26 (p. 415)
2. Transferring funds out of military budgets. Global military spending in 2018 was at
$1.8 trillion.27 The US military budget, at about $700 billion, accounted for nearly 40 per
cent of the global total. There are solid logical and ethical grounds for transferring sub
stantial shares of each country’s total military budget to supporting climate stabilization,
if we take at face value the idea that military spending is fundamentally aimed at achiev
ing greater security for the citizens of each country. But to remain within the realm of po
litical feasibility, let us assume that 5 per cent of global military spending will transfer in
to supporting climate security. That would amount to $90 billion.
3. Green Bond funding by the Federal Reserve and European Central Bank. It was demon
strated during the 2007–09 global financial crisis and subsequent Great Recession that
the Federal Reserve is able to supply basically unlimited bailout funds to private financial
markets during crises. The extensive 2017 study, The Costs of the Crisis, by Better Mar
kets concludes that the Federal Reserve committed approximately $12.2 trillion to stop
the crash of the financial system, stabilize the economy, and try to spur economic growth.
I would propose $100 billion in Green Bond financing supplied by the Fed. This would
amount to a miniscule 0.8 per cent of the Fed’s 2007–09 bailout operations during the cri
sis. The Fed’s funding support could be injected into the global economy through straight
forward channels. That is, various public entities, such as the World Bank, could issue
long-term zero interest rate Green Bonds. The Fed would purchase these bonds. The vari
ous public entities issuing these bonds would then have the funds to pursue the full range
of projects that will fall under the rubric of the global clean energy project.
This framework has not yet been introduced into policy discussions at the Federal Re
serve. But they are becoming a central area of focus at the European Central Bank. Thus,
the Financial Times reported on 12/2/19 that the recently installed ECB President Chris
tine Lagarde is moving quickly on the matter. The Financial Times reports that:
Christine Lagarde…is pushing to include climate change considerations in a re

view the central bank is due to hold into the way it conducts monetary policy. Un
til now, the expectation was for a review into purely monetary matters, such as
whether the inflation target should be revised. An explicit focus on climate change
policy would be a huge move. Because the central bank is by far the biggest influ
ence on financial conditions in the market, it can make a significant difference to
investment decisions that determine how Europe’s climate transition goes.28
The Financial Times article makes clear that the specific channels through which the ECB
would intervene to support clean energy financing will require substantial fleshing out.
The type of approach I have sketched for a Federal Reserve intervention would seem like
a relatively straightforward and modest form of intervention. I therefore propose that the
Page 23 of 44
ECB undertakes Green Bond purchases at the same level as the Federal (p. 416) Reserve,
that is, at $100 billion as of 2024, and growing over time to support clean energy invest
ments continuing at an average rate of 2.5 per cent of global GDP per year.
4. Eliminating fossil fuel subsidies and channelling 25 per cent of funds to clean energy
investments. One recent estimate of direct fossil-fuel subsidies to consumers—measured
as the difference between supply and consumer prices to purchase fossil-fuel energy—is
about $3 trillion globally as of 2015, or about 0.4 per cent of global GDP.29 Channelling
these funds, in full, into supporting public clean energy investments would therefore
more than pay for the $2.6 trillion estimate for total clean energy investments as of 2024.
This $3 trillion would also represent more than double the amount necessary to cover a
global public investment level of $1.3 trillion. However, such fossil-fuel subsidies are
largely used as a form of general support for all energy consumers. Lower- and middle-in
come households are therefore major beneficiaries of these subsidies, along with, of
course, the fossil fuel energy suppliers. Therefore, in terms of global income distribution,
eliminating these subsidies altogether would likely have a significant regressive impact,
comparable to establishing a carbon tax without an accompanying rebate programme. As
such, to continue to provide support for lower-income households, most of the funds that
are now being channelled to these households through fossil-fuel subsidies should be
redirected into either supporting lower consumer prices for clean energy or to provide di
rect income transfers for lower-income households.
Given that we will have raised $440 billion from the carbon tax, military spending trans
fers and central bank Green Bond programmes, we could then assume that 25 per cent of
the $3 trillion received as fossil-fuel subsidies be transferred into the clean energy invest
ment fund. That would amount to $750 billion. With these funds, we will have reached the
total $1.3 trillion in public investment funds necessary to attain the total of public and
private investment spending of $2.6 trillion as of 2024.
15.4.4 Channelling Financial Resources into Specific Investment

Projects
Both general purpose development banks as well as special-purpose green development

banks are already significantly engaged in financing clean energy investments. It will be
crucial to build from these efforts to achieve the necessary level of financing for clean en
ergy investments.
Germany’s KfW Bank. The case of Germany is instructive, since it has been the
(p. 417)
most successful large advanced economy to date in developing its clean energy economy.
The publicly owned development bank in Germany, KfW, has been critical to this success.
Griffith-Jones (2016) considers KfW’s impact on Germany’s overall green transformation,
including renewable energy as well as energy-efficiency investments. She finds that KfW
has underwritten roughly one-third of all financing for green investments in Germany.
KfW has thus been instrumental in moving policy ideas into effective investment projects,
with respect to both energy efficiency and clean renewables. KfW has also been highly ac
Page 24 of 44
tive in financing green investment projects elsewhere in Europe and in developing coun
tries. As Griffith-Jones writes:
KfW plays a key role, domestically and internationally, in supporting energy revo
lution, through funding major investments in renewable energy and in energy effi
ciency. In the national German case, this was to a large extent implemented within
a clear institutional and policy framework, namely the renewable energy law,
through strong policy measures, such as feed in tariffs (FITs) and reverse competi
tive auctions, which made investment in renewables commercially attractive. A
similar modus operandi existed for energy efficiency…The combination of clear
government policies and associated development bank targets has produced very
positive results in green infrastructure in Germany, which can be replicated in
emerging and developing countries. (2016: 4)30
Griffith-Jones also describes the financing terms offered by KfW in all of their areas of ac
tive lending. These include long-term loans and below-market interest rates, 100 per cent
disbursement rates, up to three years holidays in making repayments, and repayment
bonuses of up to 17.5 per cent.
Green banks. Special purpose green development banks have also become increasingly
active in recent years. A 2016 OECD study defines a green investment bank as ‘a publicly
capitalized entity established specifically to facilitate private investment into domestic
low-carbon and climate-resistant infrastructure and other green sectors such as water
and waste management’ (2016: 15). These special purpose banks have been established
at the national level in Australia, Japan, Malaysia, Switzerland, and the United Kingdom.
Within the United States, the states of California, Connecticut, Hawaii, New Jersey, New
York, and Rhode Island have created green banks. The OECD study describes the banks
as having ‘diverse rationales and goals, including meeting ambitious emissions targets,
mobilizing private capital, lowering the cost of capital, lowering energy costs, developing
green technology markets, supporting local community development and creating
jobs’ (2016: 15). The OECD study does not provide systematic evidence as to the scale at
which these institutions are currently (p. 418) providing investment financing. But consid
ering other references, it is reasonable to assume that, in general, their scale of opera
tions is much smaller than KfW.31 This raises the question as to whether the necessary
level of financing can be achieved without the full backing of large-scale national or re
gional entities, such as the equivalent of KfW.
Emerging trends in developing countries. Within developing economies there has been a
general movement in the aftermath of the 2007–09 global financial crisis away from the
predominant neo-liberal financial market policy framework that prevailed prior to the cri
sis. This trend has included the formation or expansion of development banks. For exam
ple, Grabel (2018) describes the emergence of the Development Bank of Latin America
and the New Development Bank as potentially significant new sources of subsidized long-
term financing for developing economies, including in the area of green energy invest
ments.
Page 25 of 44
Recent studies by the World Bank (Hussain, 2013) and African Development Bank
(African Development Bank Group, 2013) examine specific financial models for advancing
green investments in developing countries. Both studies consider financing arrangements
through which concessionary public financing can be mobilized to encourage, as opposed
to crowd out, private investments, thereby creating viable public–private partnerships
with clean energy investment projects. The World Bank study in particular, which focuses
on renewable energy investments, emphasizes that the long-term funding for these in
vestments has been limited by the range of risks private investors face while working
with still relatively unfamiliar technologies. These risks include uncertainty over the relia
bility of the technology within any given project and shifts in the relevant regulatory envi
ronment. The World Bank proposes a series of financing techniques for reducing these
risks for private investors. Yet the overall point remains that the public financing inter
ventions—whether they be implemented through formal development banks or otherwise
—will need to absorb a disproportionate share of these risks in order for the financing
levels to reach scale rapidly enough.
With respect to financing clean energy investments in developing countries in particular,

it is also critical that the benefits of these investments be shared fully by society’s least-
advantaged groups. Spratt, Griffith-Jones, and Ocampo emphasize this consideration in
their 2013 study ‘Mobilizing Investment for Inclusive Green Growth in Low-income Coun
tries’. This would mean, as important examples, expanding access to electricity and pro
viding clean energy for electricity and other needs at affordable prices.32 To accomplish
these ends, Spratt et al. emphasize that it is not (p. 419) realistic to expect clean energy
investments to consistently generate profits for private businesses at rates comparable to
mature investment areas, including fossil-fuel energy. The requirement that the financing
terms for clean energy investments be affordable for borrowers—that is, not always yield
ing high returns for lenders—reinforces the centrality of public investment banks with
clear social criteria guiding their financing strategies.
15.5 Domestic Resource Capacities for Clean

Energy Investments
One of the major questions that all countries will face in undertaking a clean energy
transformation will be the extent to which the large-scale expansion in clean energy in
vestment activity can be accomplished through utilizing domestic resources as opposed
to having to rely increasingly on imports. To the extent that a country runs up against do
mestic productive capacity constraints while expanding its investments in energy efficien
cy and clean renewable energy, it then faces two alternatives: either scale back the clean
energy investment project or rely increasingly on imports to maintain the ambitious in
vestment agenda.
Within this framework, it is critical to establish some measures of the range at which, in
any given country, import dependency is likely to increase as it establishes a clean energy
investment project at around 2.5 per cent of the country’s GDP. To generate a rough esti
Page 26 of 44
mate of this, I examine here the relative domestic and import content for the set of indus
trial sectors that will be mobilized to expand a country’s energy-efficiency and renewable
energy investments. I report these figures for five large economies in different regions of
the world, that is, Brazil, Germany, Indonesia, South Africa, and South Korea.
To be more specific, I undertake the following exercise. Working from the most recent
country-specific input–output tables from the OECD, those from 2015, I first calculate the
current level of domestic content for all activities that will be mobilized to undertake
clean energy investments in five major areas. These five areas are energy-efficiency in
vestments in building retrofits, industrial-efficiency and grid upgrades as well as renew
able investments in solar and wind power. Two examples of the specific set of inputs with
in a given investment project, along with the relative contributions of each of these in
puts, are as follows:
• Solar industry investments in Brazil:
▪ Computer and electronic products—35 per cent weight

▪ Construction—30 per cent weight
▪ Business sector services—18 per cent weight
(p. 420) ▪ General machinery and equipment products—12 per cent weight
▪ Basic metals manufacturing—5 per cent weight
• Grid upgrades in South Africa:
▪ Construction—25 per cent weight

▪ Computer and electronic products—25 per cent weight
▪ Electrical equipment—25 per cent weight
▪ General machinery and equipment products—25 per cent weight
Within these input–output frameworks, I then divide each of the specific activities associ
ated with each of the five investment projects into non-tradable and tradable activities.
Following from the literature, I define a ‘tradable’ activity as one in which less than 90
per cent of this activity’s inputs come from domestic sources.33
Within these definitions of ‘tradable’ and ‘non-tradable’ activities, I then assume that the
domestic content levels for non-tradable activities will remain constant as the country’s
clean energy investment project proceeds. These non-tradable activities include construc
tion, ground transportation, and administration. With tradable activities, I allow that do
mestic content will fall by up to 20 per cent. This enables us to then observe how much
overall domestic content within any given investment project area will decline when the
domestic content of specific tradable activities declines by 20 per cent.
In Table A15.6 in the Appendix, I show the results of this exercise for the five clean ener
gy investment areas and five representative countries. As we see, overall, domestic con
tent levels are generally high for all five countries with all five clean energy projects. In
Page 27 of 44
virtually all cases, domestic content levels are higher than 80 per cent. When we then al
low domestic content for tradable activities to fall by 20 per cent, we still find that, in vir
tually all cases, overall domestic content remains above 70 per cent.
Thus, after the 20 per cent decline in domestic content for tradable activities, we see that
the largest declines in overall domestic content are with grid upgrades in South Africa, in
which domestic content falls from 79 to 63 per cent; grid upgrades in South Korea, in
which overall domestic content falls from 84 to 67 per cent; and wind energy in South
Africa, in which overall domestic content declines from 83 to 68 per cent. These changes
would all represent significant increases in the respective countries’ import require
ments. But they should not entail major strains in the countries’ overall balance of pay
ments. Thus, for the most part, most countries should be able to undertake clean energy
transformations mostly through mobilizing the country’s existing supply of domestic re
sources. (p. 421)
15.5.1 Fossil Fuel Consumption and Imports/Exports
One factor in enabling the expansion of domestic production in sectors of economies

linked to clean energy will be the fact that the fossil-fuel sectors in all countries will be
correspondingly contracting. The freeing up of economic resources out of the activities
tied to the fossil-fuel sector will be substantial in all cases. These activities include ex
tracting, transporting, refining, and the retail distribution of fossil-fuel energy, along with
all the sectors that provide supplies to support these activities.
The data in Table A15.7 in the Appendix provide a sense of the magnitudes involved. The
first column of the table shows, for 2014, the extent to which each of our five representa
tive economies relies on fossil fuels to meet its overall energy consumption levels.34 As we
see, fossil fuels supply more than half of each country’s total energy consumption. Brazil
has the lowest proportion of fossil-fuel consumption, at 59 per cent of total energy con
sumption. This is because of its uniquely high levels of both hydro and biofuel production.
Indonesia is next lowest, at 66 per cent reliance on fossil fuels. But this figure includes
Indonesia’s still heavy reliance on burning peat as a high-emissions renewable energy
source. Exclusive of peat, coal, oil, and natural gas provide roughly 90 per cent of
Indonesia’s remaining energy supply. Germany, South Africa, and South Korea all rely on
fossil fuels for between about 80 and 88 per cent of their overall energy supply. These fig
ures show that, as these economies undergo transitions to clean energy sources, major
shares of their economies’ overall resources will be released from the current demands
generated by their fossil-fuel sectors.
We obtain additional perspective as to how such scenarios might play out through the fig
ures shown in the second column of Table A15.7. Here I show the import shares as a pro
portion of total energy consumption for our five selected economies as of 2014. As we
see, Indonesia and South Africa were energy exporters, both through their coal exports.
With Brazil, as the table shows, imports constituted a relatively modest 12 per cent of its
overall energy supply as of 2014, while Germany and South Korea were major energy im
Page 28 of 44
porters, at 61 and 82 per cent of their overall energy supply. These figures are represen
tative of longer-term energy consumption patterns for both countries.
Of course, the energy-importing countries, Brazil, Germany, and South Korea, are
presently utilizing a smaller share of their total domestic resources in the fossil-fuel sec
tor. Their share of total economic resources devoted to energy-linked activities could rise
as a result of increasing investments in energy efficiency and renewable energy. Howev
er, the share of total domestic resources devoted to supplying oil, coal, and natural gas in
these importing countries is still substantial. In Germany, the shares are 70 per cent for
the coal sector and 80 per cent for oil and gas. In South Korea, the proportions are 63 per
cent for coal and 79 per cent for oil and gas. Thus, even with (p. 422) Germany and South
Korea, as major energy importers, the move out of fossil fuels and into clean energy will
entail releasing domestic resources that can be repurposed for the clean energy transi
tion.
15.6 Conclusion
This chapter has demonstrated that achieving a net-zero-emissions global economy by
2050—in line with the IPCC’s climate stabilization goals—is an entirely feasible project
with respect to its technical and economic requirements. As its foundation, it will require
investments in energy efficiency and clean renewable energy at an average of $4.5 tril
lion per year globally between 2024 and 2050. This is a formidable level of spending in
absolute terms, but still amounts to only 2.5 per cent of average global GDP per year over
2021–50. Put another way, it implies that 97.5 per cent of global economic activity can
proceed largely independently of the clean energy investment project, as long as the 2.5
per cent of GDP goal is achieved each year.
The success of the global clean energy investment project—and thereby, the Global Green
New Deal—will depend on whether effective industrial and financial policies will be en
acted. Fundamentally, the Green New Deal amounts to a unified, and globally coordinated
set of industrial policies—policies, which, taken as a whole, are capable of creating an en
tirely new global energy industry infrastructure within thirty years. Accomplishing this
goal will require a range of specific policy initiatives, working in conjunction with each
other to undergird energy-efficiency and clean renewable energy investments on an un
precedented scale. The first specific requirement will be to mobilize a large enough pool
of investment funds to finance the project at the needed scale. As an illustration, I have
shown how this can be achieved through a combination of four major funding sources—a
25 per cent share of funds from carbon tax revenues; a transfer of 5 per cent of military
spending into clean energy investments; Green Bond purchases by both the Federal Re
serve and the European Central Bank, at an initial combined level of $200 billion and ris
ing with economic growth thereafter; and the elimination of all fossil-fuel subsidies and
the transfer of 25 per cent of these funds into clean energy projects.
Page 29 of 44
Working with this pool of funds, public investments will need to play the leadership role
in ramping up clean investment activities in most countries at the required pace. But pri
vate investments will be equally critical over time. Indeed, I have assumed that public
and private investment levels will need to be roughly comparable in magnitude in order
to maintain the overall project at scale. Both carrots and sticks will be needed to induce
and sustain a sufficient level of private investments. These include, as carrots, generous
financial subsidies, concessionary borrowing rates, and guaranteed markets. As sticks,
they include an ambitious renewable portfolio and energy-efficiency standards whose re
quirements cannot be readily circumvented, in contrast with some existing cap-and-trade
and renewable portfolio systems. Achieving the appropriate (p. 423) mix of these mea
sures in any given country setting will represent a major challenge in industrial policy de
sign.
Operating at the ground level, to advance this clean energy investment project at scale
will of course require the mobilization of productive resources in all countries. But as I
have shown, this should not create major problems with respect to domestic capacity bot
tlenecks, at least not after an initial adjustment period. For one thing, the domestic con
tent levels for most clean energy investment activities are already high in the relevant
productive sectors in most countries. Domestic content ratios should also remain high
even as the demands on these sectors grow with the scaling up of clean energy invest
ment activities. This is because a high proportion of the productive activity that will be re
quired are in non-tradable sectors, such as construction, ground transportation, and ad
ministration. In addition, the fossil-fuel sectors in all countries will be undergoing major
contractions as the clean energy sectors grow, thereby freeing up resources that can be
redeployed into clean energy activities.
In summary: the challenge facing humanity today with climate change is without prece
dent. Within the context of this urgent historical moment, the design and implementation
of an effective set of clean energy industrial policies will play a critical role towards
achieving the target of net-zero emissions in the global economy by 2050.
Page 30 of 44
15.7 Appendix
Table A15.1 Energy intensity ratios, global average and selected coun
tries
World Average 6.6
China 12.7
United States 5.3
Brazil 7.2
Germany 3.5
Indonesia 9.8
South Africa 20.3
South Korea 9.3
Note: Energy Intensity = Q-BTUs of energy consumed/GDP (in tril

lions of US dollars) World average for 2018; Individual Country fig
ures for 2016. GDP figures in current US dollars.
Sources: Energy consumption figures from EIA, International Energy

Statistics, http://www.eia.gov/cfapps/ipdbproject/IEDIndex3.cfm?
tid=44&pid=44&aid=2.
GDP figures from World Development Indicators: https://

data.worldbank.org/indicator.
Page 31 of 44
Table A15.2 Estimates of cost savings from energy-efficiency invest

ments
Source Regions/Countries/Sec Estimated sav

tors estimated ings in Q-BTUs
World Bank (Tay 455 projects in eleven in $1.9 billion per
lor et al., 2008: dustrial and developing Q-BTU
29) countries
McKinsey & Co. Africa, India, Middle East, $11 billion per
(2010: 27) South East Asia, Eastern Q-BTU
Europe, China
United States Na United States ~ $29 billion per

tional Academy of Q-BTU for build
Sciences (2010) ings, industry
Source: Pollin et al. (2015: 88).
Table A15.3 Average global levellized costs of electricity from utility-

scale renewable energy sources vs. fossil-fuel sources, 2010–17
2010 2017
Solar PV 36 cents 10 cents
Onshore wind 8 cents 6 cents
Geothermal 5 cents 7 cents
Hydro 4 cents 5 cents
Note: Average levellized costs for fossil-fuel generated electricity:

4.5–14 cents per kilowatt hour.
Source: https://www.irena.org/Statistics/View-Data-by-Topic/Costs/
LCOE-2010-2017.
Page 32 of 44
Table A15.4 Capital expenditure costs for building renewable electrici

ty productive equipment, present values of total lump-sum capital
costs per Q-BTU of electricity
Wind $160 billion
Solar PV $190 billion
Geothermal $112 billion
Average costs $169 billion

Assuming investments are 45 per cent wind, 45 per
cent solar, and 10 per cent geothermal
Source: EIA, https://www.eia.gov/outlooks/aeo/pdf/

electricity_generation.pdf. See Pollin et al. (2014: 136–7) for method
ology in converting levelized costs per Q-BTU into lump-sum capital
costs.
Page 33 of 44
Table A15.5 Major funding sources for global clean energy invest
ments
Investment level for 2024—Year 1 of investment cycle: $2.6 trillion in

public and private investments, at 2.5 per cent of GDP
Clean energy investment areas:

• Clean renewable energy: $2.1 trillion
○ Wind, solar, geothermal, small-scale hydro, low-emissions

bioenergy
• Energy efficiency: $500 billion
○ Buildings, transportation, industrial equipment, grid and bat

tery storage upgrades
Public sources of funds: $1.3 trillion:

• Carbon tax revenues: $160 billion
○ 25 per cent of revenues from tax; 75 per cent returned to

consumers as rebate
• Transfers from military budgets: $90 billion
○ 5 per cent of global military spending
• Green bond purchases by Federal Reserve and European Central

Bank: $200 billion
○ 1.6 per cent of Federal Reserve Wall Street bailout support

during financial crisis
• Transfers of 25 per cent of fossil-fuel subsidies: $750 billion
○ Total fossil-fuel subsidies = $3 trillion

○ 75 per cent of funds for lower clean energy prices or direct
income transfers for lower-income households
Private sources of funds: $1.3 trillion:

• Policies for Incentivizing Private Investors
○ Government procurement
○ Regulations
▪ Carbon caps and taxes

▪ Renewable energy portfolio standards for utilities
Page 34 of 44
▪ Energy efficiency standards for buildings and transporta

tion vehicles
○ Investment Subsidies
▪ Feed-in tariffs
▪ Low-cost financing through development banks and green
banks
Page 35 of 44
Table A15.6 Change in overall domestic content of clean energy investment activities after 20 per cent import in
crease with tradable activities
Energy efficiency investments Renewable investments
Building Industrial effi Grid upgrades Solar Wind

retrofits ciency
Brazil 95 per cent → 93 per cent → 87 per cent → 90 per cent → 92 per cent →
95 per cent 84 per cent 75 per cent 81 per cent 85 per cent
Germany 91 per cent → 88 per cent → 85 per cent → 88 per cent → 87 per cent →
Indonesia 91 per cent → 87 per cent → 82 per cent → 86 per cent → 83 per cent →
South Africa 86 per cent → 84 per cent → 79 per cent → 84 per cent → 83 per cent →
South Korea 89 per cent → 89 per cent → 84 per cent → 86 per cent → 87 per cent →
Sources: 2015 OECD input–output country-specific tables. Methodological details in Pollin et al. (2015: chapter 5
and appendix 2).
Page 36 of 44
Table A15.7 Reliance on fossil fuels and imports as energy sources in

selected countries, 2014
Fossil fuels as a share of Imports as a share of

total energy consumption total energy consump
tion
Brazil 59.1 per cent 11.9 per cent
Ger 79.7 per cent 60.9 per cent

many
In 66.1 per cent –103.1 per cent

done
sia
South 86.8 per cent –14.5 per cent

Africa
South 82.4 per cent 81.7 per cent

Korea
Note: Negative figures indicate net export proportion.
Source: World Development Indicators, World Bank.
References
(p. 424) Ackerman, Frank, Pat Knight, and Bruce Biewald (2016) ‘Estimat
(p. 425) (p. 426)
ing the Cost of Saved Energy’. Available at http://www.synapse-energy.com/sites/de

fault/files/COSE-EIA-861-Database-66–017.pdf.
African Development Bank Group (2013) African Development Report 2012: Towards
Green Growth in Africa. Tunis: African Development Bank Group. Available at https://
www.afdb.org/fileadmin/uploads/afdb/Documents/Publications/
African_Development_Report_2012.pdf.
Alvarez, Ramón A., Stephen W. Pacala, James J. Winebrake, William L. Chameides, and
Steven P. Hamburg (2012) ‘Greater Focus Needed on Methane Leakage from Natural Gas
Infrastructure’, Proceedings of the National Academy of Sciences of the United States of
America 109(17): 6435–40.
Azad, Rohit and Shouvik Chakraborty (2019) ‘Balancing the Climate Injustice: A Proposal
for a Differential Global Carbon Tax’, in Sevil Acar and Erinc Yeldan (eds) Handbook of
Green Economics. London: Elsevier, pp. 117–34.
Page 37 of 44
Azad, Rohit and Shouvik Chakraborty (2020) ‘Green Growth and the Right to Energy in
India’, Energy Policy 141: 111456.
Better Markets (2017) ‘The Cost of the Crisis: $20 Trillion and Counting’. Available at
https://bettermarkets.com/newsroom/20-trillion-cost-financial-crisis-3.
Boyce, James (2019) The Case for Carbon Dividends. Cambridge: Polity Press.
Coady, David, Ian Party, Louis Sears, and Baoping Shang (2017) ‘How Large are Global
Fossil Fuel Subsidies?’, World Development 91: 11–27.
Cointe, Beatrice and Alain Nadaï (2018) Feed-In Tariffs in the European Union. Bas
ingstoke: Palgrave Macmillan.
Cory, Karlynn, Toby Couture, and Claire Kreycik (2009) ‘Feed-in Tariff Policy: Design, Im
plementation, and RPS Policy Interactions’. Golden, CO: National Renewable Energy Lab
oratory. Available at http://www.nrel.gov/docs/fy09osti/45549.pdf.
EIA (US Energy Information Administration) (2012) ‘Nuclear Explained: Nuclear Power
and the Environment’. Available at https://www.eia.gov/energyexplained/index.cfm?
page=nuclear_environment.
EIA (US Energy Information Administration) (2018) ‘Levelized Costs and Lev
(p. 427)
elized Avoided Cost of New Generation Resources in the Annual Energy Outlook 2018’.
Available at https://www.eia.gov/outlooks/aeo/pdf/electricity_generation.pdf.
Grabel, Ilene (2018) When Things Don’t Fall Apart: Global Financial Governance and De
velopmental Finance in an Age of Productive Incoherence. Cambridge, MA: MIT Press.
Griffith-Jones, Stephany (2016) ‘National Development Banks and Sustainable Infrastruc

ture: The Case of KfW’. GEGI Working Paper No. 6, Boston University, Global Economic
Governance Initiative.
Howarth, Robert (2015) ‘Methane Emissions and Climatic Warming Risk from Hydraulic
Fracturing and Shale Gas Development: Implications for Policy’, Energy and Emission
Control Technologies 3: 45–54. Available at https://www.dovepress.com/methane-
emissions-and-climatic-warming-risk-from-hydraulic-fracturing--peer-reviewed-
article-EECT.
Hussain, Mustafa Zakir (2013) Financing Renewable Energy: Options for Developing Fi
nancing Instruments Using Public Funds. Washington, DC: The World Bank and Climate
Investment Funds. Available at http://documents.worldbank.org/curated/en/
196071468331818432/pdf/765560WP0Finan00Box374373B00PUBLIC0.pdf.
IEA (International Energy Agency) (2013) Energy Efficiency Market Report 2013: Market
Trends and Medium-term Prospects. Paris: OECD/IEA.
IEA (International Energy Agency) (2019) World Energy Outlook 2019. Paris: OECD/IEA.
Page 38 of 44
IRENA (International Renewable Energy Agency) (2017) Electricity Storage and Renew
ables: Costs and Markets to 2030. Abu Dhabi: International Renewable Energy Agency.
IRENA (International Renewable Energy Agency) (2019) Renewable Power Generation

Costs in 2018. Available at https://www.irena.org/publications/2019/May/Renewable-
power-generation-costs-in-2018.
Lawrence, Mark, Stefan Schafer, Helene Muri, Vivian Scott, Andreas Oschlies, Naomi E.
Vaughan, Oliver Boucher, Hauke Schmidt, Jim Haywood, and Jurgen Scheffran (2018)
‘Evaluating Climate Geoengineering Proposals in the Context of the Paris Agreement
Temperature Goals’, Nature Communications. Available at https://www.nature.com/arti
cles/s41467-018-05938-3#citeas.
McKinsey & Company (2010) Energy Efficiency: A Compelling Global Resource. New
York.
Milanes-Montero, Patricia, Alberto Arroyo-Farrona, and Esteban Perez-Calderon (2018)

‘Assessment of the Influence of Feed-in Tariffs on the Profitability of European Photovolta
ic Companies’, Sustainability, September. Available at https://www.researchgate.net/
publication/328074518_Assessment_of_the_Influence_of_Feed-
In_Tariffs_on_the_Profitability_of_European_Photovoltaic_Companies.
Molina, Maggie (2014) ‘The Best Value for America’s Energy Dollar: A National Review of
the Cost of Utility Energy Efficiency Programs’. Washington, DC: American Council for an
Energy-Efficient Economy. Available at http://assets.fiercemarkets.net/public/smart
gridnews/ACEEE-Energy-Efficiency-Costs.pdf.
NAS (National Academy of Sciences), NAE (National Academy of Engineering), and NRC
(National Research Council) (2010) Real Prospects for Energy Efficiency in the United
States. Washington, DC: National Academies Press.
Peischl, Jeff, Anna Karion, Cora Sweeney et al. (2016) ‘Quantifying Atmospheric Methane
Emissions from Oil and Natural Gas Production in the Bakken Shale Region of North
Dakota’, Journal of Geophysical Research: Atmospheres 121(10): 6101–11. https://
agupubs.onlinelibrary.wiley.com/doi/full/10.1002/2015JD024631. (p. 428)
Pierrehumbert, Raymond (2019) ‘There Is No Plan B for Dealing with the Climate Crisis’,
Bulletin of the Atomic Scientists 75(5): 1–7.
Pollin, Robert (2015) Greening the Global Economy. Cambridge, MA: MIT Press.
Pollin, Robert (2018) ‘Degrowth vs. the Green New Deal’, New Left Review 118(July/Au
gust): 5–25.
Pollin, Robert and Brian Callaci (2019) ‘The Economics of Just Transition: A Framework
for Supporting Fossil-fuel-dependent Workers and Communities in the United States’, La
bor Studies Journal 44(2): 93–138.
Page 39 of 44
Pollin, Robert, Heidi Garrett-Peltier, James Heintz, and Bracken Hendricks (2014) Green
Growth: A US Program for Controlling Climate Change and Expanding Job Opportunities.
Washington, DC: Center for American Progress. Available at https://
cdn.americanprogress.org/wp-content/uploads/2014/09/PERI.pdf.
Pollin, Robert, Heidi Garrett-Peltier, James Heintz, and Shouvik Chakraborty (2015) ‘Glob
al Green Growth: Clean Energy Industrial Investments and Expanding Job Opportunities’,
United Nations Industrial Development Organization and Global Green Growth Institute.
Available at http://gggi.org/wp-content/uploads/2015/06/GGGI-VOL-I_WEB.pdf.
Pollin, Robert, Heidi Garrett-Peltier, and Jeannette Wicks-Lim (2017) ‘Clean Energy In
vestments for New York State: An Economic Framework for Promoting Climate Stabiliz
ing and Expanding Good Job Opportunities’, Political Economy Research Institute, No
vember. Available at https://www.peri.umass.edu/publication/item/1026-clean-ener
gy-investments-for-new-york-state-an-economic-framework-for-promoting-cli
mate-stabilization-and-expanding-good-job-opportunities.
Pollin, Robert, Jeannette Wicks-Lim, Shouvik Chakraborty, and Tyler Hansen (2019) ‘A
Green Growth Program for Colorado’. Political Economy Research Institute, April. Avail
able at https://www.peri.umass.edu/publication/item/1168-a-green-growth-pro
gram-for-colorado.
Prentiss, Mara (2015) Energy Revolution: The Physics and the Promise of Efficient Tech
nology. Cambridge, MA: Harvard University Press.
Romm, Joe (2014) ‘Methane Leaks Wipe out any Climate Benefit of Fracking, Satellite Ob
servations Confirm’, Think Progress. Available at https://thinkprogress.org/methane-
leaks-wipe-out-any-climate-benefit-of-fracking-satellite-observations-
confirm-2ac26dd30381/.
Rosenow, Jan and Edith Bayer (2016) ‘Costs and Benefits of Energy Efficiency Obligation
Schemes’. Available at http://www.raponline.org/wp-content/uploads/2016/11/rap-
rosenow-bayer-costs-benefits-energy-efficiency-obligation-schemes-2016.pdf.
Ross, Michael L. (2015) ‘What Have We Learned about the Resource Curse?’, Annual Re
view of Political Science 18: 239–59.
Ruttan, Vernon W. (2006) Is War Necessary for Economic Growth? Military Procurement
and Technology Development. New York: Oxford University Press.
Spratt, Stephen, Stephany Griffith-Jones, and José Antonio Ocampo (2013) ‘Mobilising In
vestment for Inclusive Green Growth in Low-Income Countries’. Berlin: Deutsche Gessel
llschaft für Internationale Zusammenarbeit. Available at http://www.stephanygj.net/pa
pers/MobilisingInvestmentforInclusiveGreenGrowth2013.pdf.
Page 40 of 44
Teeter, Preston and Jorgen Sandberg (2017) ‘Constraining or Enabling Green Capability
Development? How Policy Uncertainty Affects Organizational Responses to Flexible Envi
ronmental Regulations’, British Journal of Management 28(4): 649–65.
Venables, Anthony J. (2016) ‘Using Natural Resources for Development: Why Has It
Proven So Difficult?’, Journal of Economic Perspectives 30(1): 161–84.
Notes:
(1) All $ values are in US dollars.
(2) See Pollin et al. (2014: 113–15).
(3) See Pollin (2015), Pollin et al. (2015), Pollin and Callaci (2019), and Pollin et al. (2019)
for discussions and further references on employment effects and Just Transition policies.
(4) Two recent survey articles on the Resource Curse are Ross (2015) and Venables
(2016).
(5) See Prentiss (2015) for her calculations on land-use requirements and a brief applica
tion of her framework in Pollin (2018: 14–17).
(6) https://www.yaleclimateconnections.org/2016/07/pros-and-cons-the-promise-and-pit
falls-of-natural-gas/.
(7) See, for example, Alvarez et al. (2012), Romm (2014), Howarth (2015), and Peischl et
al. (2016).
(8) https://earthobservatory.nasa.gov/images/1510/global-effects-of-mount-pinatubo.
(9) In his 2019 paper, ‘There Is No Plan B for Dealing with the Climate Crisis’, the leading
climate scientist and lead co-author of the Third Assessment Report of the IPCC Raymond
Pierrehumbert is even more emphatic in arguing that geoengineering does not offer a vi
able solution to the climate crisis.
(10) The IEA summarizes its three scenarios—the Stated Policies Scenario, the Sustain
able Development Scenario, and the Current Policies Scenario, on p. 751 of its 2019
World Energy Outlook.
(11) It is, however, important to note that the pattern with these ratios is highly sensitive
to the method by which one measures national GDP figures. The figures reported here
are based on nominal US dollars calculated according to each country’s exchange rate. If,
alternatively, we measured national GDP figures based on purchasing power parity, the
GDP figures would be significantly higher for the lower-income economies. This would in
turn lower their energy intensity ratios. How best to deal with these methodological is
sues is an important question, but it is beyond the scope of this chapter.
Page 41 of 44
(12) I am not aware of more recent studies that have attempted to provide comparable ag
gregated cost estimates. However, recent studies on the building sector in the US econo
my have generated results similar to those in the 2010 NAS study. These more recent
studies include Molina (2014), Ackerman et al. (2016), and Rosenow and Bayer (2016).
(13) See the discussion and references in Pollin et al. (2015: 92–6).
(14) Such detailed figures are also available in IRENA (2019).
(15) See IRENA (2017) on electricity storage costs and markets through 2030.
(16) These figures are from the US Energy Information Agency (EIA, 2018).
(17) The full methodology for generating these costs is presented in Pollin et al. (2014:
136–7).
(18) Renewable investments = $330 billion in 2018, https://about.bnef.com/blog/clean-en

ergy-investment-exceeded-300-billion-2018/; but $390 billion in 2018 according to IEA (p.
309). But it doesn’t break out bioenergy from the others. Energy efficiency, from IEA in
2014 is $240 billion (p. 307). Total is therefore $630 billion. But if we take out ~ 10 per
cent for high-emissions bioenergy, that gets us to $570 billion. $570 billion/$86 trillion =
0.7 per cent GDP.
(19) This $86 trillion global GDP figure comes from World Development Indicators. It is
derived from the prevailing exchange rates between the United States and all other glob
al currencies as of 2018. The most widely utilized alternative measure of global GDP is
derived through establishing purchasing power parities between countries. Measured ac
cording to the purchasing power parity methodology, global GDP in 2018 was $138 tril
lion (World Development Indicators). For the purposes of this exercise, it is important to,
if anything, err by underestimating the prospects for a clean energy investment pro
gramme over the thirty-year investment cycle. Therefore, for this exercise, I utilize the
lower $86 trillion figure derived on the basis of exchange rates as of 2018.
(20) See Ruttan (2006).
(21) Cory, Couture, and Kreycik (2009: 2).
(22) Cointe and Nadaï (2018) emphasize this point and contrast it with the official EU aim
of liberalizing renewable energy markets.
(23) See Boyce (2019) for an effective solution to the distributional problem, via what he
terms ‘carbon dividends’. Azad and Chakraborty (2019) expand on the idea of an egalitar
ian carbon dividend programme to the global economy.
(24) See, for example, Teeter and Sandberg (2017). There is also the problem of the caps,
or renewable portfolio standards, being established in law but then ignored in policy im
Page 42 of 44
plementation. This has been the experience, for example, in New York State. See Pollin et
al. (2017: 79–80).
(25) file:///C:/Users/RPollin/Downloads/global-wealth-report-2018-en.pdf.
(26) Azad and Chakraborty (2019) develop a more complex rebate structure, that rewards
residents of countries according to the emissions levels of each country.
(27) https://www.sipri.org/media/press-release/2019/world-military-expenditure-grows-18-
trillion-2018.
(28) https://www.ft.com/content/89f5f412-12bc-11ea-a225-db2f231cfeae.
(29) Coady et al. (2017). This study distinguishes direct fossil-fuel subsidies—what it
terms ‘pre-tax’ subsidies—and ‘post-tax’ subsidies. They define post-tax subsidies as in
cluding global warming damages, air pollution damages, and vehicle externalities, includ
ing congestion, accidents, and road damage. They estimate post-tax subsidies as amount
ing to roughly 6 per cent of global GDP. These are valuable calculations. But for the pur
poses of this discussion on financing, the standard, and much more narrowly defined,
measure of pre-tax subsidies are more directly relevant.
(30) Griffith-Jones’s conclusions are fully in line with those of other researchers. For exam
ple, the overview of the IEA’s 2013 Energy Efficiency Market Report concluded that ‘Ger
many is a world leader in energy efficiency. Germany’s state-owned development bank,
KfW, plays a crucial role by providing loans and subsidies for investment in energy-effi
ciency measures in buildings and industry, which have leveraged significant private
funds’ (IEA, 2013: 149).
(31) See, for example, Pollin et al. (2017) for a discussion of the New York State green
bank and related public financing initiatives within New York State. See also Pollin et al.
(2014) for a discussion of green banks within the US economy, and as one element within
a broader framework of measures to support clean energy investments.
(32) As one specific policy proposal, Azad and Chakraborty (2020) develop a programme
for rapidly advancing the expansion of renewable energy supply in India. The proposal in
cludes a carbon tax, with the revenues from the tax being channelled into clean renew
able energy investments that will then supply free electricity to low-income communities,
many of which still have no access to electricity.
(33) This discussion and set of calculations are an updated version of that presented in
Pollin et al. (2015: 111–19). Full references on methodology and related matters are pre
sented in this 2015 publication.
(34) The 2014 data reported in Table A15.7 are the most recent complete set of figures for
all five countries.
Page 43 of 44
Robert Pollin
Robert Pollin is distinguished university professor of economics and co-director of

the Political Economy Research Institute (PERI) at the University of Massachusetts-
Amherst. He is also the founder and president of PEAR (Pollin Energy and Retrofits),
an Amherst, MA-based green energy company operating throughout the United
States. His books include The Living Wage: Building a Fair Economy (co-authored,
1998), Contours of Descent: U.S. Economic Fractures and the Landscape of Global
Austerity (2003), An Employment-Targeted Economic Program for South Africa (co-
authored, 2007), Back to Full Employment (2012), Greening the Global Economy
(2015), and The Political Economy of Climate Change and the Global Green New Deal
(co-authored, forthcoming 2020).
Page 44 of 44
Industrial Policy and Gender Inclusivity

Stephanie Seguino
The economics of development and structural change highlight the importance of indus
trialization as a means for countries to move from poverty to higher living standards and
well-being. Industrial policy—that is, an articulated path to industrialization and the pro
duction of higher valued-added goods and services—is a means to achieve that goal. A
key question is how to make industrial policy gender-inclusive. This chapter begins by
identifying gender disparities that have emerged in the process of structural change and
industrial upgrading. The early stages of industrialization in developing countries wit
nessed the feminization of industrial employment, but this has been followed by de-femi
nization as economies industrially upgrade. That exclusion has costly effects on the suc
cess of industrial policies by limiting the talent pool. The costs of social reproduction are
disproportionately borne by women and, if not addressed, can limit the potential for in
dustrial policies to achieve their goals. The chapter concludes by identifying mechanisms
to induce and incentivize gender inclusion in employment and social reproduction.
Keywords: industrial policy, industrial upgrading, gender inequalities, stratification, social reproduction
16.1 Introduction
THE economics of development and structural change highlight the importance of indus
trialization as a means for countries to move from poverty to higher living standards and
well-being. Industrial policy—that is, an articulated policy path to industrialization and
the production of higher valued-added goods and services—is a means to achieve that
goal. Markets on their own, however, do not ensure that the benefits of industrialization
are equitably shared. With the scope of industrial policy becoming far broader than had
traditionally been conceived, integration of issues of equality and inclusion are both time
ly and needed.1
While greater equality that ensures the benefits of structural change and industrial up
grading are broadly shared is a goal in its own right, a large body of research on feed
back loops from intergroup equality to the macroeconomy2 suggests that the degree of
Page 1 of 25
equality itself impacts the effectiveness of industrial policy. For example, correctly imple
mented, industrial policies that promote greater equality generate the aggregate demand
required to absorb increased output, putting the economy on a path to equity-led struc
tural change and growth. And at the same time, greater equality has positive supply-side
effects on the quality of the labour force and knowledge assets, key components of suc
cessful industrial policies.
Attention to the role of gender equality in the success of industrial policies is yet another,
albeit relatively neglected, way in which the industrial policy frame can usefully be ex
panded. This chapter explores the two-way relationship between gender (p. 430) and in
dustrialization and the consequent implications for industrial policy design. In terms of
the effect of industrialization on gender, the employment, consumption, social reproduc
tion, and knowledge production channels are the most salient. Regarding employment, in
the early stages of industrialization of developing countries, women’s share of manufac
turing employment rose substantially, albeit under conditions of insecure employment
and low wages, with little opportunity to move up the ladder to better paid jobs. That
trend appears to reverse as countries industrially evolve, with evidence of women exclud
ed from jobs in more knowledge- and capital-intensive industries. This has occurred de
spite the fact that gender educational disparities have narrowed and even closed in a
number of countries.
Women are also underrepresented in high-tech firms, and more generally, in job cate
gories associated with innovation and technological change. Apart from the loss of
women’s talent in such jobs, this limits the types of innovations generated, and omits the
interests and needs of women as consumers in product design. It also leads to a lack of
attention to technologies that could reduce unpaid care work, with which women are dis
proportionately burdened. As outlined in this chapter, substantial economic gains could
be achieved by industrial policies that promote gender equality. This suggests that gen
der should figure prominently in industrial policy design from the start, expanding the
scope of industrial policies related to sustainability and inclusion.
16.2 The Impact of Industrial Policy and Struc

tural Change on Gender Equality
At the broadest level, the aim of industrial policy is to stimulate productivity growth in or
der to raise living standards. The way to do this is to generate the conditions for the pro
duction of high value-added goods, in particular manufactured products but also services
and agricultural goods (Oqubay, Chapter 2, this volume; Cramer and Tregenna, 2020). To
achieve this goal, industrial policies target activities that promote gains in productivity
and employment, foreign exchange earnings via export competitiveness, production
arrangements that support backward and forward linkages, and perhaps most important
ly, knowledge production, including learning by doing, in order to support dynamic com
parative advantage.
Page 2 of 25
Governments face significant social, economic, and environmental challenges that, if left
unaddressed, can undermine progress in achieving the objectives of industrial policy. For
example, the negative environmental impacts of pursuing industrialization and develop
ment with current energy technologies are by now well understood. The ‘greening’ of in
dustrial strategies has been advanced as a means to address environmental challenges
while promoting growth of output and employment (Rodrik, 2014; Pollin, Chapter 15, this
volume). The Green New Deal is an example of an industrial policy that addresses envi
ronmental goals as well as equity issues (Ocasio-Cortez, 2019).
Gender inequality is also impacted by industrial policies, and thus requires targeted ef
forts to design gender-equalizing policies. There are several channels through which in
dustrial policies affect the degree of gender inequality. Among them, the most important
are the following: (1) structural change influences access to employment and wages, and
may be gender equalizing or dis-equalizing due to the existence of gender job segrega
tion, unrelated to human capital differences; (2) industrial policies influence innovation
pathways that result in the diffusion of new varieties of industrial goods that may benefit
some groups more than others; and (3) public investment in infrastructure may reduce or
exacerbate gender inequalities, especially in the sphere of social reproduction.
16.2.1 Gender, Employment, and Industrialization
The impact of industrial policy, industrialization, and structural change on women’s rela
tive access to paid employment has long been a topic of feminist economics research. A
gender division of labour continues to be pervasive globally and is one of the main
sources of gender income inequality. Two salient aspects to this phenomenon are that: (1)
women are socially assigned responsibility for unpaid labour that produces and repro
duces the labour key to the industrialization process; and (2) even within the paid sector
of the economy, jobs are ‘gendered’, with women slotted into lower-quality occupations
and industrial sectors of the economy.
I begin with a discussion of the relationship between industrial policy, industrialization,

and gender job segregation. Industrial policies, whether in the United States in the 1800s
or in semi-industrialized countries since the 1980s, began with the development and ex
pansion of labour-intensive manufacturing industries, most notably garments and textiles.
Due to their lower wages than similarly qualified men, women have been the target
labour force in these and other labour-intensive industries. The pursuit of cheap labour
sources is intensified in economies that rely on exports to stimulate growth, with export
competitiveness enhanced by low unit labour costs and thus low-wage labour. The recent
period of globalization that began in the 1970s in response to the liberalization of trade
and investment flows has exacerbated the pressure on early industrializers to rely on low-
wage female labour as a means to attain competitiveness. Two descriptive terms have
emerged for this trend—the ‘global feminization of labour’ (Standing, 1989) and the ‘femi
nization of foreign exchange earnings’ (Samarasinghe, 1998).3
Page 3 of 25
This is not to say that industrial policies have explicitly identified women as the
(p. 432)
preferred source of low-wage workers in labour-intensive jobs. Rather, export-led labour-

intensive industrialization has been passively compatible with, and indeed, has relied on
patriarchal norms. Those norms, which assign women the responsibility for unpaid
labour, meant that women have often been barred from light manufacturing employment
once married.4 This proved essential to the success of industrial policies in so far as
women provided the unpaid labour that produced and reproduced the labour force. That
labour includes the care, socialization, and training of children and activities to ensure
the productivity of adult workers (e.g. cooking and other tasks that maintain the house
hold, as well as provision of emotional labour). Young unmarried women, as a result, have
been the target factory workers in a number of industrializing countries in the last four
decades.5 This is due to their low wages, perceived docility,6 ideologically constructed
stereotypes about women (such as their supposedly ‘nimble fingers’),7 and the flexibility
of hiring and firing women to reduce labour costs in response to fluctuations in product
demand.
What are the channels through which gender wage inequality stimulates growth?
Women’s low wages are less a function of their skills than of overt gender wage discrimi
nation as well as job segregation that reduces women’s bargaining power, depressing
their wages. There is evidence that women’s low wages have been a causal factor in the
success of the industrial policies that have stimulated the growth of export manufacturing
in a number of newly industrializing economies (Seguino, 2000; Busse and Spielmann,
2006). Women’s low wages have substituted for (or complemented) currency devaluation,
making exports cheaper than they would otherwise be and thus stimulating demand as
well as a country’s share of global supply. This serves to relax the balance-of-payments
constraint. Women’s low wages in labour-intensive manufacturing therefore act as a sub
sidy that supports the acquisition of intermediate inputs and technology required in more
capital- and skill-intensive manufacturing industries.
Neoclassical theory posits that women’s subordinate position in the industrialization

process will improve over time. This view is based on the assumption that the strong de
mand for female labour and increased educational opportunities for women would
(p. 433) contribute to rising female wages and a narrowing of the gender wage gap over
time. Scholarly work on trends in gender wage inequality indicates that such optimism is
not warranted. There is evidence that the discriminatory portion of gender wage gaps has
widened in a number of industrializing countries, such as China, India, Mexico, and Viet
nam, despite a strong demand for women’s labour (Braunstein, 2012).
Downward pressure on women’s manufacturing wages is due to several factors. Women

workers’ concentration in labour-intensive export industries as compared to men’s in non-
tradables or capital-intensive export goods production results in women’s weaker bar
gaining power. Labour-intensive firms that are more mobile than ‘male’ industries can
more easily relocate to other countries, should wages rise, holding down women’s wages
(Seguino, 2007). Further, many firms in developing countries are subcontractors in global
buyer- or producer-driven commodity chains. The outsized bargaining power of the domi
Page 4 of 25
nant firms in the chain allows them to obtain low-cost components and assembled goods
from subcontracting firms in developing countries, squeezing the wages of (women)
workers. Again, the gendered effect is a function of the forms that gender job segregation
takes in industrializing countries that have promoted export-led growth.
As semi-industrialized countries have moved up the industrial ladder, employing produc

tion technologies that are more capital and knowledge intensive, there is substantial evi
dence of de-feminization—that is, of women’s declining share of manufacturing employ
ment (Tejani and Milberg, 2016; Kucera and Tejani, 2014; Saracoglu et al., 2018). Kucera
and Tejani (2014) conducted a decomposition analysis to more fully understand this
trend, focusing on two causal mechanisms of de-feminization. The first is the reallocation
effect—that is, the impact on women’s share of employment due to the shift in production
to more capital- and knowledge-intensive goods. The second is the within-industry em
ployment effect, whereby industrial upgrading within an industry contributes to the de
cline in women’s share of employment. While they find evidence of both effects, within-in
dustry effects in the manufacturing sector dominate in explaining the declines in female
shares of manufacturing employment. Kucera and Tejani (2014: 578) explain:
Though structural change in the process of economic development is indeed char

acterized by shifts toward less labour-intensive, higher value-added manufactur
ing industries, it is developments within industries that generally matter more in
accounting for patterns of feminization and de-feminization for the manufacturing
sector as a whole.
In addition to women’s declining share of manufacturing employment as capital intensity

of production rises, there is evidence of increased gender job segregation, with a declin
ing share of women employed in industry8 (a broader measure than (p. 434) manufactur
ing) relative to the share of men employed in industry.9 This is indicative of increased ex
clusion of women (relative to men) from jobs in the industrial sector as a whole, where
employment tends to be of higher quality than in the agricultural and services sectors.10
Data in Figure 16.1 demonstrate that gender job segregation is increasing in countries at
varying stages of development over the period 1991 to 2017, reflecting women’s declin
ing access to industrial employment (and thus crowding into jobs in other sectors of the
economy). Panel A shows these trends by region.11 Gender job segregation is measured
as the ratio of the share of women employed in industry relative to the share of men em
ployed in industry—dubbed ‘women’s relative concentration in industrial-sector employ
ment’. The largest decline in women’s relative concentration (–23 per cent) occurred in
the most industrially dynamic region, East Asia and the Pacific, closely followed by the
Middle East and North Africa (MENA), and Central Europe and the Baltics. Sub-Saharan
Africa has experienced increased relative exclusion of women from industrial-sector jobs
as well, but to a lesser extent than more industrialized regions. South Asia has not experi
enced an increase in gender job segregation and this may be due to the fact that this re
gion is still engaged primarily in labour-intensive export production.
Page 5 of 25
Panel B displays trends in gender job segregation by stage of development. As would be

expected, in low-income countries, industrialization is less likely to be characterized by a
decline in women’s relative concentration in industrial-sector employment. The most sub
stantial decline (−21 per cent) is in middle-income countries. The increase in job segrega
tion in high-income countries is also strong. There, women’s relative share of industrial
jobs declined by 16 per cent over this period.
Figure 16.1 Change in women’s relative concentra

tion in industrial-sector employment, 1991–2017
Note: LAC is Latin America and the Caribbean.

MENA is the Middle East and North Africa. OECD is
the Organisation for Economic Cooperation and De
velopment. Women’s relative concentration in indus
trial-sector employment is defined as (share of em
ployed women in industry)/(% of employed men in in
dustry).
Source: Author’s calculations from World Develop

ment Indicators (accessed November 2019).
Finally, changes in women’s relative concentration in industrial-sector jobs in selected

noteworthy industrializing countries are shown in Panel C. The largest increase in job
segregation among these countries has taken place in Ethiopia, with women’s relative
concentration falling 85 per cent (from 173 per cent to 88 per cent) over this time period.
Note that this occurred in the context of expanding industrial output, such that the per
centage of women employed in the industrial sector as well as the percentage of men so
employed rose during the period, but men’s rose more than women’s (from 4.6 per cent to
12.1 per cent as compared to 7.9 per cent to 10.7 per cent (p. 435) (p. 436) for women).12
Similarly, in China, Indonesia, Vietnam, and India, the decline in women’s relative concen
tration in industrial employment is due to a more rapid increase in men’s concentration in
industrial employment than women’s. In Malaysia, Republic of Korea, and South Africa,
the share of women employed in industrial-sector jobs declined more than did the share
Page 6 of 25
of men so employed. And in Brazil and Malaysia, the share of women in industry fell while
the share of men employed in industry rose.
These trends do not bode well for the ability of structural change and industrial upgrad
ing to contribute to gender equality. Why is de-feminization and increased gender job seg
regation occurring? The mainstream explanation for women’s exclusion (whether ab
solutely or relative to men) from high value-added jobs is that women’s lack of education
renders them less qualified for skill-intensive jobs. That, however, ignores the fact that for
many workers, skills are acquired on the job through training and learning by doing—and
that educational gaps have in any case narrowed substantially over time.13
A number of other factors, including restrictive gender norms and stereotypes,14

contribute to women’s exclusion from good jobs as the economy upgrades. Moreover, as
industries become less labour intensive, labour costs are not as much of a constraint on
product demand or firm profits—and therefore, employing men at higher wages (than
women’s) is not a significant hindrance to competitiveness or profits. In other words, gen
der discrimination in access to good jobs becomes less costly as economies evolve indus
trially.
Those explanations, however, cannot fully explain the process of de-feminization and job
segregation observed in industrializing countries. To understand gender employment dy
namics occurring with structural change requires an analytical framework able to explain
the determinants of intergroup inequality—and in this particular case, gender inequality
as evidenced by women’s increased exclusion from higher-equality jobs in the industrial
sector. Stratification theory argues that intergroup inequality is created and reproduced
by dominant groups in order to maintain their privileged access to, and control over, re
sources. Two mechanisms are used to maintain dominance: exploitation (paying people
less than the value of what they produce), and opportunity hoarding (or exclusion) of
prized economic assets such as high-quality jobs (Blumer, 1958; Tilly, 1999).
(p. 437) Gender inequality persists and indeed is increasing in terms of job segregation as
a result of these dual processes. Gender norms and stereotypes saddle women with the
responsibility for unpaid labour that others benefit from (exploitation), and justify men’s
right to the highest quality and status jobs, affirming a gender hierarchy. ‘Gender police’
are not required to enforce a system of stratification. Gender norms and stereotypes work
to consolidate perceptions of group differences that justify exclusion. In the case of gen
der, for example, a widely held norm is that men are the primary breadwinners, and are
thus most deserving of high-wage jobs. More recently, the stereotype has emerged that
men are more suited to science and technology jobs than women. The gendering of newly
created types of jobs in technologically upgrading industries demonstrates that even as
old occupations associated with one gender or the other disappear in the process of
structural change, gender stereotypes recreate gender inequality in new occupations and
work environments (Ridgeway, 2011).
Page 7 of 25
The processes of de-feminization and increased gender job segregation are exacerbated
in the context of increasingly scarce industrial-sector employment. In recent years, pat
terns of stalled industrialization or premature de-industrialization have been observed in
a number of developing countries, limiting the growth of industrial-sector jobs (UNCTAD,
2016). The result is a relative downsizing of the core industrial sector. A falling share of
industrial jobs in total employment can intensify competition for the fewer jobs available,
triggering the forces of stratification that influence job access. Seguino and Braunstein
(2019) present empirical evidence that structural change, as evidenced by increases in
capital/labour ratios, combined with declining industrial employment shares of total em
ployment, have contributed to a worsening of gender job segregation since 1991.
As a result of stratification dynamics, structural change induced by industrial policies

can, and has, led to increased exclusion of women from employment in upgrading sectors
in countries, resulting in women being crowded into other sectors of the economy, and
putting downward pressure on their wages. This discussion underscores two key points.
First, industrial policies that promote structural change and the shift to higher valued-
added production using more sophisticated technologies can contribute to and exacer
bate already existing gender inequality. Women as workers may be the losers, absent at
tention to the dynamics that lead to their exclusion from newly created jobs.
Second, macroeconomic conditions matter. Productivity growth is not sufficient for work
ers to obtain the benefits of industrial upgrading. An exclusive focus on a supply-side in
crease in output through industrial policy misses the fact that aggregate demand and oth
er macro-level policies (such as exchange rate, trade, and investment policies) may not be
sufficient to ratify the increases in potential output that comes with effective industrial
policy. Without sufficient demand growth for industrial output, there will be negative ef
fects on employment, that is, de-industrialization, and ultimately, more intense gender job
competition that contributes to gender inequality.
These problems are not fatal to industrial policies that intend to be inclusive. Below I dis
cuss approaches to industrial policies that can alleviate the potential for upgrading
(p. 438) to worsen gender inequality. Suffice it to say that just as industrial upgrading is
unlikely to happen without government intervention in the form of industrial policy, so too
industrial policy is unlikely to be gender equitable, absent specific policies to override
gender dis-equalizing practices embedded in labour markets and other institutions.
16.3 Gender Equality Feedback Loops: Impact

on Success of Industrial Policies
Just as industrial policies may affect women and men differently—and potentially un
equally, via gendered employment effects—so too does gender (in)equality affect the po
tential success of industrial policies, and the desired structural change that accompanies
Page 8 of 25
those policies. The following discussion, while not exhaustive, highlights salient aspects
of feedback loops with regard to labour productivity and knowledge production.
16.3.1 Gender and Labour Productivity Growth
A key question identified in section 16.2 is how to make industrial policy inclusive, given
that gender job segregation is prevalent, with women slotted into lower-wage jobs and
unpaid labour, and men dominating in higher-wage industrial jobs. This question is all the
more pertinent due to the complementarities between production methods and a
country’s level of skills and capabilities. As a country moves up the industrial ladder to
the production of more knowledge- and capital-intensive goods and services, govern
ments must ensure that human capability improvements, the invisible foundation of a
country’s successful economic development, keep pace.
Conventional economics portrays labour productivity as merely a function of the level of

formal education and on-the-job training of the individual. That narrow characterization
misses the important fact that long before arriving at the factory or office door, people as
a resource have been produced through significant amounts of material investment as
well as time. Recognizing this, feminist economists have expanded the concept of human
capital in a way that treats it as an asset that is co-produced by families and government,
rather than simply a product of the formal education system (Folbre, 2012). To differenti
ate the use of this term from its traditional definition employed in economics, we use the
term human capacities, which reflects the full breadth of learning and socialization that
extends beyond formal education and is required for workers to be productive.
Investments in human capacities raise future productive capacity (i.e. generate

(p. 439)
economic growth) in ways similar to building more factories and equipment, though in
vesting in the future labour force is almost never treated as investment in macroeconom
ic models (Braunstein, van Staveren, and Tavani, 2011; Heintz, 2019). Although govern
ments and men also participate in human capacities development, it is primarily women
who carry out the work of social reproduction by doing both paid and unpaid caring
labour.15
Macroeconomic and economic development theory err by assuming women’s unpaid care
work is infinitely elastic. But women’s participation in the paid economy necessarily af
fects the quantity of investment in human capacities—that is, the more time spent in paid
work, the less time and energy is available for investments in human capacities. Failing to
track and ensure sufficient human capacities investments can cause industrial and other
macroeconomic policies to inadvertently undermine social reproduction with negative ef
fects on the current and future labour supply’s productivity and ability to provide the
complementary skilled labour needed as countries industrially upgrade.
How can industrial policy take account of and address this problem? Through public poli
cies and infrastructure investment, governments can redistribute and reduce16 unpaid
care work, creating the conditions for women to participate in the paid economy without
sacrificing investments in the care labour required to ensure children become a produc
Page 9 of 25
tive future labour force. Public policy can influence the distribution of care work between
women and men, and can also reduce the amount of necessary care work through public
investment in physical infrastructure (roads, sanitation, and electricity that reduce the
time necessary for care activities such as fetching water and fuel), as well as social infra
structure (spending on publicly funded care services). These expenditures, which can re
duce women’s unpaid care burden and redistribute social reproduction to the state and to
men, also support the goals of industrial policy. Attention to issues of social reproduction,
then, is another way in which the scope of industrial policies can be expanded beyond its
traditional frame.
Braunstein, Bouhia, and Seguino (2020) present a model and empirical analysis that links
structures of class-driven growth of output, employment, and productivity with those of
social reproduction and gender inequality. How social reproduction is organized (Who
does it? Does it take place in the household, or in public or market sectors?) influences
long-run productivity. The analysis covers the period 2008‒15. Principal components
analysis (PCA) scores, based on indicators of social reproduction and (p. 440) equality-led
vs. profit-led growth, are used to categorize national economies into one of four regime
types: mutual, time squeeze, wage squeeze, and exploitation. Applying PCA to this typolo
gy, the authors identify the conditions under which systems of growth on the one hand
and social reproduction on the other reinforce or contradict one another, with a focus on
the role of gender (in)equality.
Mutual regimes are the ideal—production and social reproduction dynamics mutually re
inforce each other. In this scenario, more gender equality (higher wages, more time spent
in paid labour) raises growth of output, employment, and productivity because it raises
aggregate demand17 and promotes human capacities development more than it cuts into
profits. Strong public support for care, the availability of care commodities, and men
stepping up to provide care all add to the beneficial impact of gender equality on growth
and social reproduction.
In time squeeze countries, more gender equality in the form of higher wages for women
or for care workers supports investment and growth because it raises human capacities.
However, as women’s labour-force participation increases, the time devoted to care work
decreases (because men do not contribute enough to care work, market care services and
commodities are inadequate,18 and/or due to infrastructure inadequacies). This inhibits
structural change and the success of industrial policies more generally, owing to the neg
ative effect on labour productivity growth.
In wage squeeze countries, higher wages for women raise human capacities but not by
enough to outweigh the negative effect of those higher wages on profits, investment, and
thus growth. The more profit-led (or inequality-led) the structure of the macroeconomy,
the more likely wage squeeze is likely to be obtained. The more open to the global econo
my, the greater the probability an economy is profit-led, exacerbating the contradictions
between gender equality, social reproduction, and growth (Blecker, 2016).
Page 10 of 25
And finally, the exploitation regime is one in which higher wages for women reduce
growth because they dampen profits and aggregate demand more than they raise human
capacities investment. And women’s higher wages and increased market participation
squeeze time spent on social reproduction, a problem that is accentuated in the event of
lack of public or male support, and absence of adequate care commodities and infrastruc
ture to replace women’s decreased care time. In this regime, growth is predicated on ex
ploiting women’s (unpaid) reproductive labour.
Gender/growth regimes vary widely among the countries that have adopted industrial
policies in the last several decades (Table 16.1). This variation is demonstrated by the
BRICS, among which Brazil’s is the only regime to fall into the mutual category. The re
cent increases in social protection spending there are instructive, demonstrating the im
portance of public policy to promote gender equality. But Brazil’s poor (p. 441) macroeco
nomic performance (relative to other successful industrializers) reminds us that a mutual
regime does not automatically or inherently induce economic growth, absent supportive
macroeconomic policies such as appropriate exchange-rate policy (Astorga, Cimoli, and
Porcile, 2014).19
Page 11 of 25
Table 16.1 Gender regimes in BRICS and selected developing and de
veloped countries
Regime type
BRICS
Brazil Mutual
Russia Wage squeeze
India Exploitative
China Wage squeeze
South Africa Wage squeeze
Developing Africa
Ethiopia Time squeeze
Nigeria Time squeeze
Rwanda Time squeeze
Developing Asia
Bangladesh Time squeeze
Hong Kong Wage squeeze
Indonesia Wage squeeze
Malaysia Wage squeeze
Singapore Wage squeeze
South Korea Wage squeeze
Thailand Wage squeeze
Developed economies
Page 12 of 25
Finland Mutual
France Mutual
Germany Wage squeeze
Ireland Wage squeeze
Norway Mutual
Poland Exploitation
Sweden Mutual
United States Time squeeze
Source: Braunstein, Bouhia, and Seguino (2020).
(p. 442) In contrast, China, Russia, and South Africa are characterized as wage squeeze
regimes. In those cases, higher wages for women show slow growth (these are profit-led
economies, due to the structure of production and macro-level policies) because the bene
fits to human capacities production are outweighed by the negative effect of higher
wages on profits, investment, and demand.20
India is a case of an exploitative gender/growth regime. Improvements in gender equality

in the paid economy negatively impact growth, while movement of women into the paid
economy reduces human capacities due to government’s failure to take action to redis
tribute unpaid care work, lack of adequate infrastructure, and insufficient male participa
tion in care work.
Among developing African economies that have adopted industrial policies, Ethiopia,
Nigeria, and Rwanda have time squeeze regimes (as do many other African economies).
As more women enter the labour market in these countries, the consequent strain on
women’s time limits human capacities development and the growth of labour productivi
ty. The policy implication of this combination is clear: increasing women’s paid employ
ment must be accompanied by more support for care and social reproduction to sustain
ably deliver growth—and absent attention to this, industrial policies may very well not
succeed in the longer run due to insufficiently productive workers, whose labour is com
plementary to industrial upgrading.21
In developing Asia, most regimes can be characterized as wage squeeze. Those that have
adopted extensive industrial policies fall into this category—South Korea, Malaysia, In
donesia, and Thailand. Finally, developed economies have a higher number of mutual
Page 13 of 25
regimes (Finland and Norway, for example), but here too we find wage squeeze (Germany
and Ireland), time squeeze (United States), and exploitation (Poland).
It is apparent from this analysis that industrial policies can succeed by exploiting
women’s time, and/or excluding them from participating in the development process—al
though those that structurally constrain investments in care and social reproduction are
likely to pay a cost in the longer run due to lower labour productivity growth. Ensuring a
win‒win scenario—that is, structural change that induces an (p. 443) economy’s move
ment to higher value-added production with women included in that process—requires at
tention to tendencies towards excluding women from higher-wage and more technically
sophisticated jobs and necessitates appropriate public investment and policies that pro
mote human capacities development by redistributing and reducing the care burden
women differentially shoulder.
16.3.2 Gender and Knowledge Production
Knowledge production, situated in formal education and research, as well as workplace

teams, is vital to the success of industrial policies. Gender inequality and in particular
gender exclusion from these sites, however, can hinder innovation and the acquisition of
knowledge. STEM (science, technology, engineering, and mathematics) education and re
search are essential to industrial policy, and here women are underrepresented, even
though they are now the majority of university students in most countries (UNESCO,
2017).22 The challenge is that exclusion of women results in a selection bias problem with
significant costs in terms of knowledge production and economic growth (European Insti
tute for Gender Equality, 2017). That is, gender-unequal participation in STEM pro
grammes in universities and research institutes can result in overinvestment in less quali
fied males, and failure to capture the benefits of an excluded talent pool of women, which
has costs in terms of knowledge production.23
In part, the low representation of women in STEM fields and research entities is due to
gender-unequal norms and stereotypes that shape attitudes of the gatekeepers—profes
sors, and directors and managers of research institutes. STEM subjects are typically
characterized as ‘male’ and that characterization is not too distant from the facts. The
majority of professors of STEM are male, sending an often unconscious message about
what scientists look like. This both discourages women from entering STEM, and also
causes implicit biases that can lead to subtle and overt forms of discrimination on the
part of male faculty and students that make STEM a hostile environment for women. To
make clear the point advanced here, the loss of women’s talent in STEM, a critical set of
fields for industrial upgrading, may appear to be a labour-supply problem attributed sim
ply to women’s lack of interest in science and engineering. But a more accurate under
standing is that while this is a labour-supply problem, it results from exclusionary prac
tices in both universities and research institutes (and tech-related workplaces) that cre
ate a hostile environment for women.
Page 14 of 25
Learning and upgrading for late industrializers is also predicated on the ability of
(p. 444)
organizations to incentivize and coordinate learning (Amsden, 1989). The composition of

organizations impacts their effectiveness in the problem solving so central to industrial
upgrading. A by now large body of research underscores that diverse organizations func
tion more effectively than homogeneous ones. Economist and complex systems theorist
Scott Page (2017) finds evidence, for example, that identity-diverse groups (e.g. based on
gender, race/ethnicity) can outperform homogeneous groups due to their greater func
tional diversity—that is, differences in how people define problems and approach solving
them.
With regard to the industries that are critical to industrial upgrading, Vasilescu et al.
(2015) similarly find that team diversity in open software development projects correlates
positively with team output. Male and female engineers who collaborate with both gen
ders are roughly twice as productive—that is, they produce more citations in peer-re
viewed articles—than those who only collaborate with one gender (Ghiasi et al., 2015).
Group composition matters, in other words, and in particular, lack of gender diversity in
STEM activities, industries, and organizations can limit knowledge generation.
Given that gender inclusion contributes to success in high-tech organizations, what fac
tors explain women’s low representation? Discrimination, due to implicit or explicit bias,
is an important factor.24 As an example, one study finds that women’s open-source soft
ware coding has a higher acceptance rate than men’s when a coder’s gender is unknown.
But when the gender of the coder is known, acceptance rates of women coders’ proposed
changes to a software project’s code or documentation fall (Terrell et al., 2017). Tech
workplaces more generally are widely reported to be hostile towards women scientists,
with undermining by male managers a key reason given by women for leaving tech jobs
(Mundy, 2017). This state of affairs, predicated on gender hierarchical practices embed
ded in organizations, is a market failure to be addressed by industrial policy.
16.3.3 Gender, Diversity, and Technical Design
Exclusion of groups of people (e.g. women) from technological domains and activities can
lead to their absence as users or consumers of technology because products (p. 445) de
veloped do not reflect women’s needs and preferences. According to Wajcman (2010:
149):
Empirical research on everything from the microwave oven…the telephone…and

the contraceptive pill…to robotics and software agents…has clearly demonstrated
that the marginalisation of women from the technological community has a pro
found influence on the design, technical content, and use of artefacts.
In contrast, drawing on women for design can lead to benefits for women consumers of
technology, whose needs are more likely to be taken into account—and could have the
added benefit of addressing and reducing time required for social reproduction.25
Page 15 of 25
Several examples of the failure of technological innovation to represent the needs of

women are instructive. A wide variety of products marketed as gender neutral are in fact
biased towards men’s bodies: virtual reality headsets too big for women’s heads, smart
watches too big for women’s wrists, and calorie counts on treadmills calibrated to men’s
bodies are just a few examples (Perez, 2019). The problem is particularly serious when it
comes to testing and safety. Although women are more likely to contract HIV than their
male counterparts, they represent just 19 per cent of antiretroviral drug subjects in clini
cal trials (Ovseiko et al., 2016). Crash-test dummies, used to test the safety of automo
biles, are typically male, despite the fact that women are at higher risk of serious injury in
automobile accidents (Holder, 2019).
This discussion underscores that inclusion of women in technical design can generate a
more equitable distribution of the benefits of industrialization. Inclusion of women in the
design of products could also contribute to a reduction of the grind of domestic house
work that women disproportionately bear. Addressing the time squeeze of social repro
duction, as noted in section 16.3, has a positive effect on the long-run success of industri
al policies.
16.4 Conclusion
As a number of chapters in this volume indicate, the conceptualization of industrial poli
cies has broadened, and is beginning to encompass sustainability and greening (p. 446)
policies in addition to the themes of technology, innovation, and knowledge production.
The relationship between gender and industrial policies, however, has not yet been com
prehensively explored or integrated into the industrial policy literature. To address that
gap, this chapter highlights several key channels through which gender and industrial
policy interact: employment, consumption, social reproduction, and knowledge produc
tion and innovation.
Regarding the employment channel, research has identified women’s instrumental role in
providing the labour linked to the success of early-stage late industrialization efforts via
women’s segregation in labour-intensive export manufacturing industries. As countries
move up the industrial ladder, however, women become increasingly excluded as workers
in industry, and are crowded into low-wage employment. Gender exclusion from employ
ment in the dynamic sector of the economy is an important challenge for industrial policy.
With respect to social reproduction, industrializing countries have also relied on women’s
unpaid labour to nurture the development of the human capacities required of a skilled,
productive labour force. This has unduly burdened women, and contributed to the gender
wage and income gap. Further, by failing to recognize and redistribute the burden of un
paid caring labour—or social reproduction—productivity growth is hampered.
The research discussed in this chapter also identifies the economic costs of the absence
of equitable gender inclusion in processes of innovation and technological upgrading. Ex
clusion of women from knowledge-intensive workplaces can negatively affect the efficien
cy of firms’ learning and innovation efforts. Women’s limited representation in high-tech
Page 16 of 25
firms and occupations is often attributed to their preference for fields of studies such as
humanities and the arts, rather than sciences. But this ignores the exclusionary practices
and distorted gender norms and stereotypes that contribute to women’s exclusion from
high-tech workplaces. Further, failure to include women in technology design undermines
the ability of technological innovation to produce benefits for women as consumers.
A challenge, then, is for industrial policy advocates, whether academics or policymakers,

to adopt a gender lens so as to improve the efficacy of industrial upgrading efforts and to
make those efforts more inclusive (Fontana, 2018). The tools that governments use to
nudge, cajole, and encourage the private sector and research institutions to cooperate
with industrial upgrading plans can also be used to address gender disparities identified
in employment. Reciprocal control mechanisms, such as making subsidies and other sup
ports contingent on achieving performance targets, can be used to incentivize gender in
clusion. For example, governments could require (and some already do) that research
teams on government-funded grants be gender balanced as a means of promoting gender
inclusion in research that influences product and technology design.
Governments could also use innovative monetary policy tools to achieve gender inclusion.
Asset-based reserve requirements (ABRRs) are one such mechanism, whereby central
banks use non-interest-bearing reserve requirements to guide the (p. 447) allocation of
credit towards areas deemed of strategic importance.26 Women-owned tech firms, or
firms with an equitable representation of women in high-level positions could be the tar
gets. Of course, state-owned development banks could also target their lending to achieve
strategic goals, including those that promote gender equality.
Social reproduction policies are also legitimately part of an industrial strategy. Publicly
funded physical infrastructure, targeted at reducing women’s care burden in developing
countries, can reduce time poverty, enabling women to join the paid labour force. More
over, other social policies such as publicly funded care of children, the sick, and elderly,
as well as policies that promote male participation in care, are critical to redistributing
and reducing care work, promoting gender inclusion in employment. It also reduces any
potential negative trade-off between women’s increased paid employment and investment
in children’s and workers’ human capacities and, ultimately, long-run productivity
growth. The message, then, for both scholars and policymakers is that the distributional
effects of industrial policy should be understood and interventions designed to ensure
that the benefits are broadly shared. Doing so also contributes to the effectiveness of in
dustrial policy.
Acknowledgements
I am grateful to Christopher Cramer, Fiona Tregenna, Will Milberg, and James Heintz for
providing valuable comments and suggestions on earlier versions of this chapter. Any
omissions or errors are my own.
Page 17 of 25
References
Astorga, Rodrigo, Mario Cimoli, and Gabriel Porcile (2014) ‘The Role of Industrial and Ex
change Rate Policies in Promoting Structural Change, Productivity and Employment’, in
José Manuel Salazar-Xirinachs, Irmgard Nübler, and Richard Kozul-Wright (eds) Trans
forming Economies: Making Industrial Policy Work for Growth. Geneva: International
Labour Office, pp. 79–104.
Bandara, Amarakoon (2015) ‘Economic Cost of Gender Gaps: Africa’s Missing Growth Re
serve’, Feminist Economics 21(2): 162–86.
Berg, Maxine (1992) ‘Women’s Work and the Industrial Revolution’, in Anne Digby,
Charles Feinstein, and David Jenkins (eds) New Directions in Economic and Social Histo
ry, Vol. 2. Basingstoke: Macmillan International Higher Education, pp. 22–36. (p. 448)
Blecker, Robert (2016) ‘Wage-led versus Profit-led Demand Regimes: The Long and the
Short of It’, Review of Keynesian Economics 4(4): 373–90.
Blumer, Herbert (1958) ‘Prejudice as a Sense of Group Position’, Pacific Sociological Re
view 1: 3–7.
Borrowman, Mary and Stephan Klasen (2017) ‘Drivers of Gendered Sectoral and Occupa
tional Segregation in Developing Countries’. Discussion Paper No. 22. Göttingen: Courant
Research Centre.
Braunstein, Elissa (2012) ‘Neoliberal Development Macroeconomics: A Consideration of

its Gendered Employment Effects’. Research Paper No. 2012–1. Geneva: UNRISD.
Braunstein, Elissa, Rashid Bouhia, and Stephanie Seguino (2020) ‘Social Reproduction,
Gender Equality and Economic Growth’, Cambridge Journal of Economics 44(1): 129–56.
Braunstein, Elissa, Irene van Staveren, and Daniel Tavani (2011) ‘Embedding Care and
Unpaid Work in Macroeconomic Modeling: A Structuralist Approach’, Feminist Economics
17(4): 5–31.
Busse, Matthias and Christian Spielmann (2006) ‘Gender Inequality and Trade’, Review of
International Economics 14(3): 362–70.
Cohen, Gerald A. (2001) Karl Marx’s Theory of History: A Defence. Princeton, NJ: Prince
ton University Press.
Cramer, Christopher and Fiona Tregenna (2020) ‘Industrial Policy and the Implications
for Industrial Hubs’, in Arkebe Oqubay and Justin Yifu Lin (eds) The Oxford Handbook of
Industrial Hubs and Economic Development. Oxford: Oxford University Press.
Page 18 of 25
Darity, William, Darrick Hamilton, Patrick Mason, Gregory Price, Alberto Davila, Marie
Mora, and Sue Stockly (2017) ‘Stratification Economics: A General Theory of Intergroup
Inequality’, in Andrea Flynn, Susan Homberg, Dorian Warren, and Felicia Wong (eds) The
Hidden Rules of Race. Cambridge: Cambridge University Press, pp. 35–51.
de Henau, Jerome and Sue Himmelweit (2016) ‘Developing a Macro–Micro Model for
Analysis of Gender Impacts of Public Policy’. Paper presented at Gender and Macroeco
nomics: Current State of Research and Future Directions, Levy Economics Institute, Bard
College, New York City, 9 March.
Elson, Diane (2017) ‘Recognize, Redistribute, and Reduce’, New Left Review 26(2): 52–
61.
Elson, Diane and Ruth Pearson (1991) ‘“Nimble Fingers Make Cheap Workers”: An Analy
sis of Women’s Employment in Third World Export Manufacturing’, Feminist Review 7(1):
87–107.
Epstein, Gerald (2006) ‘Central Banks as Agents of Economic Development’. Research Pa
per No. 2006/54. Helsinki: UNU-WIDER.
European Institute for Gender Equality (EIGE) (2017) Economic Benefits of Gender
Equality in the EU. Paris: EIGE.
Folbre, Nancy (2012) ‘The Political Economy of Human Capital’, Review of Radical Politi
cal Economics 44(3): 281–92.
Fontana, Marzia (2018) Inclusive and Sustainable Industrial Development: The Gender Di
mension. Vienna: UNIDO.
Fontana, Marzia and Luisa Natali (2008) ‘Gendered Patterns of Time Use in Tanzania:
Public Investment in Infrastructure Can Help’. Paper prepared for the IFPRI Project on
Evaluating the Long-Term Impact of Gender-focused Policy Interventions.
Freeman, Joshua (2018) Behemoth: History of the Factory and the Making of the Modern
World. New York: W. W. Norton. (p. 449)
Ghiasi, Gita, Vincent Lariviere, and Cassidy Sugimoto (2015) ‘On the Compliance of
Women Engineers with a Gendered Scientific System’, PLOS One. Available at https://
journals.plos.org/plosone/article?id=10.1371/journal.pone.0145931.
Grassi, Flavia, Josephine Landberg, and Sophia Huyer (2015) Running Out of Time: The
Reduction of Women’s Work Burden in Agricultural Production. Rome: Food and Agricul
ture Organization.
Greenwood, Jeremy, Ananth Seshadri, and Mehmet Yorukoglu (2005) ‘Engines of Libera
tion’, Review of Economic Studies 72(1): 109–33.
Page 19 of 25
Heintz, James (2019) The Economy’s Other Half: How Taking Gender Seriously Trans
forms Macroeconomics. Newcastle upon Tyne: Agenda Publishing.
Holder, Sarah (2019) ‘A Clue to the Reason for Women’s Pervasive Car-safety Problem’,
CityLab. Available at https://www.citylab.com/transportation/2019/07/car-accident-
injury-safety-women-dummy-seatbelt/594049/.
İzdeş, Ozge and Fiona Tregenna (2020) ‘Gender, Industrialization, and Industrial Hubs’,
in Arkebe Oqubay and Justin Yifu Lin (eds) The Oxford Handbook of Industrial Hubs and
Klasen, Stephan and Francesca Lamanna (2009) ‘The Impact of Gender Inequality in Edu
cation and Employment on Economic Growth: New Evidence for a Panel of Countries’,
Feminist Economics 15(3): 91–132.
Kucera, David and Sheba Tejani (2014) ‘Feminization, Defeminization and Structural
Change in Manufacturing’, World Development 64: 569–82.
Mitter, Swasti (2004) ‘Globalization, ICTs, and Economic Empowerment: A Feminist Cri
tique’, Gender, Technology and Development 8(1): 5–29.
Mundy, Liza (2017) ‘Why Is Silicon Valley So Awful to Women?’, The Atlantic Monthly,
April.
Nam, Jeong-Lim (1996) ‘Labor Control of the State and Women’s Resistance in the Export
Sector of South Korea’, Social Problems 43(3): 327–38.
Nelson, Richard and Edward Phelps (1966) ‘Investment in Humans, Technological Diffu
sion, and Economic Growth’, American Economic Review 56: 69–75.
Nübler, Irmgard (2014) ‘A Theory of Capabilities for Productive Transformation: Learning

to Catch Up’, in José Manuel Salazar-Xirinachs, Irmgard Nübler, and Richard Kozul-
Wright (eds) Transforming Economies: Making Industrial Policy Work for Growth. Gene
va: International Labour Office, pp. 113–49.
Ocasio-Cortez, Alexandra (2019) ‘H.Res.109: Recognizing the Duty of the Federal Govern
ment to Create a Green New Deal’. Available at https://www.congress.gov/bill/116th-
congress/house-resolution/109/text.
Ostry, Jonathan, Andrew Berg, and Charalambos Tsangarides (2014) ‘Redistribution,

Inequality, and Growth’. Staff Discussion Note No. 14/02. Washington, DC: IMF.
Ovseiko, Pavel et al. (2016) ‘A Global Call for Action to Include Gender in Research Im
pact Assessment’, Health Research and Policy Systems 14(1). Available at https://
www.ncbi.nlm.nih.gov/pmc/articles/PMC4950803/.
Page, Scott (2017) The Diversity Bonus: How Great Teams Pay off in the Knowledge Econ
omy. Princeton, NJ: Princeton University Press.
Page 20 of 25
Perez, Caroline Criado (2019) Invisible Women: Data Bias in a World Designed for Men.
New York: Abrams Press.
Ridgeway, Cecelia (2011) Framed by Gender: How Gender Inequality Persists in the Mod
ern World. Oxford: Oxford University Press. (p. 450)
469–91.
Samarasinghe, Vidyamali (1998) ‘The Feminization of Foreign Currency Earnings:

Women’s Labour in Sri Lanka’, Journal of Developing Areas 32(3): 303–26.
Saracoglu, Dürdane, Emel Memis, Ebru Voyvoda, and Burça Kizilirmak (2018) ‘Changes
in Global Trade Patterns and Women’s Employment in Manufacturing, 1995–2011’, Femi
nist Economics 24(3): 1–18.
Seguino, Stephanie (2000) ‘Gender Inequality and Economic Growth: A Cross-country

Analysis’, World Development 28(7): 1211–130.
Seguino, Stephanie (2007) ‘Is More Mobility Good? Firm Mobility and the Low Wage Low
Productivity Trap’, Structural Change and Economic Dynamics 18(1): 27–51.
Seguino, Stephanie and Elissa Braunstein (2019) ‘The Costs of Exclusion: Gender Job Se
gregation, Structural Change and the Labour Share of Income’, Development and Change
50(4): 976–1008.
Seguino, Stephanie and Caren Grown (2006) ‘Gender Equity and Globalization: Macro
economic Policy for Developing Countries’, Journal of International Development 18(8):
1091–104.
Standing, Guy (1989) ‘Global Feminization through Flexible Labour’, World Development
17(7): 1077–95.
Tejani, Sheba and William Milberg (2016) ‘Global Defeminization? Industrial Upgrading
and Manufacturing Employment in Developing Countries’, Feminist Economics 22(2): 24–
54.
Terrell Josh, Andrew Kofink, Justin Middleton, Clarissa Rainear, Emerson Murphy-Hill,
Chris Parnin, and Jon Stallings (2017) ‘Gender Differences and Bias in Open Source: Pull
Request Acceptance of Women Versus Men’, PeerJ Computer Science 3: e11. Available at
https://doi.org/10.7717/peerj-cs.111.
Thirlwall, Anthony (1979) ‘The Balance of Payments Constraint as an Explanation of Inter

national Growth Rate Differences’, Banca Nazionale del Lavoro Quarterly Review 32(128):
45–53.
Tilly, Chris (1999) Durable Inequality. Berkeley, CA: University of California Press.
Page 21 of 25
UNCTAD (2016) Trade and Development Report: Structural Transformation for Inclusive
and Sustained Growth. New York: United Nations.
UNCTAD (2018) World Investment Report 2018. New York: United Nations.
UNDP (2013) Humanity Divided: Confronting Inequality in Developing Countries. New

York: UNDP.
UNESCO (2017) Cracking the Code: Girls’ and Women’s Education in Science, Technolo
gy, Engineering, and Mathematics. Paris: UNESCO.
Vasilescu Bogdan, Daryl Posnett, Baishakhi Ray, Mark Van den Brand, Alexander Sere
brenik, Preemkumar Devanbu, and Vkadunur Filkov (2015) ‘Gender and Tenure Diversity
in GitHub Teams’, Proceedings of the 33rd Annual ACM Conference on Human Factors in
Computing Systems, pp. 3789–98.
Wajcman, Judy (2010) ‘Feminist Theories of Technology’, Cambridge Journal of Economics

34: 143–52.
Notes:
(1) On the changing landscape and scope of industrial policy, see Oqubay (Chapter 2, this
volume) and UNCTAD (2018).
(2) See, for example, UNDP (2013) and Ostry et al. (2014).
(3) The feminization of foreign exchange earnings is particularly salient in developing

countries. With few exceptions, countries cannot run persistent current account deficits.
This implies that national income cannot grow faster than the rate at which export
growth at least equals import growth (Thirlwall, 1979). Developing countries, typically de
pendent on imported intermediate and capital goods (for which demand is price inelas
tic), therefore rely heavily on exports to relax the balance-of-payments constraint.
(4) The ‘marriage bar’, once widely practised in Europe and the United States in the
1900s, and more recently by some late industrializing economies such as South Korea
and Taiwan, has resulted in employers terminating women workers upon marriage.
(5) The Industrial Revolution was also characterized by its heavy reliance on women’s fac
tory labour (Berg, 1992; Freeman, 2018).
(6) Perceptions of docility were, however, challenged by women factory workers’ militan
cy, a notable case being that of South Korea in the 1970s and 1980s, to which the govern
ment retaliated with force (Nam, 1996).
(7) This refers to the assumption of greater manual dexterity making women more pro
ductive than men in some types of factory work, like assembly production, and less suit
Page 22 of 25
able for jobs requiring ‘brawn’ such as mining and heavy manufacturing. For a ground-
breaking article on this topic, see Elson and Pearson (1991).
(8) Industry comprises mining, manufacturing, construction, electricity, water, and gas.
(9) Though a related concept, de-feminization is measured differently from gender job
segregation. The former is the percentage of manufacturing employees that are women.
The latter represents how women (relative to men) are distributed across sectors of the
economy, and in particular, in industry as compared to other sectors.
(10) That agricultural employment tends to be of lower quality than industrial employment
is not contested. It might be argued, however, that service-sector jobs can be of high
quality. Although there is some validity to that argument, service-sector jobs are bifurcat
ed in quality with low-wage service jobs typically held by women and racial/ethnic minori
ties and high-quality professional jobs in the financial sector, health services, and product
design dominated by men.
(11) Regional categories are defined using World Bank categorizations in World Develop
ment Indicators.
(12) Data are author’s calculations, based on modelled ILO employment data and reported
in the World Development Indicators. A caveat in interpreting these results is that labour
market statistics in a number of African countries have weaknesses so results should be
interpreted with caution.
(13) Moreover, Borrowman and Klasen (2017), using data on sixty-nine developing coun
tries, find that rising education levels, either overall or of females relative to males, tend
to increase rather than decrease segregation.
(14) Norms are informal ‘rules’ about behaviour that solicit punishment if violated. Stereo
types are generalizations we make about groups of people that may or may not be accu
rate. More generally, they tend to reflect limiting and often negative assumptions about
the characteristics of a particular group of people.
(15) Social reproduction is a concept with origins in Marx’s concept of the development of
productive forces (Cohen, 2001). It includes activities directly involved in the mainte
nance of life on a daily basis and intergenerationally, including the development of chil
dren and the regeneration of workers and is thus a far broader concept than human capi
tal. The latter is defined as capacities developed through formal and informal education
at school and at home, and through training and experience. For a critique of human capi
tal theory from a social reproduction perspective, see Folbre (2012).
(16) Diane Elson (2017) describes the government’s task with regard to work as the 3 Rs
—recognize, redistribute, and reduce.
Page 23 of 25
(17) One channel through which this takes place is that improvements in women’s labour-
market outcomes (both wages and employment) allow them to replace some of their un
paid labour with paid care services, thus stimulating demand.
(18) This could be due to lack of quality market (paid) care services, as well as insufficient
capital goods such as stoves or washing machines.
(19) Argentina, Colombia, and Ecuador also have mutual regimes. For a full list of regime
types in countries by region and at varying levels of development, see Braunstein, Bouhia,
and Seguino (2020).
(20) To be clear, this does not imply wages for women should be lowered in order to pro
mote growth. Macro policies, such as exchange-rate policies to maintain export competi
tiveness, can attenuate negative demand-side effects of higher female wages. Moreover,
managed trade policies and restrictions on financial and firm mobility can create condi
tions for equality-led growth (Seguino and Grown, 2006). And a shift in the structure of
production to reduce reliance on labour-intensive homogeneous export goods and a shift
to skill- and capital-intensive goods production—with women integrated into new indus
tries—can contribute to the creation of a mutual regime. That is because the latter types
of goods and services tend to be price inelastic, such that higher wages have a much
smaller negative effect on demand than on labour-intensive goods.
(21) Among economists noting the complementarity of labour’s skills with industrial up
grading, Nelson and Phelps (1966) have argued that the rate at which the gap between
the technology frontier and the current level of productivity is closed depends on the lev
el of human capital. In a more detailed analysis, Nübler (2014) offers a theory of knowl
edge-based capabilities (a broader concept than human capital) whereby capabilities and
structural change co-evolve and are mutually causative, and applies this to an analysis of
the divergent paths of South Korea and Costa Rica.
(22) It is often argued by way of explanation that women simply choose to study topics
other than science. That view, however, is inconsistent with the fact that women are the
majority of students in the natural sciences and mathematics and statistics, even as they
lag men in engineering, manufacturing, and construction (27 per cent of all students),
and information and communication technologies (28 per cent of all students) (UNESCO,
2017: 20).
(23) Several studies also show that gender inequality in education in general has negative
effects on growth (Bandara, 2015; Klasen and Lamanna, 2009). See also İzdeş and Tre
genna (2020).
(24) Explicit bias refers to conscious attitudes and beliefs we have about people or groups
of people. In contrast, implicit bias is an unconscious association or belief towards a so
cial group. Implicit bias is shaped by social conditioning in response to dominant norms
and stereotypes, such as that women are less good at science, that men are more intelli
gent, and so on. Stratification theorists argue that hierarchical norms and stereotypes are
Page 24 of 25
intentionally cultivated as a means for dominant groups (men) to legitimize their privi
leged position in institutions and broader society (Darity et al., 2017).
(25) The introduction of labour-saving consumer durables (such as washing machines and
vacuum cleaners) reduced the time required for household work, contributing to the rise
of women’s labour-force participation in the twentieth century (Greenwood et al., 2005). I
do not know of any recent studies, however, that measure the pace of household labour-
saving technological change that reduces the time needed for care work. Much of the cur
rent research on this topic focuses on estimating the gender-equalizing benefits of public
infrastructure investment. See, for example, de Henau and Himmelweit (2016) and
Fontana and Natali (2008). There is also some evidence that information and communica
tion technologies (ICT) reduce women’s time burden although, to be clear, ICT develop
ment was not originally targeted to reduce unpaid care burdens (Grassi, Landberg, and
Huyer, 2015). A concern is that technological innovation emerges from the global North
and is ill suited to the needs of women from the global South, who have a greater need
for clean water and electricity than digital technologies (Mitter, 2004).
(26) For more on ABRRs and other innovative tools central banks can utilize to promote
development, see Epstein (2006).
Stephanie Seguino
Stephanie Seguino is professor of economics, University of Vermont, United States

and fellow, Gund Institute for the Environment. Her research explores the relation
ship between intergroup inequality by class, race, and gender on the one hand, and
economic growth and development on the other. She is past president of the Interna
tional Association for Feminist Economics and an associate editor of Feminist Eco
nomics, Journal of Human Development and Capabilities and Review of Keynesian
Economics. Her work has appeared in Cambridge Journal of Economics, Develop
ment and Change, Structural Change and Economic Dynamics, and World Develop
ment, amongst others. She has consulted with numerous international organizations,
including the UNDP, UNRISD, UNCTAD, and the World Bank.
Page 25 of 25
Macro-Policy, Labour Markets, and Industrial Policy

Servaas Storm
Debates on industrialization and industrial policy have historically had a supply-side bias:
development planners focused on strengthening inter-industry linkages, mobilizing sav
ings to finance investment, and the accumulation of technological knowledge. Aggregate
demand was expected to accommodate and even facilitate the structural change brought
about by the industrialization process. However, botched industrialization experiences in
South East Asia, Latin America, and Africa demonstrate that failures to manage demand
in ways supportive of industrial policy can slow or even derail industrialization. We use an
open-economy growth model of a late industrializing economy, featuring cumulative cau
sation and a (long-run) balance-of-payments constraint, to investigate conflicts and com
plementarities between macroeconomic and industrial policies. We identify key macro
mechanisms that undermine industrialization processes—and highlight macro policies in
support of industrial diversification, structural change, and upgrading. We close by argu
ing that from a macro point of view, the widely held claim that labour laws are a ‘luxury’
developing countries cannot afford, is wrong. Labour regulation and higher real wage
growth, when given adequate macroeconomic policy support, can be made to further in
dustrialization.
Keywords: export-led growth, balance-of-payments constrained growth, wage-goods inflation, competitive ex
change rates, big push industrialization, fiscal and monetary policy, labour laws, technology-forcing regulation
17.1 Introduction
A ‘competitive’ exchange rate became Beijing’s de facto industrial policy, by default or de
sign, after China’s WTO membership (in 2001) blocked the country from utilizing more
traditional instruments, such as export subsidies or measures to promote access to for
eign technology (Rodrik, 2008; Guzman et al., 2018). Indeed, maintaining a stable, under
valued exchange rate helped the country increase its exports, and facilitated import sub
stitution and economic diversification, helping China to turn an ‘infant’ manufacturing
sector into a mature, internationally competitive industry that now employs tens of mil
lions of people and serves as a factory for the world. What China’s successful experience
Page 1 of 42
using the exchange rate as a tool for economic development and transformation reveals is
that the ‘right’ management of the macroeconomy can strengthen, supplement, or even
replace the power of traditional industrial policy instruments. The ‘right’ macro policies
‘generate the essential enabling environment for industrial policies to be effective’, write
Guzman, Ocampo, and Stiglitz (2018: 59); in the example of China, macro and industrial
policies have been complementary and mutually reinforcing.1 However, China’s success
contrasts with multiple failures by other late industrializing economies that have been un
able to employ macroeconomic policy in sync with industrial policy. In many such cases,
the opposite occurred: an unduly appreciated exchange rate limited the long-run diversifi
cation of production (p. 452) and export structures, often even leading to ‘premature de-
industrialization’ (Storm, 2015; Diao, McMillan, and Rodrik, 2019). Negative impacts of a
‘mistaken’ macroeconomic policy stance regularly outweigh any positive impacts of indus
trial policy decisions—which brings us to the central argument of this chapter: a failure to
effectively manage the macroeconomy and aggregate demand in ways that support indus
trialization and economic development disables industrial policy, often fatally so.
Macroeconomic mismanagement underlies many historical cases of failed industrializa

tion pushes in South East Asia, Latin America, and Africa. This mismanagement is in no
small measure due to clear supply-side biases of industrial policy, based on the belief that
the impediments to late industrialization were to be found and resolved on the supply
side of the economy. The aggregate-demand side of the economy, in contrast, was expect
ed to more or less passively accommodate and even facilitate the structural changes
brought about by the industrialization process (Storm, 2015). Macroeconomically speak
ing, domestic resource mobilization to finance investment was seen as the binding con
straint on the industrialization programme; hence the key challenge was to raise domes
tic savings supply:
The central problem in the theory of economic development is to understand the

process by which a community which was previously saving, and investing, 4 or 5
per cent of its national income or less, converts itself into an economy where vol
untary savings are running at about 12 or 15 per cent of national income or more.
(Lewis, 1954: 155)
Industrial policy scholars and policymakers have thus been led to focus on the building up
of capital-goods (machine-tool) industries, the strengthening of inter-industry (supply)
linkages, the rationing of scarce (often imported) inputs and foreign exchange, the mobi
lization of enough (domestic) savings to finance industrial development, and the accumu
lation of the technological, managerial and engineering knowledge necessary for efficient
manufacturing. Unfortunately, for most late industrializers things did not work out as
planned—industrialization processes ran into aggregate demand constraints, showing up
in overvalued exchange rates, unsustainable trade deficits and external debts, profit
squeezes, increased inequalities, and domestic demand collapses (Storm, 2015).
Page 2 of 42
This chapter analyses ways in which macroeconomic policy can support, or hurt, the in
dustrialization process, using a simple (Kaldorian) open-economy growth model of a ‘typi
cal’ late industrializing economy, which features ‘cumulative causation’ as well as an ex
plicit balance-of-payments (BoP) constraint. The approach builds on the theoretical analy
ses by Storm and Naastepad (2005), Naastepad (2006), Blecker (2013), and the New
Structuralist tradition which integrates Keynesian models of BoP-constrained growth
(Thirlwall, 1979, 2019; Cimoli et al., 2010) and Kaldorian-Schumpeterian insights into en
dogenous technical progress, innovation, and structural change in productive and export
specialization (Kaldor, 1967; Taylor, 2004; Cimoli and Porcile, 2014; Porcile and Sartorello
Spinola, 2018). The remainder of this chapter is organized as follows. We first present the
export-led growth model, before introducing the (p. 453) long-run BoP constraint in the
model and considering the macro impacts of a traditional (supply-side) ‘big push’ industri
alization drive. We next identify three (historically relevant) macro mechanisms through
which an aggregate demand shortage can slow or even halt the industrialization process:
(a) an ‘agrarian-cum-foreign exchange constraint’, which through higher food prices and
higher industrial wages reduces manufacturing profits and investment (a Ricardian–
Kaleckian story); (b) a drop in domestic (private and/or public) expenditure, which hurts
growth, investment, and diversification (a Robinsonian story); and (c) restrictive fiscal
and monetary policies, often intended to keep imports and inflation down, or to mop up
additional domestic savings, and/or placate foreign financial investors, which has direct
negative impacts on manufacturing investment and growth (a Keynesian story). Section
17.4 considers the ‘primary-commodity specialization trap’ and examines how macroeco
nomic policy can support economic diversification and upgrading (a Kaldorian-Thirlwal
lian narrative). Finally we zoom in on the role of wages and labour regulation and argue
that the widely held claim that strong labour laws are a ‘luxury’ developing countries can
not afford, is wrong from a Weberian-Marxian-Schumpeterian perspective. If anything,
labour regulation and higher real wage growth do further industrialization, when given
adequate macroeconomic and industrial policy support. Section 17.6 summarizes the
lessons learned.
17.2 Setting the Stage: An Export-led Model of

Economic Growth
To organize the discussion we use a model of export-led growth adapted with some modi
fications from Storm and Naastepad (2005, 2012), Naastepad (2006), and Blecker (2013)
and which captures both the central mechanisms of demand-led growth and the critical
(supply-side) constraints on the process of late industrialization. The model has been
specified in aggregate (one-sector) terms to highlight the macroeconomic mechanisms at
work in a typical late industrializing, export-led economy which is open to trade and
cross-border financial flows. Implicitly we assume that the economy has a Lewisian
‘labour surplus’ consisting of underemployed workers in agriculture and other primary in
dustries; real wages in the model reflect the cost of subsistence and are socially deter
mined. The model is expressed in growth rates and defines a stable medium-run growth
Page 3 of 42
path in which the late industrializing economy settles for a period of time, given a certain
set of exogenous structural demand and productivity conditions (Blecker, 2013). For con
venience, all variables are measured in instantaneous rate-of-change form (or differences
in natural logarithms). Table 17.1 provides empirically grounded estimates of the model
parameters, which can be read as ‘stylized facts’ describing the economic structure of
our typical late industrializing economy.
Starting on the demand side, real GDP or output growth (y) is determined by the
(p. 454)
growth of aggregate demand, as in equation (1):
(1)
where a is the growth rate of (real) autonomous demand, x is the growth rate of (real) ex
ports, m is the growth rate of (real) imports, w is (exogenous) real wage growth and z is
labour productivity growth. Coefficients are the shares in GDP of autonomous
demand, exports, and imports, respectively (Table 17.1). If real wage growth exceeds
labour productivity growth (w – z) > 0, the country’s wage share in GDP is growing. We
assume in eq. (1) that wage share growth contributes to faster growth of (p. 455) domestic
demand. This reflects the fact that, at low average levels of income, higher real wage
growth raises consumption growth strongly—and even if it does depress investment
growth, this negative effect is assumed to be more than offset by the faster growth of
consumption demand (see Table 17.3). Autonomous demand growth a is exogenous. Ex
port growth (x) is endogenous:
(2)
where ϵ is the world income elasticity of export demand, g is the growth of rest-of-the-
world income, ρ is the price elasticity of export demand, e > 0 is the rate of nominal ex
change rate depreciation (with the nominal exchange rate measured in home currency
per unit of foreign currency), and pW and p are the rates of change in the foreign and
home price levels, respectively. Note that the expression represents the rate
of real depreciation of the home currency. Foreign price inflation is taken as being exoge
nous (here we are invoking the small-country assumption). Home price inflation p is a
function of the growth in unit labour cost (w – z), the growth of unit intermediate input
cost h and the growth of the profit mark-up factor π: .2 We refrain
from explicitly incorporating changes in mark-ups here and assume that ; and since
the focus is on unit labour cost, we further simplify by assuming that . Under these
assumptions, eq. (2) can be rewritten as:
(2#)
Page 4 of 42
Import growth m is determined by growth of output y and the rate of increase in the rela
tive price of foreign goods :
(3)
Coefficients μ and δ are the income elasticity and the price elasticity of import demand,
respectively. Assuming that and , substituting (2#) and (3) into (1), and rear
ranging, gives the following expression for demand-determined output growth y:
(4)
Demand-driven growth increases when there is an increase in autonomous demand

growth (a) and world income growth (g) and when the currency loses value in nominal
terms. The sign of the impact of higher real wage growth and faster labour (p. 456) pro
ductivity growth can be positive or negative. To see this, consider the derivative of y with
respect to w:
(5)
I assume that the sign of is negative or , that is, the increase in do
mestic demand growth due to an increase in real wage growth is smaller than (the ab
solute value of) the decline in net export growth due to higher wage growth (see Table
17.3). It is in this sense that the late industrializing economy is ‘export led’: lower wage
growth does hurt the domestic market, but this damage is more than offset by its positive
impact on net export growth. Given eq. (5), the impact of higher productivity growth on
output growth must be positive:
(6)
Given that equation (4) describes demand-determined output growth, what about the sup
ply side of this economic system where productivity growth is being determined? The
supply side is given by the Kaldor‒Verdoorn relationship, according to which labour pro
ductivity growth (z) is a positive function of output growth:
Page 5 of 42
(7)
where z0 is exogenous productivity growth due to technology policies (such as R&D subsi
dies and public investment in basic research) and β is the Kaldor‒Verdoorn coefficient
(which reflects dynamic increasing returns to scale; see Storm and Naastepad, 2005,
2012). We assume that .3
Equations (4) and (7) together constitute a system of two linear equations in two endoge
nous variables (z and y) which is illustrated in Figure 17.1. The demand-growth curve is
upward sloping in the (z, y) plane (following eq. (6)); the productivity-growth curve is
(more strongly) upward sloping. Assuming that a stable growth equilibrium exists,4
equilibrium growth (yE) is determined by substituting eq. (7) into eq. (4) and rearranging:
(8)
where the multiplier term for empirically plausible reasons
(see Table 17.1). Given yE, equilibrium productivity growth becomes:
(9)
(p. 457) and we can also determine equilibrium employment growth lE as follows:
(10)
The curve for employment growth also appears in Figure 17.1. Using eq. (8), (9), and (10),
it follows that an increase in a raises equilibrium output, productivity, and employment
growth, because , , and . The same is
true for a rise in g and a nominal currency depreciation. In contrast, higher real wage
growth is not good for ‘export-led’ growth:
(11)
This drop in equilibrium output growth in turn depresses both labour productivity growth
(since ) and employment growth (because ).
Let us finally consider the macro effects of industrial policy, which succeed in bringing
about an increase in productivity growth (i.e. ), and hence contribute to faster net
export and output growth:
Page 6 of 42
(12)
Clearly, productivity growth must go up as well:
(13)
These growth impacts of industrial policy are illustrated in Figure 17.2. The curve for de
mand growth shifts up (in line with eq. (12)) and the curve for productivity growth shifts
down (as per eq. (13)). The new growth path features higher output growth as (p. 458)
well as faster productivity growth. However, the sign of the industrial policy impact on
employment growth is ambiguous, because:
(14)
Equilibrium employment growth could, in principle, increase if output growth rises more
in response to industrial policy than productivity growth. Formally, if
and this could happen only if (i) the price elasticities of exports (ρ) and imports (η) are
large; (ii) unit labour cost makes up a sizeable part of the price (i.e. γ is large); and (iii)
the shares of exports (χ) and imports (η) in GDP are large. Late industrializing economies
do not generally meet these conditions (see Table 17.1), however, and hence, (as
in Figure 17.2). Therefore, industrial policymakers face an unpleasant trade-off: raising
productivity growth increases output growth, but slows down job growth. Storm and
Naastepad (2005), Pieper (2000), Roncolato and Kucera (2014), and Junankar (2013) pro
vide empirical evidence on the ‘productivity growth–employment growth’ for late industri
alizing countries.5 Figure 17.3 illustrates the trade-off using historical evidence for a sam
ple of twenty-four industrializing countries. Only a few East Asian countries managed to
escape the trade-off, but in other Asian and Latin American developing economies a step
up in z has meant a decline in lE.
Page 7 of 42
Table 17.1 Late industrializing economies: stylized facts and parame

ter values
Structural Meaning Source

parameter
Share in GDP of autonomous demand *
Shares in GDP of exports of goods and A

services (1990‒2018)
Shares in GDP of imports of goods and A

services (1990‒2018)
The impact of a one percentage point in B

crease in real wage growth on domestic
demand growth
World income elasticity of export demand C
Income elasticity of import demand D
Relative price elasticity of export demand C
Relative price elasticity of import demand D
*
= ratio of financial inflows to foreign ex
change earnings
Marshall‒Lerner condition:
Share of unit labour cost in output price E
Kaldor‒Verdoorn coefficient F
Page 8 of 42
Growth rate of world income (1990‒2018) A
Growth rate of autonomous demand *
Exogenous (‘technology-push’) labour pro *

ductivity growth
Equilibrium
outcomes ;
Sources: (A) average for the period 1990‒2018 for low- and middle-in
come countries, World Development Indicators, World Bank; (B)
Onaran and Galanis (2014); (C) Senhadji and Montenegro (1998): me
dian of long-run estimates for sixty-six countries (1969‒93); see also
Bussière et al. (2016: table 2); (D) Senhadji (1997): median of long-
run estimates for sixty-six countries (using data for 1969‒93); see al
so Bussière et al., (2016: table 3); (E) γ is defined in fn. 2; we assume
that ; see Table 17.4 below; (F) McCombie, Pugno, and Soro
(2002); Storm and Naastepad (2005: tables 6 and 12); and (*) as
sumed.
Figure 17.1 An export-led growth model with cumu

lative causation
Source: Constructed by author.
Page 9 of 42
Figure 17.2 Successful industrial policy raises equi

librium output growth and productivity growth but
reduces employment growth
Figure 17.3 Trade-off between labour productivity

growth and employment growth in twenty-four late
industrializing countries, 1950‒2003
Notes: (1) Employment is defined as the number of

persons in civilian employment; (2) labour productivi
ty is GDP (at constant prices) per person employed;
(3) labour productivity and employment growth rates
for India, Indonesia, Malaysia, Myanmar, Pakistan,
Singapore, Korea, Sri Lanka, Taiwan, Thailand, and
Vietnam are for 1960‒2003; the figures for China are
for 1953‒2003; and (4) the statistically significant re
gression line was estimated for eighteen countries,
i.e. excluding China, Malaysia, Singapore, Korea, Tai
wan, and Thailand.
Source: Storm and Naastepad (2005).
Macro-policy can be used to overcome this ‘nasty’ trade-off. For example, a nominal cur
rency depreciation will raise output growth (assuming it can be maintained for
some time), as (from eq. (8)) we know that . This means that
the demand-growth curve in Figure 17.2 shifts up even further. Note that ex
(p. 459)
change rate depreciation will also shift the productivity-growth curve to the right. Howev
er, because , productivity growth rises less strongly than output growth—and
Page 10 of 42
employment growth must increase. This can be seen directly by taking the derivative of
employment growth with respect to e:
(15)
It is in this context that capital account regulations (CARs), especially restraints on finan
cial outflows, become a critical tool for economic development (Frenkel and Taylor, 2006;
Guzman et al., 2018). First, CARs can block a currency appreciation in a country with a
persistent current account surplus and help maintain a competitive exchange rate. Main
taining a competitive exchange rate allows developing countries to open up new lines of
(technologically more sophisticated) exports (Freund and Pierola, 2012; Cimoli et al.,
2013). Second, CARs can favour foreign direct investment and discriminate against
volatile short-term (portfolio) flows. Finally, CARs increase monetary policy autonomy by
partly curtailing the interest-rate and exchange-rate effects of (a reversal of) financial
flows.
Eq. (15) suggests that there are good grounds for combining an industrial policy push by
the state with supportive measures by a development-oriented central bank which allow
the currency to depreciate as part of its macro-management of the capital account of the
BoP. With CARs in place, monetary authorities can lower the interest (p. 460) rate, which
would weaken the currency by encouraging a financial outflow out of the country. As an
added benefit, the lowered interest rate will encourage an increase in business invest
ment; in our model, this shows up as an increase in autonomous demand growth. Higher
a increases output growth more than productivity growth, contributing to faster employ
ment growth:
(16)
Alternatively, to prop up employment growth the state could step up public investment—
which again would imply an increase in a. This way, supportive macroeconomic policy
helps to buttress and reinforce the industrial policy initiative. However, if this is true, the
reverse can occur as well: the wrong kind of macro policies, including a sustained curren
cy appreciation,6 an unsupportive increase in interest rates (Frenkel and Taylor, 2006),
and/or wrongly timed fiscal austerity (Neto and Porcile, 2017), can (partly or completely)
erode the potential cumulative growth impacts of industrial policymaking.
17.2.1 Enter the Balance-of-Payments Constraint
Successful industrial policy, when combined with supportive macroeconomic policies,

raises the equilibrium growth rates of output and productivity, while maintaining job
growth. But such a growth acceleration could be inconsistent with the long-run require
Page 11 of 42
ments of BoP equilibrium, if the growth of necessary imports (of food, capital goods, and
technology) exceeds the growth of exports and of financial inflows (Thirlwall, 1979, 2019).
The industrial policy-induced virtuous cycle of faster technological progress, rising invest
ment and exports, and higher output growth may not last, because the country is unable
to finance its growing imports through either export growth or higher financial inflows
(McCombie and Thirlwall, 2004). Following Blecker (2013), the long-run BoP equilibrium
condition can be written as follows:
(17)
where f is the growth rate of financial inflows F (measured in domestic currency) and
is financial inflows as a ratio of total ‘receipts’ in the BoP. We assume that F >
0. Substituting equations (2#), (3), and (9) into (17) gives, after rearranging, the following
general solution for the BoP-constrained growth rate:
(18) (p. 461)
We assume that the denominator of (18) is positive, which appears empirically plausible,
because , (see Table 17.4 below) and the extended Marshall‒Lerner
condition with financial flows is unlikely to hold, because the price
elasticities ρ and δ are in general low for late industrializing countries (Table 17.1). The
expression gives the effect on BoP-constrained growth of an im
provement in the country’s terms of trade. Note that if relative purchasing power parity
holds, then = 0 and the expression drops out of eq. (18).7 While
this may be the case in the long run, it is true that terms-of-trade improvements can raise
the BoP-constrained growth for some time, as illustrated by the commodity-price booms
in Africa (Bagnai et al., 2016) and Latin America (Ocampo, 2017) in the 2000s. But such
terms-of-trade effects have always been transitory (e.g. Nureldin Hussain, 1999). Let us
simplify (18) by accepting ‘price elasticity pessimism’ as a stylized fact of late industrializ
ing nations (Blecker, 2013). That is, we assume that the extended Marshall‒Lerner condi
tion is not satisfied8 and that changes in the international terms of trade have no impact
on the trade balance: . This simplifies eq. (18) to:
(19)
The BoP-constrained growth rate yB increases if world income growth (g) increases and
also following an increase in the growth rate of financial inflows f; in both cases, the BoP
constraint is relaxed and the country can have higher imports which in turn allows for
Page 12 of 42
faster growth. Note that a permanent increase in f is implausible in the long run, particu
larly if the growing financial inflows come in the form of foreign debt and increase faster
than nominal GDP. In this case, the late industrializing country runs the risk of an unman
ageable rise in its foreign indebtedness. Hence if we set f = 0 and further assume that
(which means trade must be balanced in the long run), then eq. (19) becomes:
(20)
Eq. (20) is known as Thirlwall’s Law and it highlights the structural (long-run) determi
nants of a nation’s BoP-constrained growth rate (Thirlwall, 1979, 2019): the growth rate
of world income (g) and the ratio of the world income elasticity of export demand (ϵ) and
the country’s income elasticity of import demand (μ).
This ratio of income elasticities depends on the degree of diversification of the production
structure of this economy as well as on the technological and knowledge intensity of its
pattern of (export) specialization. In general, as is shown by the data (p. 462) (p. 463) col
lected in Table 17.2, ϵ takes a low value for primary and resource-based products and
much higher values for medium- and high-tech manufactures. The more technology inten
sive the export commodity, the higher is its world income elasticity of demand (Gouvêa
and Lima, 2010, 2013; Bottega and Romero, 2019). Accordingly, the more developed, dif
ferentiated, and sophisticated the technological capabilities of a country, the higher the
income elasticity of export demand (ϵ) and hence the higher yB. ‘What you export mat
ters’ is how Hausmann et al. (2007) sum it all up. The policy message is plain: a country
should reorient its exports as much as is possible to fast-growing markets and plan for im
port substitution in those imports for which the income elasticity of demand is high (Thirl
wall, 2019). For many late industrializing nations, the ratio (see Table 17.2),
which means their BoP-constrained rate of economic growth is lower than world income
growth; this is in line with what we assume in Table 17.1. It is precisely here that industri
al policies take on a critical developmental role, because if they succeed in promoting in
vestment and technological learning, and bringing about a diversification and technologi
cal upgrading of the production and export structures, this will raise and consequent
ly yB.9 Conversely, a failure to increase the technological sophistication of the export bas
ket may result in adverse structural change, leading to a decline in and yB. This oc
curred in both Indonesia (Felipe et al., 2019) and Thailand (McCombie and Tharnpanich,
2013), as the export earnings of these countries became more heavily dependent on nat
ural resources and low value-added manufacturing after the Asian financial crisis of 1998.
If we accept that long-run growth of late industrializing countries must be based on long-
run balance in the BoP’s current account, this implies that the equilibrium output growth
yE (determined in eq. (8)) cannot permanently exceed the BoP-constrained growth rate yB
(Blecker, 2013). The reason is that if , the country is running a current account
deficit in excess of what it can structurally finance in the long run; this outcome cannot
Page 13 of 42
be sustained and restoring (long-run) external balance requires that yE is forced down to
yB. On the other hand, if , the country’s current account deficit is below what
long-run BoP equilibrium would permit and there is no need for a realignment of yE and
yB. Figure 17.3 presents the export-led growth model (of Figure 17.1) with the horizontal
BoP constraint—assuming that long-run equilibrium implies: .
Page 14 of 42
Table 17.2 Income elasticity of exports (ϵ) and imports (μ): selected late industrializing countries, 1962‒2006/2014
World income elasticity of export demand ϵ:
Primary Resource- Low-tech Medium- High- Weighted μ

products based nology technolo technolo aggregate
manufac manufac gy manu gy manu ϵ
tures tures factures factures
Argentina 0.66 0.69 0.79 0.82 0.93 0.84 1.24
Brazil 0.62 0.87 1.74 2.23 4.14 1.10 1.18
Chile 1.05 1.10 1.51 2.26 3.85 0.81 1.23
Mexico 0.77 1.12 2.26 2.83 6.91 1.67 1.28
South Ko 0.98 1.25 2.61 1.94 8.07 1.83 1.17

rea
Malaysia 0.77 0.76 8.47 12.22 4.54 1.38 1.22
Philippines 0.76 0.74 2.85 4.81 10.07 1.32 1.21
High-income OECD countries 1.61 1.22
Low-income countries 1.00 1.21
Page 15 of 42
Lower middle-income countries 1.20 1.20
—China 1.55 1.15
—India 1.42 1.14
Upper middle-income countries 1.28 1.21
Sources: The commodity-wise estimates for Argentina, Brazil, Chile, and Mexico (using data for 1962‒2014) are
from Neto and Porcile (2017); the commodity-wise estimates for South Korea, Malaysia, and the Philippines (using
data for 1962‒2006) are from Gouvêa and Lima (2010); all weighted averages for (ϵ) and (μ) are from Gouvêa and
Lima (2013) and based on data for 1965‒99.
Page 16 of 42
17.3 ‘Big Push’ Industrialization

Below a critical minimum of investment, industrial production may fail to benefit from
economies of scale and consequently private returns to individual manufacturing invest
ment turn out to be low, even if the rates of return to coordinated industrial (p. 464) in
vestments would have been high. Industrialization therefore requires a ‘big push’: the co
ordination of complementary investments, in the presence of significant scale economies
and capital market imperfections (Storm, 2015). ‘Big push’ models of (import-substitu
tion) industrialization in BoP-constrained developing countries were proposed by Rosen
stein-Rodan (1943), Mandelbaum (1945), Singer (1950), Nurkse (1953), and Prebisch
(1959). What was assumed in these models is that the demand side of the economy
would, more or less passively, support the industrialization process. Likewise, it was held
that national growth projects could remain unperturbed by global capital markets. The
failure of Latin American import-substitution industrialization (Bertola and Ocampo,
2012) and the running out of steam of India’s Nehru‒Mahalanobis industrialization strat
egy in the 1970s (Chakravarty, 1987) are proof that both assumptions were wrong.
Let us assume that the late industrializing economy raises public investment as part of a
‘big push’ strategy. In the model, this is captured by an increase in a. Its consequences for
the equilibrium growth path of the late industrializing nation are illustrated in Figure
17.4. An increase in a raises output growth from yE0 to . Productivity growth ac
celerates as well (from zE0 to zE1), while employment growth declines from lE0 to lE1, but
remains positive. The supply-driven ‘big push’ strategy appears to be working well. Let us
further suppose that the industrialization programme is not export oriented; indeed, as
Ocampo (2017) argues, state-led industrialization in Latin America during 1950‒80 was
characterized by a significant anti-export bias. We also assume that the programme is not
successful in reducing the nation’s dependency on (medium- and high-tech) imports. Tak
en together, this means that the BoP-constrained growth rate yB remains unaffected by
the ‘big push’, because ϵ and μ remain unchanged. The industrialization process thus runs
into the BoP constraint: the (p. 465) growth acceleration cannot be sustained, because the
country’s current account deficit and external debt are continuously rising relative to
GDP.
How are demand-led growth and BoP-constrained growth reconciled? East Asian late in
dustrializers succeeded in taking the high road, raising yB to yE1. They achieved this by
aggressively encouraging investment in the desired export-oriented and import-substitu
tion activities by every available means, including the offer of tax concessions, subsidies,
temporary and conditional trade protection, cheap credit, and considerable informal pres
sure (Wade, 1990; Akyüz et al., 1998; Amsden, 2001; Storm and Naastepad, 2005). Sci
tovsky (1985) appositely characterized East Asia’s investment regime as a ‘forced invest
ment’ one. But East Asia’s late industrializers are the successful exceptions (Cherif and
Page 17 of 42
Hasanov, 2019). Most other late industrializing countries went down the ‘low road’, as yE1
was ‘forced’ down to equal yB. This happened in three distinct ways.
Figure 17.4 An export-led growth model with cumu

lative causation and a balance-of-payments con
straint
Figure 17.5 Successful industrial policy raises equi

librium output growth above the BoP-constrained
growth rate
17.3.1 The ‘Wage-Goods Inflation’ Barrier: A Ricardian–Kaleckian

Mechanism
‘Everyone knows,’ wrote Ragnar Nurkse (1953: 52–3), ‘that the spectacular industrial
revolution [in England] would not have been possible without the agricultural revolution
that preceded it.’ Arthur Lewis (1954: 173) concurred:
Now if the capitalist sector produces no food, its expansion increases the demand
for food, raises the price of food in terms of capitalist products, and so reduces
profits. This is one of the senses in which industrialisation is dependent upon agri
cultural improvement; it is not profitable to produce a growing volume of manu
factures unless agricultural production is growing simultaneously. This is also why
Page 18 of 42
industrial and agrarian revolutions always go together and why economies in

which agriculture is stagnant do not show industrial development.
As shown by the experiences of South Korea, Taiwan, and China, rapid industrialization
requires parallel or prior increases in agricultural productivity and food production to
avoid the danger of running into the Ricardian trap (Saith, 1990; Storm, 2015). In our
BoP-constrained economy (Figure 17.4), the ‘big push’-induced acceleration of output
growth raises the demand of the industrial sectors for intermediate inputs and wage
goods (food and other subsistence goods) produced by the domestic agricultural sector.
Domestic agriculture may be unable to increase the growth rate of wage-goods supply,
even in the presence of a Lewisian reserve of underemployed workers, because the
growth of agricultural productivity and output is constrained by technical as well as insti
tutional factors (notably a highly unequal distribution of land ownership, the small size of
average land holdings, and exploitative and insecure tenancy systems)—and therefore the
growth of wage-goods supply is relatively price inelastic even in the longer run
(Karshenas, 2001; Kay, 2002; Storm, 1996, 2015).
Under these conditions, as argued by Kalecki (1955, 1972), faster (industrial) growth
leads to a structural excess demand for wage goods and to wage-goods inflation— (p. 466)
which (in the presence of a BoP constraint) will drive up nominal and real wage growth
(in manufacturing) and this, in turn, will raise the (industrial) wage share, lower the (in
dustrial) profit share and thus erode private investment (Chakravarty, 1987). Attempts to
finance industrial investment through a forced extraction of resources from a technologi
cally stagnant and socially unequal agriculture would inevitably dissipate in inflationary
spirals. Note that Kalecki’s diagnosis that wage-goods inflation arising from the (price) in
elasticity of agricultural supply could lower aggregate demand and output growth is one
of the earliest references in the literature to the phenomenon that came to be known as
‘stagflation’. In the model, the Ricardian–Kaleckian wage-goods inflation mechanism
would work through an increase in real wage growth w, which lowers yE (per eq. (11))
and shifts the demand-growth curve in Figure 17.5 down.10 It is possible, by combining
the expression for yB given by eq. (20) and eq. (8) for yE, to determine the ‘BoP-con
strained growth rate for real wages’:
(21)
wB is the growth rate of real wages which ensures long-run balance between yE and yB.
Noting that , it can be seen from (21) that , which means that
BoP-constrained real wage growth is forced up by the wage-goods inflation, which
(p. 467)
in turn depresses domestic demand for manufactures and thus stifles industrial growth.
This particular mechanism was at work in large, populous, and foreign-exchange-con
strained countries such as Brazil in the 1960s and 1970s (Taylor and Bacha, 1976; Kay,
2002) and India in the 1970s and 1980s (Chakravarty, 1987; Storm, 1996, 2015), when do
Page 19 of 42
mestic demand for manufactures was eroded by wage-goods inflation caused by the fact
that under-prioritized agriculture was unable to produce enough food and raw materials
for the growing industrial workforce and expanding manufacturing sector.11
17.3.2 A Drop in Domestic Expenditure Growth: A Robinsonian Mech

anism
A second mechanism realigning yE with (unchanged) yB has been suggested by Porcile

and Yajima (2019), who following Robinson (1966) propose that the growth of private-sec
tor (investment) demand will go down. Robinson argued that a growing external deficit is
likely to depress the ‘animal spirits’ of the capitalist class, and therefore private invest
ment, because it signals a growing risk of a BoP crisis, a depreciation of the currency, a
rising foreign debt burden and/or a sudden stop in foreign credit:
The most important benefit of a surplus on income account [of the BoP], which af
fects the whole economy, is that, provided there are energetic enterprises and
thrifty capitalists to take advantage of it, it permits home investments to go full
steam, while a deficit country is nervously pulling on the brake for fear of exces
sive imports.
(Robinson, 1966: 16‒17)
This fall in private investment, so Robinson adds, slows down technical progress and
hurts the country’s competitive position, tightening the BoP constraint even further. If
there is a currency depreciation, this could hurt firm profits (assuming that firms cannot
completely protect their profit mark-ups and pass higher import costs on to prices)—and
private investment would go down. In the present model, autonomous expenditure
growth a would fall, which shifts the demand-growth curve down to where it was before
the public investment drive. In this adjustment process, higher public investment ‘crowds
out’ private investment, not through higher interest rates, but rather through depressed
‘animal spirits’ of private firms and/or through a manufacturing profit squeeze.
17.3.3 Forced Savings, Restrictive Monetary Policy, and Desta

(p. 468)
bilizing Austerity: Three Keynesian Mechanisms
In terms of macroeconomic accounting, a structural current account deficit reflects the

fact that the country is unable to increase domestic savings to the same level as the ex
panded domestic (public + private) investment—and hence the country has to rely on for
eign savings (a net financial inflow) to finance its capital accumulation. For most late in
dustrializing countries, access to global capital markets is restricted and it is in this con
text of a shortage of foreign exchange that early development models—including the Har
rod‒Domar model, the Lewis model and the Mahalanobis model—emphasized the impor
tance of mobilizing domestic savings to finance capital formation growth. A failure to do
so can, in a foreign-exchange-starved economy, lead to a process of forced savings
(Storm, 2015), as the excess of demand allows firms to raise the profit mark-up factor (π
> 0), which then leads to home price inflation and a redistribution of income in favour of
Page 20 of 42
profit recipients (who have a higher propensity to save than wage earners). The higher
profit share contributes to rising inequality in household incomes and higher savings,
which reduces domestic demand and, by limiting the size of the home market, puts a
brake on industrial investment, output, and productivity growth, as argued by Kalecki
(1955, 1972) and Taylor and Bacha (1976). Lack of domestic demand slows down industri
alization, forcing yE down to yB.
This drag on growth can be avoided for as long as the current account deficit can be fi
nanced by foreign funds—but foreign indebtedness will rise and the external imbalance
will have to be rectified at some point. Exchange rate devaluations will generally not be
effective at doing this (Bahmani et al., 2013). The only way to reduce the external imbal
ance is by import compression—but to achieve this, demand must be depressed. Fiscal
austerity is one way to depress demand, and it was part and parcel of the World Bank/
IMF structural adjustment programmes adopted by, and imposed on, debt-ridden industri
alizing countries in Latin America and sub-Saharan Africa in the 1980s and 1990s. In our
model, fiscal austerity means a decline in variable a, which by depressing growth and im
ports restores BoP equilibrium. The demand-growth curve shifts down again to where it
was before the public investment drive (Figure 17.5). In Latin America, the cure of fiscal
austerity and pro-market reforms proved to be worse than the disease. As large ex
change-rate devaluations were not effective in reducing current account deficits, Latin
American governments imposed drastic cuts in public expenditure (mostly in public in
vestment) and raised taxes in order to continue servicing their high external debts (Berto
la and Ocampo, 2012). Private investment predictably suffered—and economic growth
stagnated, industrial diversification and upgrading got stalled, and Latin American
economies became more strongly dependent on the export of a few (primary) commodi
ties. As Neto and Porcile (2017: 48) write, ‘a fall in public investment reinforces back
wardness in technology, infrastructure and productivity, which heightens the external
deficit and disequilibrium’. Austerity reinforced Latin America’s comparative advantage
in natural resource- and labour- (p. 469) abundant sectors, led to a re-commodification of
their production and export structures (Ocampo, 2017), and resulted in considerable de-
industrialization (Diao et al., 2019).
Persistent current account deficits will also alarm monetary policy authorities, who will
raise interest rates in an attempt to mop up additional domestic savings and/or attract (or
keep attracting) foreign finance. If successful, the central bank’s restrictive monetary pol
icy raises domestic savings (thereby lowering consumption) and depresses (or ‘crowds
out’) private investment—and as a result, domestic demand and import demand are
forced down; this will also realign yE with yB. These mechanisms are called ‘Keynesian’
because they all constitute a failure to manage aggregate demand in ways supportive of
the big push industrialization process.
Page 21 of 42
17.4 Vicissitudes in the Life of Export-led In

dustrialization Strategies
The challenge for late industrializing countries is to escape from the primary-commodity
specialization trap in which they find themselves (Ocampo, 2017; Cherif and Hasanov,
2019). The question is how to defy comparative advantage, and how far from comparative
advantage the state should push. To illustrate, South Korea’s development path after
1960 would have been radically different if it had followed the policy advice it was given
at the time to continue focusing on what were then its main exports (rice, raw silk and
wolfram), instead of embarking on developing shipbuilding, electronics and automotive
industries (Cherif and Hasanov, 2019). In principle, a commodity-price boom could pro
vide the launch pad for an industrialization drive and a structural transformation of the
economy. Exactly such an opportunity for industrialization was created during 2003‒13
as the windfall export gains (caused by a global commodity-price boom) and a (China-led)
surge in natural resource-seeking foreign direct investment revived economic growth in
Africa and Latin America (Ocampo, 2017). This commodity boom turned African
economies into the fastest growing on the planet (Bagnai et al., 2016) and this, for some
time, led to much optimism about prospects for accelerated industrialization based on
commodities. But overall, sub-Saharan African and Latin American countries have been
unable to fully capture the benefits from commodity booms. Industrialization based on
commodities is not at all easy—mostly because macroeconomic imbalances can throw the
growth process off the rails.12 (p. 470)
17.4.1 The ‘Dutch Disease’
Let us explore the macroeconomic consequences of a prolonged commodity-price boom in

terms of the growth model of Figure 17.4, by assuming that the growth rate of world
prices pW increases. The higher export earnings constitute a windfall gain, and assuming
(for the moment) that these are not invested in projects intended to diversify and upgrade
the country’s production and export structures, yB stays the same. Note further that a
higher pW does not affect the long-run BoP-constrained growth rate yB, because we as
sume . However, the commodity-price boom drives up equilibrium
output growth, since:
(22)
and as a result, the demand-growth curve shifts up as is illustrated in Figure 17.6. The
labour productivity growth does not shift, but through the Kaldor‒Verdoorn mechanism,
productivity rises (along the curve) from zE0 to zE1. Employment growth increases as well
(from lE0 to lE1), because output growth increases more than productivity growth (since
). The commodity-price hike thus triggers a process of cumulative growth.
Page 22 of 42
Before proceeding, we should note that while the late industrializing nation is enjoying a
surplus in its current account due to the increase in export earnings, the step-up in its
economic growth has raised the growth rate of the country’s imports; the growth rate of
real exports has remained unchanged. Accordingly, in real terms, the country is running a
larger current account deficit than is allowed by the BoP constraint (eq. (19)). Hence,
notwithstanding the fact that the country is enjoying a temporary current account surplus
(in nominal terms), it is running a trade deficit in (p. 471) real terms—which it is able to fi
nance using the higher export earnings only for as long as the commodity-price boom en
dures.
But the new equilibrium growth path is unlikely to last, if the monetary authority fails to
maintain a ‘competitive’ exchange rate. Global demand for the country’s currency in
creases as a consequence of the export commodity-price boom, and the nominal ex
change rate will appreciate . A ‘non-competitive’ exchange rate may strangle the
development of the tradable manufacturing industries. The currency appreciation re
aligns equilibrium growth yE with yB as in Figure 17.6. The currency appreciation (a fall in
e) shifts the demand-growth curve downward, as , and the re
source-rich country remains locked into the commodity-specialization trap. The mecha
nism is known as the ‘Dutch disease’,13 and is one manifestation of a larger pathology
summarized under the heading of ‘natural-resource curse’: countries with abundant nat
ural resources tend to have lower and more unstable economic growth than countries
with fewer natural resources (Venables, 2016; Ocampo, 2017). A big part of the reason for
this is that coping with massive fluctuations in export earnings (or with huge inflows of
foreign finance) is challenging for any government.14 An escape from this trap is possible
only when the country succeeds in diversifying and upgrading its production and export
structures. For Hans Singer (1950) and Raúl Prebisch (1959) it was clear that this re
quired strict and careful management of the BoP in the context of a long-run strategy of
import-substitution industrialization.
Figure 17.6 The macroeconomic impacts of a com

modity-price boom
Page 23 of 42
17.4.2 Escaping the Primary-Commodity Specialization Trap
Economic growth yE in our late industrializing economy can only be increased in a lasting
way if the country succeeds in simultaneously raising its BoP-constrained growth rate yB.
The BoP-constrained growth rate is a reflection of the developing economy’s (initial) posi
tion in the international division of labour, specializing in natural resources and agricul
tural exports and importing mostly medium- and high-tech manufactured goods. But a
country’s pattern of specialization is not given—to paraphrase the famous dictum of Joan
Robinson—by God and factor endowments15 (Porcile and Sartorello Spinola, 2018), but
can be changed and improved by deliberate, state-guided strategies to change the struc
ture of production towards activities with (p. 472) higher productivity—which means pro
ductively absorbing existing new technologies, catching up with the technological leading
countries (as described most clearly by Gerschenkron, 1962), mobilizing resources and
guiding investment to more sophisticated industries (Wade, 1990), prioritizing credit to
targeted industries (with the strongest linkages and spillovers), enforcing export orienta
tion, maintaining a ‘competitive’ exchange rate (Guzman et al., 2018) and managing the
BoP, and learning how to manufacture by just doing it and keeping on doing it (as de
scribed most powerfully by Amsden, 2001). To capture the power of a developmental
state to (gradually but critically) improve the pattern of specialization of the economy, we
modify our growth model following Cimoli and Porcile (2014), assuming that:
(23)
Equation (23) defines progressive structural change in the economy, away from natural
resources-based activities and towards manufacturing, as a function of mission-oriented
public investment in infrastructure and more risky, innovation-related activities (Mazzu
cato, 2013; Neto and Porcile, 2017) and state-guided private investment (Wade, 1990;
Amsden, 2001), which are both elements of autonomous demand growth a. A successfully
coordinated public and private investment drive will diversify the production system, re
ducing the country’s import dependence (which lowers μ), while raising the share in ex
ports of more technology-intensive manufacturing goods (which raises ϵ). A higher ratio
raises the country’s yB in eq. (20). The relaxation of the BoP constraint next allows the
late industrializing country to step up its demand-determined rate of growth.
This is illustrated in Figure 17.7. Initially, the economy grows at . The state then
initiates a big public investment programme and imposes effective industrial policy guid
ance on business investment—both elements of this industrialization strategy lead
(p. 473) to successful diversification and upgrading of national production and export
structures. In Figure 17.7, higher public investment (higher a) pushes up the demand-
growth curve. But this time (and after a while), the industrial policy thrust raises and
yB shifts up as well. The economy can now settle in a new, higher-growth equilibrium
, which will be determined by how much industrial policy manages to raise .
Page 24 of 42
Equilibrium productivity growth rises along the productivity-growth curve from zE0 to zE1
in line with Kaldor–Verdoorn relation (in eq. (7)). Figure 17.7 further shows that employ
ment growth is higher in the novel growth equilibrium. A final point to note is that the
structural relaxation of the BoP constraint lowers the BoP-constrained growth rate of real
wages wB. To see this, we differentiate wB in eq. (21) with respect to the ratio of income
elasticities of demand and get:
(24)
which is negative because (see Table 17.1). What eq. (24) tells us
is that the structural loosening of the BoP constraint has lowered the pressure of wage-
good inflation in the economy—the spectre of stagflation has been exorcized by success
ful industrial diversification and the system has more space for demand-led development.
However, actual real wage growth can be higher than before, because equilibrium pro
ductivity growth has increased. Workers gain because employment growth and real wage
growth will rise.
Figure 17.7 Escaping the primary-commodity spe

cialization trap
17.5 Labour Laws as Instruments of Industrial

Policy
Successful late industrialization that steers clear of persistent external disequilibria and
an unsustainable external debt, requires the pattern of specialization to be re-oriented to
wards more technology-intensive industries with higher income elasticities of demand for
exports. The market mechanism is unlikely to bring about the growth-enhancing structur
al change that is needed (Storm, 2015) and hence the ‘leading and guiding hand of the
state’ (Amsden, 2001; Wade, 2018; Cherif and Hasanov, 2019) must design and use (selec
tive) industrial policy instruments to target exports with growth potential and identify im
Page 25 of 42
ports where there is potential for import substitution (Storm and Naastepad, 2005; Thirl
wall, 2019). Crucially, late industrializing nations can complement their industrial policy
by using labour regulation designed in such a way that it serves as a ‘beneficial con
straint’, forcing firms to upgrade, diversify, and become more productive (Storm and Ca
paldo, 2019; Storm, 2019). This suggestion may appear outlandish: after all, how can the
late industrializing economy benefit from labour laws and regulations, as these will just
raise labour costs and prices, and therefore harm net exports, business profits, and firm
investment?
It is true that not so long ago, it was received opinion that labour protections for
(p. 474)
ordinary workers were a ‘luxury developing countries cannot afford’, because ‘laws creat
ed to help workers often hurt them’, as the World Bank (2008: 8) concluded. However, it
is now clear that this ‘perversity thesis’, while remaining influential among ‘madmen in
authority, hearing voices in the air’, is no longer empirically tenable. Meta-analyses of the
empirical literature on labour regulation and economic performance by Nataraj et al.
(2014) and Broecke et al. (2017) find that the effects of regulation on growth and employ
ment are generally small or absent, and negligible compared to those on income distribu
tion. The reason why the ‘perversity thesis’ is wrong is that it sees labour laws as exclu
sively driving up unit labour cost, while neglecting their considerable positive impacts on
domestic demand, productivity growth, and diversification and upgrading (Storm and Ca
paldo, 2019; Storm, 2019).
17.5.1 Wages and Demand
Higher real wage growth (in combination with stable and growing employment) raises
consumption growth and expands the home market (Ocampo et al., 2009; Storm, 2015;
UNIDO, 2017). However, if wage growth exceeds labour productivity growth, this raises
unit labour cost and prices, which damages net exports and consequently business profits
and investment. It follows that the macro impact of higher wage growth is the net out
come of faster domestic demand growth, captured in the model by the coefficient ξ in eq.
(8), and lower net export growth, reflected by the expression in eq. (8)
(Bhaduri and Marglin, 1990). So far we have assumed that , mean
ing that on balance, higher real wage growth does lower economic growth. While the
negative sign is empirically plausible, we must recognize that the macro impact of higher
real wage growth in most late industrializing nations is surprisingly small (in absolute
terms) and should not be exaggerated.
The point is illustrated by Table 17.3, which presents internationally comparable econo
metric evidence on ξ and for seven major late industrializing economies
(Onaran and Galanis, 2014). First, it can be seen that a one percentage point increase in
real wage growth raises domestic demand growth by 0.35 percentage points on average
(ξ > 0) for six countries (excluding South Africa). Second, a one percentage point in
crease in real wage growth reduces net export growth by ‒0.35 percentage points on av
erage (when we exclude the estimate for China).16 This limited sensitivity of net export
growth to real wage growth is perhaps surprising but the reasons are not difficult to un
Page 26 of 42
derstand. To begin with, as illustrated by Table 17.4, (nominal) unit labour costs (multi
plied by the profit mark-up) make up only one-third of the price (p. 475) level—which in
terms of our model means that ν, defined in footnote 1, takes a value of ⅓ and hence γ is
½ (Table 17.4). This means that a one percentage point increase in real wage growth will
increase home price inflation by around 0.5 percentage points, on the assumption that
firms decide to pass on the full increase in wages. However, empirical evidence indicates
that this is not what firms actually do: operating under (p. 476) competitive conditions,
firms are found to pass only around 50 per cent of the increase in wages on to final
prices, absorbing the other half in lower profit mark-ups.17 This then means that a more
realistic value for γ is around ¼—which explains why the impact of higher real wage
growth on net export growth is relatively limited (in Table 17.3). Taking Table 17.3’s find
ings on ξ and together, the net impact of higher real wage growth on eco
nomic growth is about zero—as is indeed the case in Argentina, India,
Mexico, and South Korea. In sum, aggregate economic growth is not very sensitive to
variations in real wage growth, and consequently, the effects of labour regulation on em
ployment must also be small. This corroborates the conclusions of World Bank re
searchers Kuddo et al. (2015) that the effects of regulation on growth and employment
are much smaller than the heat of the debates suggests, and of the IMF (2016: 115) that
‘reforms that ease dismissal regulations with respect to regular workers do not have, on
average, statistically significant effects on employment and other macroeconomic vari
ables’. It follows that attempts to industrialize by a strategy of ‘beggaring one’s
neighbour’ (suppressing national real wage growth below wage growth abroad) are just
pointless.
17.5.2 ‘Technology-forcing’ Labour-Market Regulation
Labour laws may constitute ‘beneficial constraints’ on firms’ decision-making (Streeck,

2004), which, far from undermining markets and capitalism, raises the static as well as
the dynamic efficiency of economic activity in distinct ways. A first—Weberian—efficien
cy-enhancing mechanism operates by improving legitimacy. Since labour-market institu
tions inject fairness into the employment contract (by limiting inequality and facilitating
risk sharing), they create goodwill and political legitimacy amongst workers and enhance
productivity, perhaps as an unintended consequence (Lee and McCann, 2011). A second
channel through which labour regulations enhance productivity is a Marxian one: higher
real wage growth (made possible by regulation) raises the rate of labour-saving technical
progress. Such ‘Marx-biased technical progress’ can be included in eq. (7):
(25)
Page 27 of 42
Table 17.3 Impact on economic growth of a one percentage point in

crease in real wage growth, selected late industrializing countries
Change in Change in net Change in eco

domestic export growth nomic growth
demand
growth ξ
Ar 0.14 ‒0.19 ‒0.05

genti
na
China 0.41 ‒1.99 ‒1.57
India 0.29 ‒0.31 ‒0.02
Mexico 0.28 ‒0.38 ‒0.10
South 0.02 ‒0.51 ‒0.49

Africa
South 0.42 ‒0.36 0.06

Korea
Turkey 0.49 ‒0.28 0.21
Source: Onaran and Galanis (2014). The estimation period is 1970‒

2007 for all countries (but 1978‒2007 for China).
Page 28 of 42
Table 17.4 Share of wages in the gross output price, selected regions and countries
Labour-in Profit share Share of inter

come share mediate in
puts in gross
output
Africa 0.467 0.533 0.54* 0.28 0.39
South Africa 0.515 0.485 0.54 0.30 0.44
Latin America 0.495 0.505 0.51* 0.33 0.48
Brazil 0.587 0.413 0.51 0.36 0.37
East Asia 0.514 0.486 0.69* 0.19 0.23
China 0.495 0.505 0.69 0.18 0.23
South East Asia 0.416 0.584 0.51* 0.29 0.40
India 0.544 0.456 0.51 0.35 0.53
Note: *The regional estimate is assumed to be equal to the corresponding country estimate.
Page 29 of 42
Sources: Labour-income shares and profit shares are from ILOSTAT https://ilostat.ilo.org/topics/labour-income/. Da
ta on the share of intermediate inputs in gross output are based on the input‒output tables of Brazil, China, India,
and South Africa, OECDSTAT https://stats.oecd.org/Index.aspx?DataSetCode=IOTS. See footnote 2, this chapter, for
an explanation of the symbols.
Page 30 of 42
where ϑ represents the impact of higher real wage growth on labour productivity growth
through labour-saving innovations. Empirical evidence on ϑ for a sample of developing
countries suggests that ϑ takes a value of around 0.3.18 Using eq. (25), the derivative of
equilibrium growth with respect to real wage growth becomes:
(26)
Higher real wage growth may still continue to depress export-led growth, but the
(p. 477)
negative effect dwindles (in absolute terms) the closer ϑ is to unity. A third and final chan
nel through which labour regulation may enhance labour productivity, competitiveness,
and industrial upgrading is a Schumpeterian one, which operates by spurring innovation
(Streeck, 2004). Tougher labour rules favour the stronger, more productive firms, as
these will change work practices and upgrade production technology, and in the process
these firms displace established, but less productive, competitors. This ‘cleansing’ or
‘technology-forcing’ effect of labour-market standards has been observed to have been at
work in China (Huang et al., 2014) and India (Acharya et al., 2013).
From a Weberian-Marxian-Schumpeterian perspective, it makes sense to use labour stan

dards deliberately designed to ‘force’ firms to comply with technological norms that are
not currently viewed as economically feasible. This way, labour regulation and resulting
higher wages can be made to contribute to export diversification:
(27)
Eq. (27) expresses the ‘technology-forcing’ potential of labour market regulation and
higher real wage growth: higher wage growth forces firms that can no longer compete on
low wages and low-tech (labour-intensive) products to diversify and move into higher-tech
more capital-intensive goods, and as a result of this, increases—raising the country’s
BoP-constrained rate of growth in turn. If the ‘technology-forcing’ labour regulation is
combined with a mission-oriented public investment programme (an increase in a) to di
versify and upgrade exports, we are back to Figure 17.7, as this will push up the BoP-con
strained growth rate yB even more. Properly designed labour regulation thus reinforces
the process of progressive structural change through which the late industrializing coun
try can free itself from the shackles of primary-commodity specialization and climb up the
industrial ladder.
17.5.3 The Wage and the Profit Rate
To hammer home the point that labour regulation and higher wages are not automatically
antithetical to business interests, let me define the rate of profit (ρ) as the real return on
invested capital, as follows:
Page 31 of 42
(28)
where π = (Π/Y) = the share of real profits (Π) in real income (Y), = capacity
utilization, and = the ‘normal’ output‒capital ratio. K is the capital stock (at
constant prices) and is ‘normal’ (trend) output. κ is assumed to be a long-run constant.
Eq. (28) can be extended using this definition of the profit share π (Storm and Naastepad,
2012):
(29) (p. 478)
where θ = the wage share, W = the real wage, and Z = the level of labour productivity.
Substituting (29) into (28) gives this decomposition of ρ:
(30)
The business profit rate thus has three determinants: W, Z and u. A higher real wage rais
es the wage share, reduces the profit share and therefore decreases ρ. Higher productivi
ty, on the other hand, reduces the wage share, increases the profit share and therefore
raises the profit rate. And if demand and capacity utilization go up, this must raise ρ. We
use eq. (30) to assess the impacts of a higher real wage rate on firms’ profitability. First, a
higher real wage rate directly and one-for-one reduces the profit share and hence the
profit rate. Second, higher real wages have positive and negative impacts on aggregate
demand and growth—as in eq. (26). We assume that (Table 17.3). Third, higher
real wages spur labour-saving technical progress (via Weberian–Marxian–Schumpeterian
channels, as in eq. (25)), and thus raise z in eq. (30). Finally, the ‘technology-forcing’ po
tential of higher wages and labour regulation raises and hence yB. Demand-led
growth yE will adjust (going up), pushing up capacity utilization u in eq. (30). If the ‘tech
nology-forcing’ labour regulation is combined with public investment to diversify and up
grade exports, this will further push up yB, yE, and u. Empirically, the sign of the net im
pact of a higher wage on the profit rate ρ is probably negative, but (as Figure 17.8 makes
clear), the impact may be surprisingly small (in absolute terms), if the Marx-biased tech
nical progress channel and the ‘technology-forcing’ mechanism are strong. Plus, with a
little help from fiscal stimulus, which raises u and hence ρ, higher real wages can be
made compatible with rising business profit rates. The ‘perversity thesis’ does not hold
water.
Page 32 of 42
Figure 17.8 The (+/–) impacts of a higher real wage

on the profit rate (ρ)
17.6 Macro-Policy Lessons for Late Industrial

ization
If developing countries are to improve their long-run growth performance, they must in
dustrialize, building up a broad and technologically sophisticated domestic industrial
base with strong potential for high productivity, income, and export growth, in
(p. 479)
ways that improve the trade balance and raise BoP-constrained growth. The challenge is
to develop industries with growth potential, in defiance of static comparative advantage,
and to keep pushing innovation, structural change, and industrial diversification to the
next level of sophistication. Fiscal, monetary, exchange-rate and labour-market policies
must be designed in ways that support the industrialization strategy—otherwise, the in
dustrial policy drive will be frustrated by growing external deficits and unsustainable
debts, or alternatively, by processes of domestic stagflation or austerity-induced deflation.
We can derive a few guiding principles for macroeconomic policymaking based on our
model analysis, but with the qualification that successful macro-management requires
pragmatism, and rule-based ‘one-size-fits-all’ approaches should be avoided.
A first lesson is that it is critical for late industrializing nations to use CARs to maintain
competitive exchange rates, because a competitive exchange rate can strengthen the ef
fectiveness of industrial policy and reward export orientation. CARs also increase mone
tary policy autonomy, increasing the policy space for central banks to use interest rates
and prioritized credit allocation in support of industrialization. But undervalued exchange
rates cannot permanently sustain growth transitions. Furthermore, we must recognize
that undervaluing one’s exchange rate is essentially a mercantilist, ‘beggar-thy-neigh
bour’ strategy, which will be counter-productive if adopted by many countries at the same
time. Second, fiscal policy is of paramount importance as an enabler of industrial policy—
public investment increases long-term (productivity) growth, helps diversification and
raises employment, while stabilizing short-term demand. Progressive taxation, luxury
consumption taxes, import tariffs (to protect infant industries), counter-cyclical export
taxes and taxes on foreign financial inflows (of equity and portfolio finance) can be used
to mobilize resources for financing public investment in physical and social infrastructure
Page 33 of 42
in non-inflationary ways, which helps to resolve the limited supply-responsiveness of criti

cal wage goods-producing sectors such as agriculture.19 Wrongly timed fiscal austerity
can be destabilizing and may undermine the industrialization process.
Third, persistent trade deficits and the threat of a sudden financial outflow may force cen
tral banks to raise interest rates, but overly restrictive monetary policy will trigger a re
cession. The way for the monetary authorities to maintain a ‘development-oriented’ policy
regime, consistent with a targeted exchange rate and accommodating to fiscal and indus
trial policy, is to impose exchange controls and restrictions on financial outflows. A ‘devel
opment-oriented’ monetary policy also implies that inflation targeting cannot be the only
goal of central bank policy. Fourth, wage restraint is fundamentally ineffective and unde
sirable as a strategy for trying to improve external competitiveness and raise growth.
Sustained growth of domestic demand requires sustained growth of real wages—which
raises capacity utilization and profit rates. Labour regulations can be designed as being
‘technology forcing’ in nature, and thus (p. 480) support and amplify the impacts of indus
trial policy aimed at export diversification and upgrading. However, the biggest lesson to
be learned from the analysis is that successful macro-management of late industrializing
economies requires discretion, rather than rigid fiscal, monetary, or exchange rate policy
rules. What is required are pragmatism, a flexibility to deal with country-specific condi
tions and historical context, and above all, as John Maynard Keynes wrote in a letter to
Roy Harrod (dated 4 July 1938), ‘vigilant observation of the actual working of our system’
in order to be able to choose good models. Elaborating his point, Keynes wrote:
Economics is a science of thinking in terms of models joined to the art of choosing

models which are relevant to the contemporary world. It is compelled to be this,
because, unlike the typical natural science, the material to which it is applied is, in
too many respects, not homogeneous through time. The object of a model is to
segregate the semi-permanent or relatively constant factors from those which are
transitory or fluctuating so as to develop a logical way of thinking about the latter,
and of understanding the time sequences to which they give rise in particular cas
es.
The open-economy, BoP-constrained growth model used here has been shown to be rele
vant to the contemporary world, not least because it has highlighted the critical impor
tance of a development-oriented management of aggregate demand for successful late in
dustrialization.
References
Acharya, Viral V., Ramin P. Baghai, and Krishnamurthy V. Subramanian (2013) ‘Labour
Laws and Innovation’, Journal of Law and Economics 56(4): 997‒1037.
Akyüz, Y., Ha-Joon Chang, and Richard Kozul-Wright (1998) ‘New Perspectives on East
Asian Development’, Journal of Development Studies 34(6): 4‒36.
Page 34 of 42
Amsden, Alice H. (2001) The Rise of ‘The Rest’: Challenges to the West from Late-Indus
trializing Economies. New York: Oxford University Press.
Bagnai, Alberto, Arsène Reiber, and Thi Anh-Dao Tran (2016) ‘Sub-Saharan Africa’s
Growth, South–South Trade and the Generalised Balance-of-payments Constraint’, Cam
Bahmani, Mohsen, Hanafiah Harvey, and Scott W. Hegerty (2013) ‘Empirical Tests of the
Marshall–Lerner Condition: A Literature Review’, Journal of Economic Studies 40(3):
411‒43.
Bertola, Luis and José Antonio Ocampo (2012) The Economic Development of Latin Ameri
ca since Independence. Oxford: Oxford University Press.
Bhaduri, Amit and Stephen A. Marglin (1990) ‘Unemployment and the Real Wage: The
Economic Basis for Contesting Political Ideologies’, Cambridge Journal of Economics
14(4): 375‒93.
Blecker, Robert A. (2013) ‘Long-run Growth in Open Economies: Export-led Cumulative

Causation or a Balance-of-payments Constraint?’, in Geoffrey C. Harcourt and Peter
Kriesler (eds) Oxford Handbook of Post-Keynesian Economics. Oxford: Oxford University
Press, pp. 390‒414. (p. 481)
Bottega, Ana C. and João P. Romero (2019) ‘Innovation, Export Performance and Trade
Elasticities across Different Sectors and Countries’. CCEPP Working Paper 04–19. Cam
bridge: University of Cambridge, Department of Land Economy.
Broecke, Stijn, Alessia Forti, and Marieke Vandeweyer (2017) ‘The Effect of Minimum
Wages on Employment in Emerging Economies: A Survey and Meta-analysis’, Oxford De
velopment Studies 45(3): 366‒91.
Bussière, Matthieu, Guillaume Gaulier, and Walter Steingress (2016) ‘Global Trade Flows:
Revisiting the Exchange Rate Elasticities’. Working Paper No. 608. Paris: Banque de
France.
Chakravarty, Sukhamoy (1987) Development Planning: The Indian Experience. Oxford:

Clarendon Press.
Named: Principles of Industrial Policy’. Working Paper No. WP/19/74. Washington, DC:
IMF.
Cimoli, Mario and Gabriel Porcile (2014) ‘Technology, Structural Change and BoP-con
strained Growth: A Structuralist Toolbox’, Cambridge Journal of Economics 38(1): 215‒
37.
Page 35 of 42
Cimoli, Mario, Sebastian Fleitas, and Gabriel Porcile (2013) ‘Technological Intensity of
the Export Structure and the Real Exchange Rate’, Economics of Innovation and New
Technology 22(4): 353‒72.
Cimoli, Mario, Gabriel Porcile, and Sebastian Rovira (2010) ‘Structural Change and the
BoP Constraint: Why Did Latin America Fail to Converge?’, Cambridge Journal of Econom
ics 34(2): 389‒411.
Diao, Xinshen, Margaret McMillan, and Dani Rodrik (2019) ‘The Recent Growth Boom in
Developing Economies: A Structural Change Perspective’, in Massiko Nissanke and José
Antonio Ocampo (eds) The Palgrave Handbook of Development Economics. London: Pal
grave Macmillan, pp. 281–334.
Elbadawi, Ibrahim A., Linda Kaltani, and Raimundo Soto (2012) ‘Aid, Real Exchange Rate
Misalignment, and Economic Growth in Sub-Saharan Africa’, World Development 40(4):
681‒700.
Felipe, Jesus, Matteo Lanzafame, and Gemma Estrada (2019) ‘Is Indonesia’s Growth Rate
Balance-of-payments Constrained? A Time-varying Estimation Approach’, Review of Key
nesian Economics 7(4): 537–53.
Frenkel, Roberto and Lance Taylor (2006) ‘Real Exchange Rate, Monetary Policy and Em
ployment’. UN-DESA Working Paper No. 19. New York: United Nations.
Freund, Caroline and Martha Denisse Pierola (2012) ‘Export Surges’, Journal of Develop
ment Economics 97(2): 387‒95.

Gouvêa, Raphael Rocha and Gilberto Tadeu Lima (2010) ‘Structural Change, Balance-of-
payments Constraint and Economic Growth: Evidence from the Multi-sectoral Thirlwall’s
Law’, Journal of Post Keynesian Economics 33(1): 169‒204.
Gouvêa, Raphael Rocha and Gilberto Tadeu Lima (2013) ‘Balance-of-payments-con

strained Growth in a Multisectoral Framework: A Panel Data Investigation’, Journal of
Economic Studies 40(2): 240‒54.
Guzman, Martin, José Antonio Ocampo, and Joseph E. Stiglitz (2018) ‘Real Exchange Rate
Policies for Economic Development’, World Development 110(1): 51‒62.
Hausmann, Ricardo, Dani Rodrik, and Jason Hwang (2007) ‘What You Export Matters’,
Journal of Economic Growth 12(1): 1–25. (p. 482)
Holz, Carsten and Aaron Mehrotra (2013) ‘Wage and Price Dynamics in a Large Emerg
ing Economy: The Case of China’. BIS Working Paper No. 409. Basel: Bank for Interna
tional Settlements.
Page 36 of 42
Huang, Yi, Prakash Loungani, and Gewei Wang (2014) ‘Minimum Wages and Firm Em
ployment: Evidence from China’. IMF Working Paper No. WP/14/184. Washington, DC:
IMF.
IMF (2016) ‘Time for a Supply-side Boost? Macroeconomic Effects of Labor and Product
Market Reforms in Advanced Economies’, World Economic Outlook 2016. Washington,
DC: IMF, pp. 101–42.
Junankar, Pramod N. (2013) ‘Is There a Trade-off between Employment and Productivi
ty?’. IZA Discussion Paper No. 7717. Bonn: Institute for the Study of Labor.
University Press.
Kalecki, Michal (1955) ‘The Problem of Financing Indian Economic Development’, Indian
Economic Review 2(3): 1‒22.
Kalecki, Michal (1972) ‘Problems of Financing Economic Development in a Mixed Econo

my’, in Michal Kalecki, Selected Essays on the Economic Growth of the Socialist and
Mixed Economy. Cambridge: Cambridge University Press, pp. 145‒61.
Karshenas, Massoud (2001) ‘Agriculture and Economic Development in Sub-Saharan

Africa and Asia’, Cambridge Journal of Economics 25(3): 315‒42.
Kay, Cris (2002) ‘Why East Asia Overtook Latin America: Agrarian Reform, Industrializa
tion and Development’, Third World Quarterly 23(6): 1073‒102.
Kuddo, Arvo, David A. Robalino, and Michael Weber (2015) ‘Balancing Regulations to Pro
mote Jobs: From Employment Contracts to Unemployment Benefits’. Working Paper No.
101596. Washington, DC: World Bank.
Lee, Sangheon and Deirdre McCann (eds) (2011) Regulating for Decent Work: New Direc
tions in Labour Market Regulation. London: Palgrave Macmillan.
Lewis, W. Arthur (1954) ‘Economic Development with Unlimited Supplies of Labour’, The
Manchester School 22(2): 139‒91.
Mandelbaum, Kurt (1945) The Industrialisation of Backward Areas. Oxford: Blackwell.
McCombie, John S. L. and Nat Tharnpanich (2013) ‘Balance-of-payments-constrained

Growth, Structural Change, and the Thai Economy’, Journal of Post Keynesian Economics
35(4): 569–95.
McCombie, John S. L. and Anthony P. Thirlwall (eds) (2004) Essays on Balance of Pay
ments Constrained Growth: Theory and Evidence. London: Routledge.
Page 37 of 42
McCombie, John S. L., Maurizio Pugno, and Bruno Soro (2002) Productivity Growth and
Economic Performance: Essays on Verdoorn’s Law. London: Palgrave Macmillan.
Naastepad, C. W. M. (2006) ‘Technology, Demand and Distribution: A Cumulative Growth

Model with an Application to the Dutch Productivity Growth Slowdown’, Cambridge Jour
nal of Economics 30(3): 403‒34.
Nataraj, Shanti, Francisco Perez-Arce, Sinduja V. Srinivasan, and Khrisna B. Kumar

(2014) ‘The Impact of Labor-market Regulation on Employment in Low-income Countries:
A Meta-analysis’, Journal of Economic Surveys 28(3): 551‒72.
Neto, Antonio S. M. and Gabriel Porcile (2017) ‘Destabilizing Austerity: Fiscal Policy in a
BoP-dominated Macro-dynamics’, Structural Change and Economic Dynamics 43(1): 39‒
50. (p. 483)
Nureldin Hussain, Mohammed (1999) ‘The Balance of Payments Constraint and Growth
Rate Differences amongst African and East Asian Countries’, African Development Review
11(1): 103–37.
Nurkse, Ragnar (1953) Problems of Capital Formation in Underdeveloped Countries. New

Ocampo, José Antonio (2017) ‘Commodity-led Development in Latin America’, in Gilles

Carbonnier, Humberto Campodónico, and Sergio Tezanos Vázquez (eds) Alternative Path
ways to Sustainable Development: Lessons from Latin America. International Develop
ment Policy Series No. 9. Geneva: Brill-Nijhoff, pp. 51‒76.
Ocampo, José Antonio, Codrina Rada, and Lance Taylor (2009) Growth and Policy in De
Onaran, Özlem and Giorgos Galanis (2014) ‘Income Distribution and Growth: A Global
Model’, Environment and Planning A 46(10): 2489‒513.
Pieper, Ute (2000) ‘Deindustrialization and the Social and Economic Sustainability Nexus
in Developing Countries: Cross-country Evidence on Productivity and Employment’, Jour
nal of Development Studies 36(4): 66‒99.
Porcile, Gabriel and Danilo Sartorello Spinola (2018) ‘Natural, Effective and BoP-con
strained Rates of Growth: Adjustment Mechanisms and Closure Equations’, PSL Quarter
ly Review 71(285): 139‒60.
Porcile, Gabriel and Giuliano Toshiro Yajima (2019) ‘New Structuralism and the Balance-
of-payments Constraint’, Review of Keynesian Economics 7(4): 517–36.
Prebisch, Raúl (1959) ‘Commercial Policy in Under-developed Countries’, American Eco

nomic Review, Papers and Proceedings 492: 251–73.
Page 38 of 42
Robinson, Joan (1966) The New Mercantilism: An Inaugural Lecture. Cambridge: Cam
Rodrik, Dani (2008) ‘Second-best Institutions’, American Economic Review 98(2): 100‒4.
Rodrik, Dani (2018) ‘An African Growth Miracle’, Journal of African Economies 27(1): 10–
27.
Roncolato, Leanne and David Kucera (2014) ‘Structural Drivers of Productivity and Em
ployment Growth: A Decomposition Analysis for 81 Countries’, Cambridge Journal of Eco
nomics 38(2): 399‒424.
Rose, Andrew K. (1991) ‘The Role of Exchange Rates in a Popular Model of International
Trade: Does the “Marshall–Lerner” Condition Hold?’, Journal of International Economics
30(3–4): 301‒16.
Rosenstein-Rodan, Paul N. (1943) ‘Problems of Industrialisation of Eastern and South

Eastern Europe’, The Economic Journal 53(210/211): 201‒11.
Saith, Ashwani (1990) ‘Development Strategies and the Rural Poor’, Journal of Peasant
Studies 17(2): 171‒244.
Scitovsky, Tibor (1985) ‘Economic Development in Taiwan and South Korea 1965–81’,
Stanford Food Research Institute Studies 19(3): 1–50.
Senhadji, Abdelhak (1997) ‘Time-series Estimation of Structural Import Demand Equa

tions: A Cross-country Analysis’. IMF Working Paper No. WP/97/132. Washington, DC:
IMF.
Senhadji, Abdelhak and Claudio Montenegro (1998) ‘Time-series Estimation of Export De
mand Equations: A Cross-country Analysis’. IMF Working Paper No. WP/98/149. Washing
ton, DC: IMF.
Singer, Hans W. (1950) ‘The Distribution of Gains between Investing and Borrowing Coun
tries’, American Economic Review, Papers and Proceedings 40(2): 473‒85. (p. 484)
Souza, Joao Paulo de (2017) ‘Real Wages and Labor-saving Technical Change: Evidence
from a Panel of Manufacturing Industries in Mature and Labor-surplus Economies’, Inter
national Review of Applied Economics 31(2): 151‒72.
Storm, Servaas (1996) ‘Domestic Constraints on Export-led Growth: A Case Study of In
dia’, Journal of Development Economics 52(1): 83‒119.
Storm, Servaas (2015) ‘Structural Change’, Development and Change 46(4): 666‒99.
Storm, Servaas (2019) ‘Labor Laws and Manufacturing Performance in India: How Priors
Trump Evidence and Progress Gets Stalled’. Working Paper No. 90. New York: Institute
for New Economic Thinking.
Page 39 of 42
Storm, Servaas and Jeronim Capaldo (2019) ‘Labour Institutions and Development under
Globalization’, in Massiko Nissanke and José Antonio Ocampo (eds) The Palgrave Hand
book of Development Economics. London: Palgrave Macmillan, pp. 553‒87.
Storm, Servaas and C. W. M. Naastepad (2005) ‘Strategic Factors in Economic Develop

ment: East Asian Industrialization 1950–2003’, Development and Change 36(6): 1059‒94.
Storm, Servaas and C. W. M. Naastepad (2012) Macroeconomics beyond the NAIRU.

Streeck, Wolfgang (2004) ‘Educating Capitalists: A Rejoinder to Wright and Tsakalotos’,

Socio-Economic Review 2(3): 425‒38.
Taylor, Lance (2004) Reconstructing Macroeconomics: Structuralist Proposals and Cri

tiques of the Mainstream. Cambridge, MA: Harvard University Press.
Taylor, Lance and Edmar Bacha (1976) ‘The Unequalizing Spiral: A First Growth Model
for Belindia’, Quarterly Journal of Economics 90(2): 197–218.
Thirlwall, Anthony P. (1979) ‘The Balance of Payments Constraint as an Explanation of In

ternational Growth Rate Differences’, Banca Nazionale del Lavoro Quarterly Review 128:
45‒53.
Thirlwall, Anthony P. (2019) ‘Thoughts on Balance-of-payments-constrained Growth after

40 Years’, Review of Keynesian Economics 7(4): 554–67.
UNIDO (2017) Industrial Development Report 2018. Demand for Manufacturing: Driving
Inclusive and Sustainable Industrial Development. Vienna: UNIDO.
Venables, Anthony J. (2016) ‘Using Natural Resources for Development: Why Has It
Proven So Difficult?’, Journal of Economic Perspectives 30(1): 161‒84.
Wade, Robert H. (1990) Governing the Market: Economic Theory and the Role of Govern
Wade, Robert H. (2018) ‘What Room for the Developmental State in an Era of Financial
ization, Global Production Chains and Knowledge Monopoly?’, Development and Change
49(2).
World Bank (2008) Doing Business: An Independent Evaluation Report. Washington, DC:
World Bank.
Notes:
(1) While a currency devaluation may initiate growth acceleration (as we argue later in
this chapter), it is unlikely to sustain it indefinitely without supportive industrial and fis
cal policies (Storm and Naastepad, 2005; Rodrik, 2018; Thirlwall, 2019).
Page 40 of 42
(2) This equation can be derived as follows. First, assume that the price level P=W̿Z+H,
where Π= 1+ the profit mark-up, W̿ = the nominal wage, Z = the level of labour produc
tivity, and H = intermediate input cost per unit of output. Expressed in instantaneous
rate-of-change form (differences in natural logarithms), we get this equation for home
price inflation: p = π + ν w̿-z + 1 - ν h. ν = W̿Z–1P or the ratio of unit labour cost, the profit
mark-up factor and the price level. Nominal wage growth w̿ is defined as w̿ = w + p; sub
stituting this into the equation for home price inflation and rearranging, we obtain: p = π
+ γ w – z + h, where γ = 1–ν.
(3) Empirical evidence on the Kaldor‒Verdoorn coefficient for China, Malaysia, Singapore,
South Korea, Taiwan, and Thailand by Storm and Naastepad (2005) shows that β=0.4.
(4) The system solves for a unique and stable equilibrium, as we assume the slope of the
productivity-growth curve to be steeper than the slope of the demand-growth curve.
(5) Long-run econometric estimates by Storm and Naastepad (2005) and Junankar (2013)
for a panel of low- and middle-income countries indicate that dlE/dz0 ≈ –1. We obtain the
same elasticity value here, using the stylized facts in Table 17.1.
(6) Elbadawi et al. (2012) provide evidence that currency overvaluations had a negative
impact on export diversification for eighty-three sub-Saharan countries (1970‒2004).
(7) The econometric evidence on long-run relative purchasing power parity is inconclu
sive, and is sensitive to the currencies, price indices, time periodization, and econometric
methods used (Blecker, 2013).
(8) For empirical evidence, see Rose (1991) and Bahmani et al. (2013).
(9) The point is illustrated by comparing the world’s most successful industrializer China
(ϵ/μ = 1.4 in Table 17.2) and a less successful industrializer such as Brazil (ϵ/μ = 0.9). For
a given rate of growth of world income (of, say, 3.5 per cent), China’s yB will be 4.9 per
cent compared to 3.2 per cent in Brazil.
(10) This can be justified as follows. The real wage is w = W̿/P; we can write:
, where PF = the price of wage goods (food). The term is the
‘product wage’, or the nominal wage deflated by the price of wage goods (Chakravarty,
1987). If wF is socially determined and fixed and if food prices rise faster than the general
price level P, then w must increase. This is the mechanism underlying eq. (21).
(11) Note that in Latin America, Africa, and India, the resulting more unequal distribution
of rural incomes restricted the growth of the domestic market, contributed to a distorted
industrial structure and (via imports of consumer goods) contributed to chronic BoP prob
lems (Kay, 2002; Karshenas, 2001; Chakravarty, 1987).
(12) Cycles in commodity prices introduce (enhanced) cyclicality in fiscal revenues (be
cause government revenue is heavily dependent on taxation of natural resource-intensive
industries) and in business investment (which tends to increase as commodity prices go
Page 41 of 42
up). Short-run stabilization emergencies (for instance in response to a sudden collapse of

commodity prices) may seriously conflict with a long-run policy focus on industrialization
and structural change (Ocampo et al., 2009).
(13) The term was coined in 1977 by The Economist to describe the decline in manufactur
ing in the Netherlands after the country had become a big natural gas producer and ex
porter in the 1960s.
(14) Part may be due to weak governance, which has, in some cases, created space for pa
tronage politics and domestic conflicts. See Venables (2016).
(15) In 1970, only three South Korean export industries within the ‘machinery and trans
port equipment’ category (at the SITC 4-digit level) had a revealed comparative advan
tage index greater than one, singling them out as promising sectors for future develop
ment. But by 1990, all three had disappeared as export industries, while Korea’s top ex
porting industries in 1990 did not even exist in 1970 (Cherif and Hasanov, 2019).
(16) China is the ‘outlier’ in Table 17.3: its net export growth is found to decline by almost
two percentage points in response to a one percentage point increase in real wage
growth. But note that Onaran and Galanis (2014) find much higher price elasticities of ex
ports and imports for China than Bussière et al. (2016) and Bottega and Romero (2019).
(17) See Holz and Mehrotra (2013) for evidence on China.
(18) See Storm and Naastepad (2005) and De Souza (2017).
(19) Agrarian modernization requires more than just public investment. East Asia’s di
rigiste states relied on egalitarian technology policy (hybridization and chemical fertiliz
ers), infrastructure and irrigation investment, rural credit, and radical institutional and
land reforms (Storm, 2015).
Servaas Storm
Servaas Storm is a senior lecturer at Delft University of Technology. He obtained a

PhD in economics from Erasmus University Rotterdam. His work has appeared in
scholarly journals including Cambridge Journal of Economics, Development and
Change, Eastern Economic Review, Industrial Relations, International Journal of Po
litical Economy, Journal of Post-Keynesian Economics, and Structural Change and
Economic Dynamics. His latest book, co-authored with C.W.M. Naastepad, is Macro
economics beyond the NAIRU (Harvard University Press, 2012), winner of the 2013
Myrdal Prize of the European Association for Evolutionary Political Economy. He is
one of the editors of Development and Change.
Page 42 of 42
Technological Disruptions, GVCs, and Industrial Policy

David Bailey and Lisa De Propris
This chapter examines the impact of technological change on global value chains (GVCs)
and what initiatives and instruments governments in advanced economies can deploy to
support firms and people during the transition. Drawing on an emerging debate on de-
globalization, we discuss how global production is slowly shifting from being organized in
GVCs to continental platforms with shorter and geographically closer relationships as
firms seek to co-locate manufacturing and innovation activities. This offers regions and
places the opportunity to upgrade and transform their economies and thereby to anchor
high-technology industries, leveraging industrial legacy with frontier technologies. We
will discuss the implications for a transformative place-based industrial policy that aims to
connect embedded industries to new technologies; to repopulate embedded industries
with new firms and start-ups, and to use regulation and procurement to create new mar
kets and allow exploration.
Keywords: industrial policy, global value chains, Fourth Industrial Revolution, new technologies, manufacturing,
place-based industrial policy
18.1 Introduction
THE organization of production inside firms and along the supply chain change to re
spond to disruptions in the technology and in the nature and geography of demand. A
wealth of literature going back to the 1970s has dissected the drivers of firms’ production
internationalization strategies (the ‘why’) and related operations (the ‘how’) (Hymer,
1960; Dunning, 2000; Mathews, 2006). Such internationalization strategies of multina
tional corporations (MNCs) have been driven by the contribution that different ‘places’
can make to MNCs’ overall division of value. The globalization of production activities
that emerged from the combined offshoring of manufacturing activities created so-called
global value chains (GVCs) (Gereffi and Fernandez-Stark, 2011). Consequently, the Smith
sonian division of labour became a division of ‘value additive functions’, whereby each
production stage contributes to a higher or lower extent to the value addition of the final
Page 1 of 22
output. This space-neutral approach has been criticized by Scholte (2000) who argued
that the globalization of production activities has ‘de-territorialized’ production choices.
The global organization of production in GVCs started to be challenged with the global fi
nancial crisis in 2008–09 when it became clear that two decades of mass offshoring had
left a painful mark on some regions and places in advanced economies, which were de-in
dustrialized or hollowed out. The socio-economic costs of offshoring were unveiled and
this prompted a process of soul searching that has, belatedly, led to a more critical ap
proach to globalization (Bailey and Tomlinson 2017).
At the same time, the global economy found itself at the end of a technological cycle with
incremental innovations increasingly exhausting technological opportunities (p. 486)
(Rifkin, 2011). A wave of new technologies is emerging associated with digital and green
technologies and these are expected to disrupt firms, value chains, regions, markets, and
ultimately the whole economy.
Against this backdrop, this chapter explores the impact of technological change on GVCs
and what initiatives and instruments governments in advanced economies can deploy to
support firms and people during the transition. There are signs that the global economy is
‘de-globalizing’ with firms seeking to co-locate manufacturing and innovation activities;
this offers an opportunity to regions and places to upgrade and transform their
economies, leveraging industrial legacy with frontier technologies. However, this is not
going to happen automatically, as has been discussed with—for example—so-called
‘reshoring’ in developed economies; rather, an active industrial policy is required (Bailey
and De Propris, 2014; Bailey et al., 2018a, 2018b).
In this chapter we discuss the implications for a transformative industrial policy that we
see comprising three objectives: (1) to connect embedded industries to new technologies;
(2) to repopulate embedded industries with new firms and start-ups; and (3) to use regu
lation and procurement to create new markets and allow exploration.
18.2 Technological Change and the Fourth In

dustrial Revolution
A wave of new technologies is driving a shift in our techno-socio-economic paradigm
(Pérez, 2010). These new technologies are expected be pervasive and disruptive, chang
ing not only the organization of production inside the firm and thereby the workplace by
altering the balance between capital and labour, but also re-drawing the organization of
production between firms and along the value chain whether local or global. These new
technologies include a range of advances in biotech, nanotech, neuro-technologies, green
and renewables, ICT and mobile tech, cloud technology, big data, 3D printing, artificial in
telligence, Internet of things, robotics, sensoring, space technology, and drones. Some of
the current debate on the impact of these technologies on the economy and society has
drawn on the seminal work of Kondratiev (1935) on technological waves to argue that
Page 2 of 22
they amount to a Fourth Industrial Revolution (FIR) (De Propris, 2016; OECD, 2017a;
Rifkin, 2011; Schwab, 2016).
The First Industrial Revolution coincided with the British Industrial Revolution between
the last decades of the eighteenth century and the first half of the nineteenth century
with new technologies related to steam power, cotton, steel, and railways, coupled with
mechanization and the surge of the factory system. The Second Industrial Revolution was
triggered by the introduction in the post-war period of electricity, heavy and mechanical
engineering (automotive and aviation) and synthetic chemistry (oil-based production).
New business forms emerged as large firms started producing away from the domestic
market, becoming multinational firms. The Third Industrial Revolution coincided with the
introduction of electronics and computers, petrochemicals and (p. 487) aerospace in the
last decades of the twentieth century. Volatile demand required more flexible organiza
tional forms, such as firm clusters and industrial districts, which became global hubs due
to their competitive advantage (De Propris, 2016). New technologies also enabled faster
communications and transport, kicking off a process of globalization in production that
led to multiplications of GVCs (Gereffi, 1999).
The ‘digital revolution’ argument is captured by the debate on Industry 4.0 (Deloitte,
2015; European Commission, 2016; PWC, 2016) and on technologies enabling the emer
gence of smart and connected factories. Often referred to as cyber-physical spaces (GTAI,
2014), production places will see humans working side by side with digitally connected
robots or other forms of automation with the novelty that the latter will communicate
with each other rather than just being operated by humans. This will transform the flow
and use of inputs in the factory, with capital investment in automation expecting to re
place some human tasks. Therefore, in factories we expect automation to displace labour-
intensive productions for increasingly complex and non-routine functions.
While FIR technologies will bring about a disruptive change that has been wrongly re
duced by some to simply a ‘digital revolution’ linked to artificial intelligence and the In
ternet of things (Deutsche Bank, 2014), we believe that as a technological revolution, its
360-degree transformative power will be felt not only in the world of production but also
in the world of consumption and usage and in everybody’s ways of life.
To fully appreciate the transformational shift that the deployment of new technologies
will have on firms’ value creation, it has been argued that a broader understanding of the
impact that new technologies will have must be acknowledged; on this De Propris and
Bailey (2020) propose a more holistic conceptualization of Industry 4.0 that takes into ac
count five interlinked transformations brought about by new technologies (see Figure
18.1).
Page 3 of 22
FIGURE 18.1 Industry 4.0+
Source: Authors.
The concept of Industry 4.0+ allows us to understand changes in the nature of business
and markets (Bailey and De Propris, 2019). These transformations are:
a. the creation of new radical markets

b. the design and manufacture of radically new products
c. the adoption of new business models
d. the experimentation of new product-service innovations
e. the exploration of new customer-centred innovations.
These transformations will possibly trigger two major new trends: one is a recoupling of
innovation and production; and two, the scaling-down of production for niche markets. In
particular, radically new products can create radically new markets that either address
new demand needs or push new technology-driven offers. Either way, radical innovations
require exploration, experimentation, and testing that are likely to call for a recoupling of
innovation with production. As new products are launched or tested in the market, a
smoother, direct, closer, more frequent, and more meaningful interaction between innova
tion and production will be necessary. The experimental nature of these high-tech mar
kets and products means that a co-location between innovation and production is likely to
emerge. Other disruptive transformations arising (p. 488) from the FIR include the serviti
zation of manufacturing and changes in firms’ business models. The servitization of man
ufacturing introduces a product–service innovation (Bustinza et al., 2018; Dimache and
Roche, 2013; Baines et al., 2017) that is linked to the choice of businesses and customers
to move from a product-based business model to a service-based business, whereby buy
ers overcome the need to own products and rather prefer to hire or lease them. As the re
lationship between producer and user is woven around the specific characteristics of the
servitized good, again such relationship intensifies with constant exchanges that in this
latter case extend beyond the moment when user receives the ‘good’ because they are
not buying ‘the good’ but the function; that is, the usefulness and the service attached to
it (Vendrell-Herrero and Bustinza, 2020). As the servitized good is designed, produced
and supplied around the specifications of the user, the relation between innovation, pro
duction, and consumption becomes closer. The need for an innovation–production–con
Page 4 of 22
sumption continuum may result in firms strategically experimenting with new business
models where co-innovation with consumers occurs in the design of products—with per
sonalized offers—and in delivery with elements of servitization.
The other major trend is the efficient scaling-down of production. Some new technologies
such as cloud technology, big data, and data analytics will enable the design, manufactur
ing, and delivery of personalized products (goods and services) thanks to a two-way flow
of information and data between the producers and its customers. The flourishing of a
growing number of start-ups or entrepreneurial activities linked to digital technologies
but with the ethos of producing something has been widely documented in the ‘makers
movement’. These ‘digital artisans’ utilize new technologies (p. 489) to create new—
emerging or existing—niche markets or to address the demand for small-scale, specific
products. The key point here is that digitally enabled technologies can be efficiently
scaled down to support small-scale productions whose markets are very thin and geo
graphically dispersed. Makers tend to appear in places where there is already an ‘indus
trial commons’ (Anderson, 2012) with localized specialized competences because their at
tachment to the tacit and cumulated knowledge of the place provides a unique advantage
that they couple with an ability to master digital technologies. Another example of an effi
cient scaling-down of production is the use of additive manufacturing or 3D-printing that
has enabled production at the point-of-consumption with the related benefits of reducing
time lags, lower disruption, storage, stockpiling, and so on (De Propris and Bailey, 2020).
Digital technology and automation are therefore expected to transform buyer–supplier re
lationships along the whole supply chain. Indeed, the Internet of things, sensoring, space
technology, and mobile technology are, for instance, expected to enable remote machine-
to-machine communications that can allow for the coordination of complex production op
erations via a smooth integration of functions geographically dispersed.
The attribution of value creation along the value chain is likely to be completely re-drawn
by FIR technologies. We are used to visualizing value attribution along the value chain
with the smile curve (Mudambi, 2008) and this framework has (arguably) explained three
decades of production decoupling (Baldwin, 2016) as MNCs adopted strategies to shave
costs from low-value-adding functions by relocating them in low-cost economies. Howev
er, we would argue that the five transformations above-mentioned are likely underpinned
by a re-distribution of value creation across production functions in ways that are not yet
fully understood. The adoption and application of new technologies will disrupt the para
meters and the processes of value creation for firms, leading to a radical rethink of value
creation along the supply chain. This means that production is no longer a low-value-
added function but it can embody equal, if not more, value addition than innovation (de
sign and R&D) since the two go hand in hand. We can think therefore of a flat or even of a
‘sad curve’ (see Figure 18.2).
Page 5 of 22
FIGURE 18.2 The sad curve
Source: Authors’ elaboration.
In summary, technological change is expected to bring about disruptive transfor

(p. 490)
mations that will encourage a re-coupling of innovation with production, while also easing
a scaling-down of efficient production. Overall this will likely reshape value creation
along the value chain by redistributing value addition differently from before. The big
question is what impact this will have on the geographical dispersion of the value chain.
We are proposing that there is some initial evidence suggesting an increasing interest in
co-locating innovation and production leading to some de-globalizing trends.
18.3 GVCs and Manufacturing Hollowing Out

A wealth of literature going back to the 1970s has dissected the drivers of MNCs’ produc
tion internationalization strategies (the ‘why’) and related operations (the ‘how’) (Hymer,
1960; Dunning, 2000; Mathews, 2006). The globalization of production activities that
emerged from the combined offshoring of manufacturing activities created so-called
GVCs (Gereffi and Fernandez-Stark, 2011). Consequently, the Smithian division of labour
became a division of ‘value-added functions’, whereby each production stage contributes,
to a higher or lower extent, to the value addition of the final output. The location choices
of multinational enterprises (MNEs) were motivated by the contribution that different
‘places’ could make to their overall division of value, but in practice such choices were
place-neutral in the sense that MNEs were indifferent across places given equal features.
Production choices were fundamentally ‘de-territorialized’ (Scholte, 2000) and places
were replaceable, especially for low-value functions.
In particular, what was relocated were low-value-adding functions, namely labour-inten

sive manufacturing functions, to low-labour-cost countries via the creation of GVCs. This
framework sees the production process as a value chain, that is ‘the full range of activi
ties which are required to bring a product or a service from conception, through the dif
ferent phases of production…delivery to final consumers and the final disposal after use…
Production per se is only one of a number of value added links’ (Kaplinsky and Morris,
2002). So each function is weighted in the firm’s strategy depending on its value creation
contribution: what contributes the most to the value added of the final production is
clearly what is likely to be seen by the firm as a ‘core activity’ or a ‘strategic activity’ and
Page 6 of 22
is therefore internalized, while what contributes the least, on the contrary, is seen as sec
ondary and can be outsourced. GVCs comprise, therefore, components that embody dif
ferent value-adding content and which contribute differently to the overall value added
(Timmer et al., 2014). In this framework, manufacturing, production, and assembly are
deemed to produce low levels of value added, because when the technology is mature, in
imperfectly competitive markets (i.e. (p. 491) global markets), a reduced price elasticity
can only be constructed by embodying in products a high content of intangible value
through product development, advertising, and marketing. These are the functions that
most contribute to firms’ extraction of market power and are therefore higher in value
added. This trend has been extensively documented in the literature on GVCs (OECD,
2017b).
The GVC framework was introduced by Gereffi and Fernandez-Stark (2011) to describe
and analyse the impact that manufacturing offshoring had on developing and emerging
economies, especially in terms of working conditions and labour standards (Barrientos et
al., 2011), and trajectories of upgrading (Humphrey and Schmitz, 2002). One of the key
tenets of the GVC framework is that inter-firm relationships are determined by their rela
tive decision-making powers, namely their governance (Gereffi et al., 2005). Suppliers
producing low-value-added and standardized components are constantly competing on
price; they are likely to be in a captive relationship with their buyer who can easily re
place them with a cheaper supplier. Lee and Gereffi (2015) argue that to overcome this
vulnerable position, suppliers attempt to upgrade their capabilities to extend the portion
of the value chain they can cover. This allows them to extend their bargaining power by
moving towards building more relational transactions based on the uniqueness and value
of the suppliers’ competences.
Governance and upgrading are therefore the ‘twin pillars of the GVC framework’ (Gereffi,
2019: 240) to the extent that they determine the dynamics of economic development in
developing economies. Under the assumption of a value chain shaped like the ‘smile
curve’ above-described, any aspiration to greater value capture does not rest only on indi
vidual firms, suppliers, but it requires ‘economic actors—nations, firms, and workers—to
move from low-value to relatively high-value activities in global production
networks’ (Gereffi, 2019: 240). In other words, economic upgrading takes on a systemic
connotation that supports and strengthens any upgrading attempts of individual suppli
ers. For such systemic upgrading to occur, national or regional governments are required
to proactively intervene in the macro- and micro-components of the economy. Gereffi and
Sturgeon (2013) suggest moving away from import-substitution policies or export-orient
ed policies, to embrace value-chain-oriented policies that strategically target specific in
dustries where growth is favoured. Economic upgrading in such specific industries—us
ing the GVC approach—would mean supporting suppliers to develop a global customer
base, develop a competitive advantage in high-value niches, in specialized value chains,
and finally moving towards the top end of the value chain (Gereffi and Sturgeon, 2013).
Page 7 of 22
There are two points worth making as we are building up to discuss how IP can leverage
the reshaping of global production due to technological change. One is that GVCs de
scribe models of organizing production based on decoupling and offshoring at a time
when the technology was mature, innovations were incremental, and competition was on
price or on market imperfections and segmentations (e.g. branding). Therefore, firms
could only innovate incrementally and adopt cost-saving strategies. The second point is
that in the global economy there was a convenient divide between countries holding tech
nological capabilities and countries holding lower-costs (p. 492) resources; such ‘comple
mentarity’ in our view drove what some saw as a ‘turbocharged globalization’ (Friedman,
1999). Indeed, the need and policy will of emerging/transition/developing economies to
enter the global economy met with the strategic determination of MNEs to seek cost effi
ciencies.
These dynamics played out until 2008, when the global financial crisis occurred. The relo
cation of manufacturing to Asia introduced a novel international division of labour that
turned out to have significant implications for the erosion of the manufacturing base of
advanced economies with a dangerous concentration on untradable sectors. The manu
facturing hollowing-out presented a number of costs for EU economies. Various streams
of research contested the assumption that manufacturing processes and innovation might
easily be decoupled and relocated independently of each other (Pisano and Shih, 2009;
Ketokivi and Ali‐Yrkkö, 2009). Indeed, the offshoring of labour-intensive manufacturing
functions started pulling along also more innovation-intensive ones, thereby destabilizing
the EU innovation basis. This is already occurring in pharmaceutical, advanced engineer
ing, and ICT.
The GVC framework has very recently been applied to describe the internationalization of
Italian industrial districts as a systemic strategy that had implications for the home
economies (De Marchi et al., 2017) in terms of changing the reconfiguring and breaking
up local value chains as well as losing crucial competences to offshored functions. It be
came clear that the demise of manufacturing activities resulted in an impoverished ‘in
dustrial commons’ eroding advanced economies’ innovation capabilities (Pisano and Shih,
2009), while allowing suppliers in emerging economies to gradually climb up the value
ladder, through so-called processes of learning by supplying (Alcacer and Oxley, 2014).
The loss of skills, competences, and tacit knowledge across a sufficiently diversified suite
of sectors has had a long-lasting effect on the ability of EU regions to maintain an indus
trial base able to secure long-term prosperity for its citizens. Indeed, critiques of novel
forms of organization pointed to the larger societal drawbacks of the offshoring bandwag
on, highlighting how the ‘hollowing-out’ of the large corporation and the relocation of
manufacturing abroad contributed to the increasing levels of inequality recorded in ad
vanced societies.
Sectoral imbalances in some EU economies and the United States towards services and
untradable sectors also reduced the ability of these economic systems to respond to the
financial shocks, which led to a much more precarious and lengthy economic recovery
(Aiginger, 2015), in some economies like the United States overcome only by significant
Page 8 of 22
and sustained fiscal expansionary policies. Those countries that had chosen to maintain a
solid manufacturing base, such as Germany, Belgium, or France, were found to reset their
economy more quickly and therefore to be more resilient. Other economies with an equal
ly healthy manufacturing base lingered, however, in a long and painful recession due to
stringent austerity policies, such as Italy (Aiginger, 2015). Post-financial crisis, a lively de
bate emerged on the importance of manufacturing for advanced economies to maintain a
balanced economic and skills base: what was often referred to as a need for ‘rebalancing’
or a new manufacturing renaissance (Bailey and Tomlinson, 2017).
This renewed interest in manufacturing had three aspects. First, it engaged policymakers
who understood a call to defend and promote domestic jobs and factories by (p. 493) sup
porting domestic firms and sectors. Second, protectionist trade policies and more recent
ly aggressive trade wars have been defended to protect, for example, ‘American
jobs’ (Forbes, 2019); however, this fails to understand the interconnectedness of global
production and the consequences of such actions. Finally, it coincided with the surfacing
of the debate on technological change and a ‘new’, ‘smart’ manufacturing model; in other
words, there was an opening for high-cost economies to develop and anchor new techno
logical capabilities across EU manufacturing sectors to support jobs creation and a high
er standard of living. This also meant a new competitive scenario for firms where low-val
ue-adding labour-intensive functions had to be redefined.
The adoption of new technologies by the manufacturing sector will transform the manu
facturing model in ways discussed above, which we argue will also entail a new global or
ganization of production.
18.4 Signs of De-globalization

The renewed interest in manufacturing of advanced economies translated into policymak
ers’ commitment to re-build their manufacturing base. This meant attempting to slow
down offshoring, but also encouraging a new trend that saw firms which had previously
offshored some production functions looking to bring those activities back home. Indeed,
evidence emerged that some firms with operations abroad were considering or actively
pursuing reshoring strategies. Reshoring has been used to capture the choice of a multi
national firm to move back to their home economy a production operation previously off
shored—either via a new investment or domestic outsourcing (Bailey and De Propris
2014). Reshoring has since become something of a buzzword with many country case
studies emerging in relation to the United States, Spain, Italy, and Germany (Kinkel,
2020; Kinkel et al., 2020; Ernst & Young, 2015) over the last few years. Company case
studies have also emerged, documenting individual firms’ strategic rationale for
reshoring part of their operations back to the home economy, such as Walmart, Apple,
Google, Caterpillar, and General Electrics (Reshoring Institute, 2019; Bals et al., 2016) or
the case of the Finnish bicycle company which moved production from Taiwan back to
Finland (Gylling et al., 2015).
Page 9 of 22
Many studies have discussed some of the push factors that have convinced firms to recon
sider the actual cost and benefit of offshoring (Kinkel et al., 2020). Value chain frictions
and disruptions due to natural disasters or geo political factors have shown that global
operations have so far ignored important hidden costs (Gray et al., 2017; Kinkel and Mal
oca, 2009) linked to the increasing complexity of value-chain governance (Lavissiére et
al., 2016), a lack of flexibility and responsiveness due to long lags, and overlooked quality
and control limitations (Bailey and De Propris, 2014). On the other hand, the decision to
move back to relatively high-cost locations in advanced economies has also been driven
by some pull factors that have made them attractive, especially in relation to the emer
gent technological change: such as greater flexibility (p. 494) and responsiveness; closer
integration with design and product development; better control on quality and variety,
shorter lead time, and the use of new technologies to make cost savings. These location
decision strategies have been linked to changes in the distribution of value creation along
the value chain, as discussed above, with a shift from a smiling curve to a flatter one (see
Figure 18.2).
Some scepticism has been expressed on the small scale of the reshoring phenomenon,
notwithstanding the visible and active commitment of policymakers to leverage it to re
populate their industrial base (see, for instance, the US tax change to encourage MNEs to
repatriate their profits, New York Times, 2019). This has been in part due to difficulties in
capturing empirically the changes in firms’ production decisions: data on firms’ decision
to swap a foreign supplier for a domestic supplier is hard to find. Easier to trace have
been changes in firms’ investment decisions; however, these have tended to be more rigid
in the short term and in some countries very much constrained. However, reshoring
trends have been an important signal of a change in the perception of business over
whether choosing to have globalized operations is always superior to having national or
locally bound operations. Indeed, we see reshoring representing firms’ longer-term strat
egy to better face a new competitive environment (Młody, 2016; Navarro, 2013), and must
therefore be read as an early sign of a transformation that is being unleashed at the mo
ment.
Firms’ supply chains are shortening and the cost-saving attraction of production off
shoring is receding. This opens two levels of questions: one is to what extent we can talk
about de-globalizing forces shaping a new production order, and the other is whether re
gional economies in advanced economies are capable of picking up this challenge and be
come fertile ground for a flourishing of new dense and geographically short value chains.
An emerging literature on de-globalization is starting to look into what this might mean
(Livesey, 2017, 2018; van Bergeijk, 2018; Martin et al., 2018). Some have drawn on data
on long-term changes in FDI and import/export trends and have argued that there is evi
dence of a reduction in the global production activities (Pegoraro et al., 2020) with supply
chains being geographically shorter.
Page 10 of 22
18.5 Industrial Policy and Beyond

The term ‘industrial policy’ has gone through many fashions and has meant different
things at different times. Government intervention in the productive parts of the economy
was justified by instances of market failure or any time the market was unable to allocate
resources efficiently or equitably. This view justified national champions in industries
deemed to be natural monopolies, for instance in the utilities. All this was of course
turned on its head by a massive wave of privatizations in the 1980s in Europe, where
competition and liberalization allowed market forces in industries of strategic societal im
portance such as water, energy, and telecoms. At the same time, (p. 495) industrial policy
has been associated with interventionist governments promoting national champions or
picking winners, that is, sectors, in a global competitive environment characterized by in
ter-country competition and home-rooted large firms. This was the case when industrial
policies were aimed at supporting the growth of domestic champions and turning a blind
eye to their sometimes-anticompetitive dominant position domestically, in view of allow
ing them to become competitive in foreign markets (Bianchi and Labory, 2018; Bianchi et
al., 2019).
We would argue that the industrial policy of picking sectors of the 1970s and 1980s oc
curred at a time of technological change coinciding roughly with the Third Industrial Rev
olution, so government intervention to promote industry development was aimed at sup
porting domestic champions positioning themselves in new industries vis-à-vis foreign
competitors in a competitive environment shaped by absolute advantage and trade.
The emergence in the 1980s of MNEs controlling more complex GVCs changed the bal
ance of power between large firms and their home government as MNEs started dealing
with foreign governments keen to host their operations and offering favourable terms to
attract them. In some countries, such as Ireland, the attraction of FDI was part of a na
tional industrial policy to promote the emergence and development of some industries
(Bailey et al., 2016). In such a globalized context and a time when technology was mature
and known, the term industrial policy fell from grace, replaced by a heightened attention
to the contribution of places and regions with their industrial specializations to create na
tional competitive advantages (Porter, 1990). Indeed government intervention was recast
to promote, not national champions, but a favourable national business environment, a
national innovation system (Chaminade et al., 2018), and to create systemic conditions for
businesses to thrive from telecoms infrastructure and education/skills, to transport con
nectivity and regulatory framework. These fell under different labels though, such as re
gional policy, innovation policy, or competitiveness policy. The latter had in common in
our view two elements; one is that their prescriptions assumed maximum impact with
minimum government intervention, as their gestation and delivery occurred at times
where the dominant thinking was free market and deregulation. The second common ele
ment is that they introduced, for the first time, an appreciation of the importance of
places as local engines of growth contributing to a country’s competitiveness.
Page 11 of 22
Yet, industrial policy is now back on the agenda after years of neglect (Bailey et al.,
2019), with this revived enthusiasm coming at a time of technological transformation.
Over the last decade, ‘place-based’ approaches have increasingly been used to develop
regional industrial policies, as dissatisfaction with spatially blind approaches has grown.
Place-based approaches emphasize tailoring policy to develop place-specific specialisms
and capabilities, on which a region can thrive. Barca (2009), for instance, defines a place-
based approach as a long-term strategy aimed at tackling persistent underutilization of
potential and persistent social exclusion in specific places through external interventions
and multilevel governance. This place-based focus has begun to be applied to a wide
range of policy settings including industrial policy (Bailey et al., 2018a, (p. 496) 2018b),
research on place-based leadership (Beer et al., 2019), and in managing the impacts of
economic shocks (RSA, 2019). The place-based approach chimes with modern perspec
tives on industrial policy more broadly, where industrial policy is seen as a process of dis
covery requiring strategic collaboration between the private sector and state in unlocking
growth opportunities, where policy ideally has the quality of ‘embedded autonomy’ (see
Rodrik, 2008). It is not captured by firms and sectors, but focuses on bringing together
actors in a discovery process, where firms and the state learn about underlying costs and
opportunities and engage in strategic coordination.
To be successful, place-based approaches require strengthened local and regional institu

tions, a need for local stakeholders to be active in order to deliver success, the develop
ment of human capital, and the critical embrace of innovation (Tomaney, 2010). However,
as the Regional Studies Association (2019) notes, government initiatives badged as
‘place-based policy’ often fall well short, with governments simply relabelling long-estab
lished programmes as ‘place-based policy’, or only partially innovating.
In the context of the EU’s regional policy, smart specialization strategies at the regional
level were designed to empower regional actors to identify trajectories of industrial re
newal drawing on their embedded specializations. Although smart specialization strate
gies are seen as having paved the way in place-based approaches, Gianelle et al. (2019)
suggests that regions and countries have, in reality, put in place mechanisms that may
circumvent the very rationale of smart specialization. They note that this could arise from
the result of lobbying activities, higher political return from widespread public support
measures, a risk-averse attitude of policymakers, and a lack of adequate institutional and
administrative capacity that can be observed at national and regional levels. Further
more, Barzotto et al. (2019a and 2019b) argue that specialization strategy projects are
more likely to be identified in stronger regions, which boast a strong entrepreneurial tal
ent pool and business networks; with the result that policy implementation may inadver
tently be a-spatial and could exacerbate regional imbalances.
The arrival of radical and disruptive FIR technologies associated with Industry 4.0, poses
challenges for policymaking and the role of governments. Technological change will fun
damentally transform industries and markets, as this will inevitably impact on regions
and places. The big question is whether the focus on regional policy and concerns about
wider socio-economic cohesion and growth should still dominate, or whether industrial
Page 12 of 22
policy can regain centre stage as a strategic intervention to shape a country’s productive
competitiveness. Recent work has stressed that the disruptions brought about by such
technologies have the potential to introduce new layers of socio-economic divides (De
Propris and Bailey, 2020), with a concomitant call for a more holistic transformative in
dustrial policy that brings together technology, sectors, and place. Against this backdrop,
the policymakers risk underestimating the fundamental role that policy and public inter
ventions need to play just as the economy and wider society are embarking on transfor
mations that will shape work, industry, mobility, communications, and more for decades
ahead.
(p. 497) 18.6 Transformative Industrial Policy

As suggested in the definition of Industry 4.0+, new technologies will not only fundamen
tally change the organization of production, but crucially the nature of industries and
markets and with them society. The global organization of production that we have wit
nessed so far is responding to these changes with visible reconfigurations of value chains
in more compact networks that stretch within continents as production spaces and final
markets (Schwab, 2018).
Such radical and disruptive change cannot occur without the pro active role of govern
ments and policymaking. Industrial policies must move towards having a truly transfor
mative power, by enabling new technologies to penetrate and redefine existing sectors, as
well as create brand new markets and industries. Indeed, awareness and adoption are
crucial steps to ensure that the existing manufacturing base shifts onto the new techno
logical paradigm despite the inevitable resistance from incumbent actors. This can, how
ever, only be successful if the technological transformation of industries is rendered
against the place-based unique characteristics of such localized industries in regions. We
therefore advocate for a transformative place-based industrial policy that acts at the in
tersection across three dimensions: the place (cluster and region), the industry with its
value chain, and the new technologies.
This requires three levels of intervention. First, industrial policy must connect the indus
tries already embedded in regions to new technologies through a process of accessing,
adopting, and adapting to the latter. This could, in the European context for example, em
body smart specialization strategies with a heightened attention towards the creation and
adoption of FIR technologies in different regions so as to bring together technology, sec
tors, and place. A transformative industrial policy needs to think beyond sectors alone
and, rather, identify, nurture, and diffuse the key cross-cutting technologies (such as digi
talization, the Internet of things, robotics, and artificial intelligence) that have an en
abling role across manufacturing and services.
Industrial and regional policy needs to recognize and exploit such technologies by making
them accessible to businesses in different regions. Furthermore, a transformative region
al industrial policy needs to be developed in a holistic sense (for example, on skills, ac
Page 13 of 22
cess to finance, clusters, supply chains, and innovation) so as to enable policy to be better
suited to the distinctive characteristics and advantages of different scales.
The latter requires regionally based industrial development strategies promoting ‘related
diversification’ capitalizing on the FIR. Such strategies need to recognize (i) the need to
bring together different but related activities in a region via cross cutting FIR technology
platforms (such as via Living Labs or digital demonstration hubs) and (ii) the differing po
tentials of regions to diversify, due to different industrial, knowledge, and institutional
structures linked to specific regional historical trajectories. This requires tailor-made pol
icy actions embedded in and linked to the specific needs and available resources of re
gions, starting with the existing knowledge and institutional (p. 498) base in that region.
These need to capitalize on region-specific assets, rather than attempting to replicate and
apply policies that may have worked in quite different places (Bailey et al., 2018b).
Other elements of a transformative regional industrial policy would need to involve, inter
alia, the need for: new skills to be developed and constant re-skilling and up-skilling
processes as the FIR progresses. However, to transform the region’s potential based on
‘unrelated variety’, and to broaden and renew the region’s industrial structure by helping
it branch into new related activities, policy must also encourage crossovers between man
ufacturing and service industries and between manufacturing and new technologies. This
could come via knowledge-transfer mechanisms that connect related and unrelated indus
tries.
Second, a place-based industrial policy must favour the creation of new firms and start-
ups in the embedded industries. This new generation of entrepreneurs is crucial to ex
plore, create, and experiment with market niches often at the intersection between sec
tors. This will require SMEs to have access to funding and finance to embrace digital
technologies; recognizing and exploiting possibilities to reposition firms, industries, and
regions on new parts of the GVC as the value added of manufacturing changes over time;
creating and seizing re-shoring opportunities as re-localization opportunities open up,
namely involving policies to rebuild supply chains in Europe. The EU Industrial Policy
2030 (European Commission, 2019) suggests that new technologies offer a great opportu
nity to strengthen European firms to export in ‘fast growing world markets’ (European
Commission, 2019: 17) and to leverage firms’ competitive advantage in new technologies
by gaining a ‘strategic independence from foreign suppliers’ (European Commission,
2019: 11). This confirms, in the case of the EU (likewise for Asian and US competitors;
Schwab, 2018), the commitment to control the emerging new technologies and to be able
to deploy them horizontally across sectors in tightly knit networks of buyers and suppli
ers.
At the same time, new technologies open up endless possibilities for exploration and inno
vation and this should stimulate the entrepreneurial, risk-taking nature of new businesses
and start-ups. Now more than ever, initiatives—like incubators, science parks, accelera
tors—that enable the realization of such ventures should be supported and expanded.
Drawing on the triple helix model, the embeddedness of these facilities in regional spaces
Page 14 of 22
with a dynamic (albeit mature) manufacturing specialization, can offer true path renewal
strategies for local firm clusters, as they connect systemically businesses with academia
or research organizations. These new innovative businesses will provide an injection of
technological novelty in the local industry with transformative powers: a holistic regional
industrial policy (Harrison et al., 2019). An example of this is the Vinnväxt programme of
Vinnova (the Innovation System Agency of the Swedish Government) which supports re
gions that have declining industries to identify new industrial trajectories thanks to a sys
temic effort that brings together universities, local institutions, and the local business
community. The Vinnväxt programme provides prime funding to help regional stakehold
ers work together to graft one or more (p. 499) relevant new technologies onto the local
industry with the promotion of new firms and new local value chains.
Finally, the redefinition and use of regulation and procurement to create new markets for
technological change also means technological uncertainty as the old paradigm is setting,
but many new ones are still leaving space for doubt and hesitation. In this transition, a
pro active and far-sighted government is necessary to share the risk of technology adop
tion with business. This means first of all, significant infrastructural investment to em
brace new technologies (e.g. 5G) and enable firms and citizens to become aware of users
of such technologies. More significantly, the transformative power of an industrial policy
that is prioritizing technological transformation requires the use of ‘smart regulation’ to
change behaviours and incentives and a risk-sharing procurement policy that creates new
markets and allows the exploration of new products or services. This idea was introduced
first by Porter and van der Linde (1995) who controversially suggested that to overcome
firms’ resistance to adopt new technologies (in their article linked to sustainability) ‘prop
erly designed—government—regulations’ can support firms’ innovation via technological
adoption and applications at times of technological change. In protected domestic mar
kets, firms can achieve levels of innovation and performance that allow them to compete
internationally and engage in GVCs at the highest levels of value creation. Advocates of a
pro active role of government have become more vocal recently thanks to the work of
Mazzucato (2011) who argued for an entrepreneurial role for government by inspiring,
risk-sharing, and supporting businesses’ mission to innovate. By designing mission-orient
ed policies associated to innovation and technological change, government can use pro
curement to favour the creation of small-scale value-chain networks of firms with pur
poseful links with research organizations. The design and deployment of new products or
processes for protected public markets allow firms to gain a competitive advantage that
will redraw completely their position and value creation contributions in GVCs.
18.7 Conclusions
This chapter has examined the impact of technological change on GVCs and what initia
tives and instruments governments in advanced economies might deploy to support firms
and people during the transition. There are signs that the global economy is ‘de-globaliz
ing’, with firms seeking to co-locate manufacturing and innovation activities; this offers
an opportunity for regions and places to upgrade and transform their economies, leverag
Page 15 of 22
ing industrial legacy with frontier technologies. We have discussed the implications for a
transformative industrial policy that we see comprising a number of dimensions. A trans
formative industrial policy needs to think beyond sectors alone and, rather, identify, nur
ture, and diffuse the key cross-cutting (p. 500) technologies (such as digitalization, the In
ternet of things, robotics, and artificial intelligence) that have an enabling role across
manufacturing and services. Policy also needs to recognize and exploit such technologies
by making them accessible to businesses in different regions. Furthermore, a transforma
tive regional industrial policy needs to be developed in a holistic sense (e.g. on skills, ac
cess to finance, clusters, supply chains, and innovation) so as to enable policy to be better
suited to the distinctive characteristics and advantages of different scales.
Acknowledgements
The writing of this chapter has been supported by the EU Horizon 2020 project ‘MAK
ERS’, which is a Research and Innovation Staff Exchange under the Marie Sklodowska-
Curie Actions, grant agreement number 691192, and via the Economic and Social Re
search Council’s ‘UK in a Changing Europe’ programme, Grant Reference: ES/T000848/1.
References
Aiginger, Karl (2015) ‘Industrial Policy for a Sustainable Growth Path’, in David Bailey,
Keith Cowling, and Philip R. Tomlinson (eds) New Perspectives on Industrial Policy for a
Modern Britain. Oxford: Oxford University Press, pp. 365–94.
Alcacer, Juan and Joanna Oxley (2014) ‘Learning by Supplying’, Strategic Management
Journal 35(2): 204–23.
Anderson, Chris (2012) Makers: The New Industrial Revolution. New York: Random
House.
Bailey, David and Lisa De Propris (2014) ‘Manufacturing Reshoring and its Limits: The
UK Automotive Case’, Cambridge Journal of Regions, Economy and Society 7(3): 379–98.
Bailey, David and Lisa De Propris (2019) ‘6. Industry 4.0, Regional Disparities and Trans
formative Industrial Policy’, Regional Studies Policy Impact Books 1(2): 67?78.
Bailey, David and Philip R. Tomlinson (2017) ‘Industrial Policy since the Crisis’, in Philip
Arestis and Malcolm Sawyer (eds) Economic Policies since the Global Financial Crisis. In
ternational Papers in Political Economy Series. London: Palgrave Macmillan, pp. 221–63.
Bailey, David, Carlo Corradini, and Lisa De Propris (2018a) ‘“Home-Sourcing” and Closer
Value Chains in Mature Economies: The Case of Spanish Manufacturing’, Cambridge Jour
nal of Economics 42(6): 1567–84.
Page 16 of 22
Bailey, David, Ann Glasmeier, Philip R. Tomlinson, and Peter Tyler (2019) ‘Industrial Poli
cy: New Technologies and Transformative Innovation Policies?’, Cambridge Journal of Re
gions, Economy and Society 12(2): 169–77.
Bailey, David, Helena Lenihan, and Alex de Ruyter (2016) ‘A Cautionary Tale of Two
Tigers: Industrial Policy “Lessons” from Ireland and Hungary’, Local Economy 31(8): 873–
91.
Bailey, David, Christos Pitelis, and Philip R. Tomlinson (2018b) ‘A Place-based Develop
mental Regional Industrial Strategy and/for Sustainable Capture of Co-created Value’,
Cambridge Journal of Economics 42(6): 1521–42.
Baines, Tim, Ali Ziaee Bigdeli, Oscar F. Bustinza, Victor Ghuang Shi, James Baldwin, and
Keith Ridgway (2017) ‘Servitization: Revisiting the State-of-the-art and Research Priori
ties’, International Journal of Operations and Production Management 37(2): 256–78.
(p. 501)
Baldwin, Richard (2016) The Great Convergence. Cambridge, MA: Harvard University
Press.
Bals, Lydia, Jon F. Kirchoff, and Kai Foerstl (2016) ‘Exploring the Reshoring and Insourc
ing Decision-making Process: Toward an Agenda for Future Research’, Operations Man
agement Research 9(3–4): 102–16.
Barca, Fabrizio (2009) ‘An Agenda for a Reformed Cohesion Policy: A Place-based Ap
proach to Meeting European Union Challenges and Expectations’. Available at http://
www.interact-eu.net/downloads/1224/
An_Agenda_for_a_Reformed_Cohesion_Policy.pdf.
Barrientos, Stefanie, Gary Gereffi, and Arianna Rossi (2011) ‘Economic and Social Up
grading in Global Production Networks: A New Paradigm for a Changing World’, Interna
tional Labour Review 150(3–4): 319–40.
Barzotto, Mariachiara, Carlo Corradini, Felicia Fai, Sandrine Labory, and Philip R. Tomlin
son (2019a) Revitalising Lagging Regions: Smart Specialisation and Industry 4.0. London:
Routledge.
Barzotto, Mariachiara, Carlo Corradini, Felicia Fai, Sandrine Labory, and Philip R. Tomlin
son (2019b) ‘Enhancing Innovative Capabilities in Lagging Regions: An Extra-Regional
Collaborative Approach to RIS3’, Cambridge Journal of Regions, Economy and Society,
12(2): 213–32.
Beer, Andrew, Sarah Ayres, Terry Clower, Fabian Faller, Alessandro Sancino, and Markku
Sotarauta (2019) ‘Place Leadership and Regional Economic Development: A Framework
for Cross-regional Analysis’, Regional Studies 53(2): 171?82. doi:
10.1080/00343404.2018.1447662
Page 17 of 22
Bianchi, Patrizio and Sandrine Labory (2018) A Concrete Experiment of Industrial Policy
for the Manufacturing Revolution. Cheltenham: Edward Elgar.
Bianchi, Patrizio, Clemente Ruiz Durán, and Sandrine Labory (eds) (2019) Transforming
Industrial Policy for the Digital Age: Production, Territories and Structural Change. Chel
tenham: Edward Elgar.
Bustinza, Oscar F., Ferran Vendrell-Herrero, Emanuel Gomes, and Esteban Lafuente
(2018) ‘Product–Service Innovation and Performance: Unveiling the Complexities’, Inter
national Journal of Business Environment 10(2): 95–111.
Chaminade, Cristina, Bent Åke Lundvall, and Shagufta Haneef (2018) Advanced Introduc
tion to National Innovation Systems. Cheltenham: Edward Elgar.
Deloitte (2015) Industry 4.0. Challenges and Solutions for the Digital Transformation and
Use of Exponential Technologies. Zurich: Deloitte.
De Marchi Valenina, Elisa De Maria, and Gary Gereffi (2017) Local Clusters in Global Val
ue Chains. London: Routledge.
De Propris, Lisa (2016) ‘How the Fourth Industrial Revolution is Powering the Rise of
Smart Manufacturing’, World Economic Forum. Available at https://www.weforum.org/
agenda/2016/06/how-the-fourth-industrial-revolution-is-powering-the-rise-of-
smart-manufacturing.
De Propris, Lisa and David Bailey (2020) ‘Disruptive Industry 4.0+: Key Concepts’, in Lisa
De Propris and David Bailey (eds) Industry 4.0 and Regional Transformations. London:
Taylor and Francis.
Deutsche Bank (2014) ‘Industry 4.0: Upgrading of Germany’s Industrial Capabilities on

the Horizon’. Available at https://www.dbresearch.com/PROD/RPS_EN-PROD/
PROD0000000000451959/
Industry_4_0%3A_Upgrading_of_Germany%E2%80%99s_industrial_ca.pdf.
Dimache, Aurora and Thomas Roche (2013) ‘A Decision Methodology to Support Servitisa
tion of Manufacturing’, International Journal of Operations and Production Management
33(11–12): 1435–57. (p. 502)
Dunning, John H. (2000) ‘The Eclectic Paradigm as an Envelope for Economic and Busi
ness Theories of MNE Activity’, International Business Review 9(2): 163–90.
Ernst & Young (2015) Reshoring Manufacturing—Time to Seize the Opportunity. London:
Ernst & Young.
European Commission (2016) Industry 4.0. Brussels: European Commission.
European Commission (2019) Industry 2030 Report. Luxembourg: Publication Office of

the European Union.
Page 18 of 22
Forbes (2019) ‘The Trade War That Never Stops’. Available at https://www.forbes.com/
sites/stevehanke/2019/10/13/the-trade-war-that-never-stops/#5528d92c74dd.
Friedman, Thomas L. (1999) The Lexus and the Olive Tree. New York: Farrar, Straus &
Giroux.
Gereffi, Gary (1999) ‘International Trade and Industrial Upgrading in the Apparel Com
modity Chain’, Journal of International Economics 48(1): 37–70.
Gereffi, Gary (2019) ‘Economic Upgrading in Global Value Chains’, in Stefano Ponte, Gary
Gereffi, and Gale Raj-Reichert (eds) Handbook on Global Value Chains. Cheltenham: Ed
ward Elgar, pp. 240–54.
Gereffi, Gary and Karina Fernandez-Stark (2011) ‘Global Value Chain Analysis: A Primer’,
Center on Globalization, Governance & Competitiveness (CGGC), Duke University, North
Carolina, United States.
Gereffi, Gary and Timothy Sturgeon (2013) ‘Global-value-chain-oriented Industrial Policy:

The Role of Emerging Economies’, in Deborah K. Elms and Patrick Lows (eds) Global Val
ue Chains in a Changing World. Geneva: WTO, pp. 329–60.
Value Chains’, Review of International Political Economy 12(1): 78–104.
Gianelle, Carlo, Fabrizio Guzzo, and Krzysztof Mieszkowski (2019) ‘Smart Specialisation:
What Gets Lost in Translation from Concept to Practice? Regional Studies May: 1–12.
10.1080/00343404.2019.1607970.
Gray, John V., Gökçe Esenduran, Johnny M. Rungtusanatham, and Keith Skowronski
(2017) ‘Why in the World Did They Reshore? Examining Small to Medium-sized Manufac
turer Decisions’, Journal of Operations Management 49: 37–51.
GTAI (2014) ‘Industrie 4.0. Smart Manufacturing for the Future’, German Trade and In
vestment. Available at https://www2.deloitte.com/content/dam/Deloitte/ch/Docu
ments/manufacturing/ch-en-manufacturing-industry-4-0-24102014.pdf.
Gylling, Michael, Jussi Heikkilä, Kari Jussila, and Markku Saarinen (2015) ‘Making Deci
sions on Offshore Outsourcing and Backshoring: A Case Study in the Bicycle Industry’, In
ternational Journal of Production Economics 162: 92.
Harrison John, Mercedes Delgado, Ben Derudder, Isabelle Anguelovski, Sergio Montero,
David Bailey, and Lisa De Propris (2019) ‘Pushing Regional Studies beyond its Borders’,
Regional Studies 54(1): 129–39.
Humphrey, John and Hubert Schmitz, (2002) ‘How Does Insertion in Global Value Chains
Affect Upgrading in Industrial Clusters?’, Regional Studies 36(9): 1017–27.
Page 19 of 22
Hymer, Stephen H. (1960) The International Operations of National Firms: A Study of Di
rect Foreign Investment. Cambridge, MA: MIT Press.
Kaplinsky, Rafael and Mike Morris (2002) A Handbook for Value Chain Research.
Brighton: Institute of Development Studies.
Ketokivi, Mikko and Jyrki Ali‐Yrkkö (2009) ‘Unbundling R&D and Manufacturing: Postin
dustrial Myth or Economic Reality?’, Review of Policy Research 26(1–2): 35–54. (p. 503)
Kinkel, Steffen (2020) ‘Industry 4.0 and Reshoring’, in Lisa De Propris and David Bailey
(eds) Industry 4.0 and Regional Transformations. London: Taylor and Francis.
Kinkel, Steffen and Spomenka Maloca (2009) ‘Drivers and Antecedents of Manufacturing
Offshoring and Backshoring: A German Perspective’, Journal of Purchasing & Supply
Management 15(3): 154–65.
Kinkel Steffen, Diletta Pegoraro, and Rosemary Coates (2020) ‘Reshoring in the US and
Europe’, in Lisa De Propris and David Bailey (eds) Industry 4.0 and Regional Transforma
tions. London: Taylor and Francis.
Kondratieff, Nikolai D. (1935) ‘The Long Waves in Economic Life’, The Review of Econom
ics and Statistics 17(6): 105–15.
Lavissière, Alexandre, Tibor Mandják, and Laurent Fedi (2016) ‘The Key Role of Infras
tructure in Backshoring Operations: The Case of Free Zones’, Supply Chain Forum: An In
ternational Journal 17(3): 143–55.
Lee, Joonkoo and Gary Gereffi (2015) ‘Global Value Chains, Rising Power Firms and Eco
nomic and Social Upgrading’, Critical Perspectives on International Business 11(3–4):
319–39.
Livesey, Finbarr (2017) From Global to Local. London: Profile Books.
Livesey, Finbarr (2018) ‘Unpacking the Possibilities of Deglobalisation’, Cambridge Jour

nal of Regions, Economy and Society 11(1): 177–87.
Martin Ron, Peter Tyler, Michael Storper, Emil Evenhuisd, and Ann Glasmeier (2018)
‘Globalization at a Critical Conjuncture?’, Cambridge Journal of Regions, Economy and
Society 11: 3–16.
Mathews, John A. (2006) ‘Dragon Multinationals: New Players in 21st-century Globaliza

tion’, Asia Pacific Journal of Management 23: 5–27.
Mazzucato, Mariana (2011) The Entrepreneurial State. London: Demos.
Młody, Michal (2016) ‘Backshoring in Light of the Concepts of Divestment and De-inter
nationalization: Similarities and Differences’, Entrepreneurial Business & Economics Re
view 4(3): 167–80.
Page 20 of 22
Mudambi, Ram (2008) ‘Location, Control and Innovation in Knowledge-intensive Indus

tries’, Journal of Economic Geography 8(5): 699–725.
Navarro, Peter (2013) ‘China 2013: The Year of Reshoring to America?’, Financial Execu
tive 29(1): 33–5.
New York Times (2019) ‘Trump’s Push to Bring Back Jobs to U.S. Shows Limited Results’,
13 August.
OECD (2017a) The Next Production Revolution: Implications for Governments and Busi
ness. Paris: OECD.
OECD (2017b) Global Value Chain Development Report 2017: Measuring and Analyzing
the Impact of GVCs on Economic Development. Paris: OECD.
Pegoraro, Diletta, Lisa De Propris, and Agnieszka Chidlow (2020) ‘De-globalisation, Value
Chains and Reshoring’, in Lisa De Propris and David Bailey (eds) Industry 4.0 and Region
al Transformations. London: Taylor and Francis.
Pérez, Carlota (2010) ‘Technological Revolutions and Techno-economic Paradigms’, Cam

Pisano, Gary P. and Willy C. Shih (2009) ‘Restoring American Competitiveness’, Harvard
Business Review 87(7–8): 114–25.
Porter, Michael E. (1990) ‘The Competitive Advantage of Nations’, Harvard Business Re
view 68(2): 73–93. (p. 504)
Porter Michael E. and Claas Van der Linde (1995) ‘Toward a New Conception of the Envi
ronment–Competitiveness Relationship’, Journal of Economic Perspective 9(4): 97–118.
PWC (2016) ‘Industry 4.0: Building the Digital Enterprise’. Available at https://
www.pwc.com/gx/en/industries/industries-4.0/landing-page/industry-4.0-building-
your-digital-enterprise-april-2016.pdf.
Reshoring Institute (2019) ‘Companies Case Studies’. Available at https://

reshoringinstitute.org/reshoring-knowledge-library/reshoring-case-studies/.
Rifkin, Jeremy (2011) The Third Industrial Revolution: How Lateral Power is Transform
ing Energy, the Economy, and the World. New York: Palgrave Macmillan.
Rodrik, Dani (2008) One Economics, Many Recipes: Globalization, Institutions, and Eco
nomic Growth. Princeton, NJ: Princeton University Press.
RSA (Regional Studies Association) (2019) Place-based Leadership. Available at http://

placeleadershipnetwork.org/rsa-policy-expo-enabling-best-practice-in-place-
based-policy.
Scholte, Jan A. (2000) Globalization: A Critical Introduction. New York: Macmillan.
Page 21 of 22
Schwab, Klaus (2016) The Fourth Industrial Revolution. Geneva: World Economic Forum.
Schwab, Klaus (2018) Shaping the Future of the Fourth Industrial Revolution. London:
Penguin Random House.
Timmer Marcel P., Abdul A. Erumban, Bart Los, Robert Stehrer, and Gaaitzen J. de Vries
(2014) ‘Slicing up Global Value Chains’, Journal of Economic Perspectives 28(2): 99–118.
Tomaney, John (2010) Place-based Approaches to Regional Development: Global Trends

and Australian Implications. Sydney: Australia Business Foundation.
Van Bergeijk, Peter A. G. (2018) ‘On the Brink of Deglobalisation…Again’, Cambridge

Journal of Regions, Economy and Society 11: 59–72.
Vendrell-Herrero, Ferran and Oscar F. Bustinza (2020) ‘Servitization in Europe’, in Lisa

De Propris and David Bailey (eds) Industry 4.0 and Regional Transformations. London:
Taylor and Francis.
David Bailey
David Bailey is professor of business economics at the University of Birmingham and

senior fellow of the ESRC’s ‘UK in the Changing Europe’ programme, exploring the
impacts of Brexit on the UK automotive and manufacturing industries. He has writ
ten extensively on industrial and regional policy, especially in relation to manufactur
ing and the auto industry.
Lisa De Propris
Lisa De Propris is professor of regional economic development, University of Birm

ingham. She has written extensively on competitiveness in clusters and regions; tech
nological change and innovation; clusters and globalization; and regional and indus
trial policy.
Page 22 of 22
Industrial Policy: A Long-term Perspective and Overview of Theoretical Ar
guments
Industrial Policy: A Long-term Perspective and

Overview of Theoretical Arguments
Erik S. Reinert
The aim of this chapter is to give a brief overview of the historical arguments that have
been used to argue for industrial policy in its widest sense, that is, that what a nation (or
region) specializes in producing may be of key importance to the wealth and welfare of its
inhabitants. Historically it has been generally agreed that symmetrical trade—trade in
similar goods between nations at similar levels of technological development—has tended
to be beneficial to both trading partners. In these cases, employing Ricardian trade theo
ry has not been detrimental to the trading partners. This chapter explains the situations
when Ricardian trade theory is not beneficial to one of the trading partners, and—at the
same time—the economic mechanisms which have been identified as making industrial
policy desirable. That manufacturing matters has, in various forms, been presented as a
main reason for industrial policy, at least since England’s import-substitution policies dur
ing the 1400s: adding value to English wool by spinning it into woollen cloth and gar
ments. This was mainly achieved by raising export duties on raw wool, making English
wool cheaper for domestic manufacturers than for foreign ones. However, the reasons
why manufacturing matters have varied. And that understanding has gone from intuitive
inferences to scientific evidence. This chapter will historically present this process and
the most common arguments for industrial policy over time.
Keywords: industrial policy, theoretical arguments, historical perspectives, increasing and diminishing returns,
symmetrical trade, trade theory, import-substitution policies, manufacturing
Wee felt it before in sense; but now wee know it by science.
Edward Misselden, The Circle of Commerce or the Balance of Trade, London,

Dawson for Nicholas Bourne (1623: 130)
Page 1 of 53
guments
19.1 New Perspectives on Cold War Economic

Theory: Adam Smith, David Ricardo, and Paul
Samuelson Revisited
INITIALLY it is of some importance to gain a broader perspective of what has developed
into ‘general truths’ of the neoclassical economics during the Cold War.
The fact that David Ricardo’s theory of comparative advantage in international trade
dates back to 1817 conveys an impression that this principle has been ruling economic
theory since then. It is also assumed that David Ricardo merely solidified the free-trade
principles of Adam Smith. However, the following quote from the young Adam Smith
shows how far away his principles were from the logic of comparative advantage and neo-
liberalism:
When the legislature establishes premiums and other encouragements to advance

the linen or woollen manufactures, its conduct seldom proceeds from pure sympa
thy with the wearer of cheap or fine cloth, and much less from that with the manu
facturer or merchant. The perfection of police (i.e. policy), the extension of
(p. 508) trade and manufactures, are noble and magnificent objects. The contem
plation of them pleases us, and we are interested in whatever can tend to advance
them. They make part of the great system of government, and the wheels of the
political machine seem to move with more harmony and ease by means of them.
(Smith, 1812 [1759], vol. 1: 320; my emphasis)1
It is important to note that in his main work The Wealth of Nations (1776), Adam Smith
refers to ‘the invisible hand’ only once: when private individuals prefer English goods to
imported goods, which would happen around this time after about 300 years of England
protecting its own manufacturing industry. But this is an argument for when to give up
protectionism rather than against protectionism as a principle.
It is also worth noticing that the term ‘free trade’ historically did not at all have the un
equivocal meaning it is usually given today. In a 1920 volume Edwin Seligman—Colombia
University’s eminent economics professor and an avid collector of economics books—
makes the following point as regards the meaning of ‘free trade’ in Edward Misselden’s
work:
Free trade…denoted in those days something very different from what it signifies
today. It did not mean freedom to import goods without the payment of duty. On
the contrary…freedom to export goods as over against the companies which pos
sessed a monopoly of trade, like the East India Company…Almost all free traders
were in fact what we should today call protectionists.
(Seligman, 1920: ix)
Page 2 of 53
guments
Cold War economics—the theories that stood victorious after the fall of the Berlin Wall—
had its roots in David Ricardo in 1817. However, recent n-gram technology has made it
possible to illustrate how David Ricardo and his theory of ‘comparative advantage’ were
virtually neglected until Paul Samuelson brought them into the core of economics at the
start of the Cold War with two articles in The Economic Journal in 1948 and 1949. Com
munism advanced under the utopian slogan ‘from each according to his ability, to each
according to his needs’. With his renewed interpretation of David Ricardo, Paul Samuel
son produced a counter-utopia: under the standard assumptions of neoclassical econom
ics free trade would produce a tendency towards factor price (p. 509) equalization: the
prices of labour and capital would tend to equalize across the planet. This became the
‘noble lie’ of neoclassical economics and neo-liberalism and appeared to make industrial
policy superfluous.
The n-grams below show how Cold War economics brought David Ricardo out of the shad
ows. Compared to other English economists and economic philosophers—such as father
and son James and John Stuart Mill—David Ricardo had indeed been much less important
during the first 100 years after his 1817 theory (Figures 19.1 and 19.2).
FIGURE 19.1 The frequency of ‘David Ricardo’ (in

English) during the first 100 years after the 1817
publication of his main work, Principles of
Economics, compared to that of two other, then much
more famous, English economists
Source: Author and Google N-Gram.
Figure 19.2 Frequency of the term ‘comparative

advantage’ (in English) from 1817 until today
Note: As is clearly shown, the term was very little

used for the first 100 years of its existence, but its
use started with the birth of the planned economy
and exploded with the start of the Cold War in the
late 1940s.
Source: Author and Google N-Gram.
Page 3 of 53
guments
On the theoretical level, the Cold War (1947‒89) was fought between two cosmopolitical
theories. Neither in neoclassical/neo-liberal theory nor in communism was the nation
state a unit of analysis. In both theories the nation state was not seen as having a place.
Neoclassical economics is built on methodological individualism—no state needed—and
also in Marxism the state was supposed to wither away as obsolete after a brief ‘dictator
ship of the proletariat’. In practice, of course, it was not the state but the rights of individ
uals that withered away under communism.
Both political extremes were far too abstract to be practical guides to human soci
(p. 510)
eties. The implicit conclusion in 1989, however, was that because communism had been
proven to be wrong, neo-liberalism—the other political extreme—had to be perfect. This
belief has led to increasing poverty in many countries. A key economist in the historical
tradition in which this chapter is written, Gustav Schmoller (1838‒1917), clearly saw that
both political extremes were unfit for practical purposes. In his 1897 inaugural speech as
Rector of the University of Berlin, Schmoller expressed the hope that he had seen the end
of the two ideological extremes, Manchester Liberalism (today’s neo-liberalism) and com
munism. His characterization of both these ideologies was harsh: ‘the naïve optimism of
“laissez-faire” and the childish and frivolous appeal to revolution, the naïve hope that the
tyranny of the proletariat would lead to world happiness, increasingly showed their real
nature, they were twins of an ahistorical rationalism’ (Schmoller, 1897, my translation
and emphasis).
In practice, the ideological extremes of ‘the irrational twins’ opened up a wide spectrum
of possible economic policies. In Western Europe, Germany’s soziale Marktwirtschaft
(social market economy) (Müller-Armack, 1947)2 and Sweden’s Middle Way (Childs, 1936)
were successful models navigating the broad spectrum of opportunities between the ‘irra
tional twins’. After the 1989 fall of the Berlin Wall and with the rule of the Washington
Consensus, these ‘middle ways’ were in practice outlawed.
19.2 The Historical Roots of Industrial Policy

Theory
From manufacturing you may expect the two greatest ills of humanity, superstition
and slavery, to be healed.
Ferdinando Galiani (1728–87), Italian economist
More than two decades ago, when I was working on another paper on the history of eco
nomic policy, David Landes, the eminent Harvard economic historian, gave me a serious
warning: be careful, if not you are likely to end up with Adam and Eve. The point is well
taken: it is possible to argue that in his Poroi of 355 BC (also called Ways and Means),
Xenophon, arguing that a city’s economic problems could be improved by increasing the
size of the population, could be seen as understanding increasing returns or economies of
scale. In order to avoid the Adam and Eve problem, the story told in this chapter starts in
Page 4 of 53
guments
the 1400s, with practical policies, and with a systemic theoretical understanding that es
sentially starts in 1589 with Giovanni Botero’s bestselling On the Greatnesse of Cities
(Botero, 1589; Figure 19.3),3 relatively unknown today.
One important note on earlier European history, though: if we allow US economic

(p. 511)
historian Richard Goldthwaite to be our guide, the history of the rise of European civiliza
tion was in fact a process of import substitution. From the twelfth century onwards manu
factured goods which had previously been imported from the Levant started to be pro
duced in Europe. At the core of what is generally referred to as ‘the commercial revolu
tion’ was the growth of an industrial sector, Goldthwaite argues:
Initially Europeans had an unfavourable balance of payment…and initially they

made up the difference through piracy and plundering, but above all through the
export of silver and gold…With the growth and development of the European
economy, the Italians steadily improved their situation. They began to produce for
themselves the manufactured goods they had been importing, and by the fifteenth
century they had gone beyond import substitution and were exporting these very
goods to the Muslim world.
(Goldthwaite, 2009: 6)
This was later the logic of all European countries that have succeeded in upgrading their
industrial sector, starting with England in the 1400s up until and including classical de
velopment economics and the Marshall Plan.4 Only during very brief periods—the last of
them being from around 1989 until about now—has this principle of creating wealth
through structural upgrading of national productive sectors been abandoned. Just as in
dustrial companies almost by definition initially go through several years of loss, and
need ‘subsidies’ from the owners, so whole national systems also need years of subsidies
and protection before they are profitable. This period of protection—which the great lib
eral John Stuart Mill (1848) called infant industry protection—is still needed.
Old economic policies were carried out with varying degrees of understanding of the un
derlying principles. These policies were, in our meaning of the word, not based on what
we would normally call scientific analysis, but on ‘clues’, on a mode of inference called
abduction—or phronesis, Aristotle’s third form of knowledge (Reinert and Daastøl, 1997).5
This tradition is continued by the Italian philosopher Giambattista Vico (1668‒1744), in
the US philosophical tradition of C. S. Pierce and in economics in Nicholas Kaldor’s ‘styl
ized facts’. According to Pierce: ‘[Induction] can never originate any idea whatever. No
more can deduction. All the ideas of science come to it by the way of abduction. Abduc
tion consists of studying facts and devising a theory to explain them. Its only justification
is that if we are ever to understand things at all, it must be in that way’ (Lawson, 1989:
59–78).6 Pierce here describes the role played by the formulation of hypotheses as the
fundamental element in the creation of new knowledge. This reasoning is also in line with
the philosophy that lies at the root of the German historical school of economics, in the
eighteenth-century philosophers Gottfried von Leibniz and Christian Wolff.
Page 5 of 53
guments
No doubt most historical arguments have their points, but history shows that it
(p. 512)
has also been possible ‘to be right for the wrong reason’. We shall try to illustrate how the
creation of industrial policy was generated through abductive reasoning with a parallel
from the history of medicine. From the twelfth century sailors in the Mediterranean used
lemons to prevent scurvy (Villner, 1986: 110–13). This was a very effective policy. Howev
er, the explanation as to why this policy worked only appeared in the early twentieth cen
tury with the discovery of Vitamin C (Mervin, 1981: 14–15). In the meantime, acidity had
been seen as the curative element, which led to disastrous experiments with vinegar in
stead of citrus fruits.
Likewise, we would claim that it is entirely possible to establish good economic policies
for a time, without fully understanding the factors involved. For example, identifying
‘progress’, or the ability to pay more taxes, with the use of machinery in an increasing
number of industries would result in a beneficial public policy, even if the causal relation
ship between the use of machinery and wealth were not clearly established, or had been
‘unlearned’. The intuitive abduction often precedes what we would think of as a more ‘sci
entific’ type of knowledge. This view that abduction anticipates ‘science’ was expressed
in the above quote from English economist Edward Misselden—an economist who was
heavily influenced by Giovanni Botero—in 1623: ‘Wee felt it before in sense, but now wee
know it by science.’
England is an example of a country which appears to have created an industrial policy

without much theory, other than a clear recognition of what at Harvard Business School
used to be called ‘we are in the wrong business’. In the fifteenth century, England was a
poor nation, heavily indebted to her Italian bankers. Her main export was wool. But over
a relatively short period, England went from being a poor nation on the periphery of Eu
rope to being the leading nation of the world—from being a poor farming country to pos
sessing a global empire on which the sun never set.
There are different versions of the story as to how this happened. Probably a policy origi
nally intended to increase national revenues—a tax on the export of raw wool—ended up
having the unintended by-product of creating a domestic industry of woollen cloth. The
story is likely to have started earlier, but let us look at the version of Daniel Defoe—the
polymath historian best known as the author of Robinson Crusoe—who described the
English strategy retrospectively in his book Plan of English Commerce in 1728 (Defoe,
1728).
Henry VII, who came to power in 1485, had grown up in exile in Burgundy, where English
wool was being spun into cloth. The wealth he observed there contrasted sharply with the
poverty he later found in England. But, the Prince observed, the wealth in Burgundy de
pended totally on the import of English raw materials: wool and the Fuller’s earth used to
clean it. When he came to the throne of England, Henry employed the anti-Ricardian logic
which during subsequent centuries dominated, not only in England, but also on the conti
nent: don’t accept your comparative advantage, shape it. Manufacturers are rich, produc
ers of raw materials are poor. Therefore, to get rich and develop the country, we must
Page 6 of 53
guments
promote the production of manufactures. Selling manufactures is (p. 513) ‘good’ trade—in
today’s language, it makes us competitive—while selling raw materials is ‘bad’ trade
(King, 1721, vol. I: 3).
The Tudor strategy, which started with King Henry, was to bring England into the wealth-
creating downstream activities in wool manufacturing that he had observed abroad. Prac
tice preceded theory, which is not uncommon when historical circumstances conspire to
make certain facts, if not the deeper economic mechanisms at work or the theory explain
ing them, palpably visible, along with obvious policy measures. England’s economic
growth (her economy had been essentially static, like all pre-modern agriculture soci
eties, for centuries before this) started by observing the economic structure of richer
countries and emulating—copying and trying to improve upon—their economic structure
(Reinert, 2009b). This meant pushing into manufacturing, the quintessential (if today not
the only) home of ‘advantageous’ economic activities, the key to sustained growth.
The English strategy was gradual, and started with import substitution, which to this day
is a common first move in development plans. In 1489, tariffs on cloth were increased,
and local cloth manufacturing was encouraged. The Crown paid for skilled foreign work
ers to be brought in, and businessmen were paid bounties (in modern terms, subsidies)
for establishing textile manufacturing firms. And when sufficient manufacturing capacity
had finally been achieved to process all domestic wool production, England prohibited the
export of raw wool. During the next century, in the reign of Elizabeth I, the death penalty
was introduced for the export of raw wool from England. See Figure 19.4 for an example
of similar harsh punishments for sins against ‘the cult of added value’ in the Republic of
Venice.
19.3 The Key to Wealth as Urban Synergies

Created by Adding Value to Raw Materials un
der a Large Division of Labour: Giovanni
Botero (1589)
I first met practical industrial policy as a young student and assistant to the professor of
Spanish at the Latin American Institute at the University of St. Gallen in Switzerland. In
1972, the Swiss Federal Technical Cooperation, in cooperation with UNCTAD in Geneva,
organized an export-promotion training course for representatives of Spanish-speaking
South American nations. I was recruited to travel to Latin America to select the candi
dates who had been presented by the local governments, and also to organize the part of
the course that took place in St. Gallen in the summer of 1972.
The core idea of the course was to promote Latin American exports with higher value
than the traditional raw materials. In cooperation with the local Swiss embassies, (p. 514)
my task was also to pick one product from each country on which the participants, in ad
dition to theoretical courses, should do practical market research during their two
Page 7 of 53
guments
months’ stay in St. Gallen and Geneva. Although the idea was completely in line with clas
sical development economics, as it was still practised at the time, neither I—nor probably
anyone else—had any idea that this idea of added value had been at the very core of theo
ries explaining the differences in national incomes. Value added had been the key expla
nation for what created wealth in the few cities of wealth in Europe—like Amsterdam,
Florence, and Venice—since Giovanni Botero’s work On the Greatness of Cities (Botero,
1589). By 1671, Botero’s book had been published in around forty-two editions in all the
important European languages,7 and his thoughts were also spread in German-speaking
areas by Veit Ludwig von Seckendorff’s 1656 book on The German Principality
(Seckendorff, 1656)8—which remained continuously in print for 100 years—and in English
through the works of Francis Bacon.9
Two apparently different economic traditions, cameralism and mercantilism, seem to have
grown out of the extremely widely diffused works of Giovanni Botero (1544–1617)10 as a
common platform and point of reference. Botero, in turn, built on two much older tradi
tions: his work Ragion di Stato—of which On the Greatness of Cities is part—satisfied the
oldest tradition in economic policy advice, the tradition of Fürstenspiegel, a kind of
owner’s manual to the numerous small states of Europe.11 Botero’s other main work, Re
lazioni Universali (Botero, 1622) satisfied another very old tradition: the need for surveys
and the fact-finding missions’ quest for geographical, cultural, and anthropological knowl
edge.12 All in all, at the time when the knowledge of the whole world and its cultures be
came codifiable, Giovanni Botero provided an unusually complete range of social sci
ences.13 It is worth noting that in contrast to the many utopias of the period, Botero’s rea
soning was based on observations of history and facts. In his work he clearly distances
himself from ‘bullionism’—the idea that a nation’s wealth consists in the amount of pre
cious metals owned—of which mercantilism is sometimes accused.
Giovanni Botero was born in the small town of Bene Vagienna in the province of
(p. 515)
Cuneo in the Italian Piedmont region. As a Jesuit, he was keenly interested in non-Euro
pean cultures. From the point of view of now long-standing Western Eurocentrism, the
ability of the Jesuits to engage in two-way cultural communication reminds us (p. 516) that
Eurocentrism is not necessarily a ‘natural’ state of affairs. Jesuit Matteo Ricci (1552–
1610), a contemporary of Botero, ventured with a small group to China, where he trans
lated not only Christian and Western scientific texts into Chinese, but also Chinese texts
into Latin. By entering foreign cultures—from the Chinese to the Guaraní in South Ameri
ca—Jesuit travellers also played the role of anthropologists. As one observer says, Botero
‘brought together an immense mass of geographical and anthropological information,
which he tried to organize according to broad methodological categories (like “re
sources,” “government” and “religion”)’ (Rubies, 2000: 294).
Little appears to unite Sir Walter Raleigh (1554–1618), Francis Bacon (1561–1626), utopi
an Tommaso Campanella (1568–1639), English economist Edward Misselden (1608–54),
Spanish economist Gerónimo de Uztáriz (1670–1732) and Swedish technologist and econ
omist Christopher Polhem (1661–1751). But one thing does: they all convey key insights
found originally in Giovanni Botero, but following the practice of the time they do not
Page 8 of 53
guments
quote him or anyone else as to the origins of these insights. As mentioned, the work of
the first German bestseller, Veit von Seckendorff (1626–92) is clearly also very much in
fluenced by Botero (Reinert, 2005). There are still thirty editions of Botero’s works (main
ly uncatalogued) from the time of Seckendorff’s librarianship in the Gotha Library, which
he formed for Ernest the Pious (Ernst der Fromme) of Sachsen-Gotha-Altenburg, and
Botero was on the reading list Seckendorff made for the education of princes. The large
number of translations of Botero’s works testify to his strong influence on the European
seventeenth-century zeitgeist (Reinert and Reinert, 2018).
Botero argues that one of the reasons for the economic superiority of cities over the coun
tryside is that the ability to invent new things is much greater there than in the country
side. Here we find an early trace of ‘Schumpeterian’ thinking, which was followed up by
Francis Bacon’s 1625 essay ‘Of Innovations’.14
Botero’s Ragion di Stato (1589) was the first modern economics bestseller. In English, Ra
gion di Stato came to be called Reason of State and in German Staatsräson. In his 1925
work on Staatsräson, Friedrich Meinecke mentions Botero’s many followers and the ‘true
catacombs of forgotten literature’ which follow in Botero’s path.15 (A new translation of
Botero’s The Cause of the Greatnesse of Cities has an excellent introduction by Geoffrey
Symcox.)16
Figure 19.3 The cover of the first edition of the book

that laid the earliest foundations for a theory of in
dustrial policy: Botero’s 1589 volume
Note: Botero (1589) was translated into all important

European languages—the first translations in Ger
many were from Italian into Latin—and the book
reached a record of forty-two editions between 1589
and 1671.
Source: Reinert family library.
Page 9 of 53
guments
The understanding that grew out of Botero’s work was that only in barren areas lacking
natural resources and with limited possibilities for food production—but in favourable ge
ographical positions such as Venice and Amsterdam—would economic development tend
to arise ‘naturally’. In virtually all countries, heavy-handed government (p. 517) policies
were required during the transition from diminishing-returns activities (agriculture) to in
creasing-returns activities (manufacturing), as they were identified by Antonio Serra in
1613 (S. Reinert, 2011); or from ‘natural activities’ to ‘artificial activities’, to use the later
terminology of Thomas Mun (Mun, 1664). This was the essence of the thinking that
Botero’s influence turned into the economic mainstream at the time. What Venice and the
Dutch Republic had achieved—rather than the policies of Venice and the Dutch Republic
—was the object of attention of foreign economists and foreign rulers alike. Edward Mis
selden argued in 1620 that it was necessary to understand the difference between heaps
of stones and logs and a house. Between them was the added value of human knowledge
and skills. This rings a bell when reading Botero, the first English translation of which
was in 1605:17
Some will aske me; whether Fertilitie of Land, or Industrie of Man, importeth
more to make a place Great, or populous? Industrie, assuredly. First because Man
ufactures framed by the skilfull hand of Man, are more in number,18 and price,19
than things produced by Nature: For Nature giveth matter, and subject: but the
Curiositie and Art of Man addeth unspeakable varietie of formes. Wool, from Na
ture, is a rude and simple Commoditie: What fair things, how various, and infinite,
doth Art make out of it?20
Compare the Marbles, with the Statues, Colossuses, Columns, Borders, and infi
nite other Labours, taken. Compare the Timber, with the Galleys, Galleons, Vessels
of many sorts, both of Warre, Burthen, and Pleasure: Compare also the Timber,
with the Statues, the Furnitures for Building, and other things innumerable, which
are built with the Plane, Chesill, and Turners-Wheele. Compare the Colours with
the Pictures.21
Page 10 of 53
guments
Figure 19.4 An example of the practical conse

quences of ‘the cult of value added’
Source: Contemporary Venetian broadside. Reinert

family collection.
Botero’s basic ideas around the geographical clustering of economic activities leading to
progress have had many modern followers. Although geography was obviously present in
other economists, in relatively modern times it was not until the publication of Johann
Heinrich von Thünen’s first volume of Der Isolirte (sic) Staat (The Isolated State) (von
Thünen, 1826)22 that location theory based around a core industrial city—as Botero had—
was rediscovered. Von Thünen (1783‒1850) also places the industrial city at the geo
graphical and economic core of the modern state (Figure 19.5).
Figure 19.5 von Thünen’s map of a modern state,

with the industrial city at its core
Source: von Thünen (1826).
Page 11 of 53
guments
Von Thünen drew a map of civilized society with four concentric circles around a core of
increasing returns activities—the city. Moving outwards from the city core, the use of cap
ital and advanced skills gradually decreases and the use of nature gradually (p. 518)
(p. 519) increases. Near the city the most perishable products are produced, such as dairy
products, vegetables, and fruit; grain for bread is produced further out, and in the periph
ery there is hunting in the wilderness. Economists today have rediscovered von Thünen’s
approach to economic geography, but many miss the crucial point he stresses, which
stands on the lines of the first page of Der Isolirte Staat: ‘Man denke sich eine sehr große
Stadt in der Mitte einer fruchtbaren Ebene gelegen’; a very big city in the middle of a fer
tile plain is at the core of society. As with Botero, here also the city is at the core of the
system. It is worth noticing that already very early on economists distinguished between
manufacturing cities (like Venice or Milan) from which wealth spread, and administrative
cities (the typical example was Madrid) that played more of a parasitic role.
Since von Thünen was a farmer and mainly interested in the improvement of agriculture,
he does not pay too much attention to the factories in the city, even though they are also
mentioned in his book. Von Thünen did not argue against the accepted knowledge of the
time that a state needed manufacturing industry, and that this industry needed tariff pro
tection. Underlying what happened in von Thünen’s outer circles was a development ma
chine at the core of the concentric circles—the urban increasing-returns industries (man
ufacturing)—which, for a time at least, needed targeting, nurturing, and protecting. In
other words, the presence and state of development (p. 520) of the core city would also de
termine the standard of living in the rest of the country, in these outer circles.
In von Thünen’s map the most ‘modern’ sector—manufacturing—formed the city core,
and the most ‘backward’ sector—hunting and gathering—formed the periphery furthest
from the city. Moving outwards away from the city, the use of nature as a factor of pro
duction increases and the use of capital decreases. Only the city will have authentic in
creasing returns, free from nature’s flimsy cyclicality and supply of resources (land, min
erals) of different qualities.
As one moves from the city towards the periphery, man-made comparative advantage
(subject to increasing returns) gradually diminishes and nature-made comparative advan
tage (subject to diminishing returns) increases. As we move outwards in the circles, the
carrying capacity of the land in terms of population also diminishes.
The importance of the linkages and synergies for agricultural development, seeing the
benefits accruing to agriculture from the proximity of manufacturing, was not uncommon
in eighteenth-century economics: ‘Husbandry…is never more effectually encouraged than
by the increase of manufactures,’ says David Hume in his History of England (1767).
Von Thünen’s model pictures all the stages of development inside one nation state, one
labour market, one school and university system, and one social security system. The syn
ergies that David Hume points to are partly the result of an equal access to basic institu
tions and government services accruing to the ‘hunters’ in the outermost circle as well as
to the city dwellers. The local city market does to national agriculture what an interna
Page 12 of 53
guments
tional market can never do. Proximity to a city in the same labour market, rather than
abroad, assures employment for the second and third son on the farm. The wage pressure
from the city activities makes labour more expensive in the countryside, allowing for
technological change that would never be profitable with low wage rates. The proximity
to the city gives access to advanced technology and expertise that a rural-only nation
would never achieve. All in all, von Thünen’s model provides a useful picture for develop
ment as a synergy between town and countryside.
Late in the nineteenth century, in his Principles of Economics (Marshall, 1890) and in an
earlier work (Marshall and Marshall, 1879), Alfred Marshall introduces ‘industrial dis
tricts’. In 1909, Alfred Weber publishes Über den Standort der Industrie (Theory of the
Location of Industries). After the Second World War, Botero’s ideas of geography-based
economic agglomerations appear in August Lösch’s The Economics of Location: A Pioneer
Book in the Relations between Economic Goods and Geography (1954), with French econ
omist François Perroux as ‘growth poles’ (Perroux, 1955) and with Harvard Business
School’s Michael Porter as ‘clusters’ (Porter, 1990). In Italy, Giacomo Becattini re-intro
duced ‘Marshallian industrial districts’, and ‘The Third Italy’—the economic power of the
many small and medium-sized enterprises in Central Italy—generated much attention. In
the Third World, the importance Albert Hirschman gave to ‘linkages’ also reflects this
way of thinking.
It is worth noticing that some of these ideas had clear Schumpeterian influences. August
Lösch was a student of Schumpeter in Bonn, and François Perroux wrote a 216- (p. 521)
page introduction to the French translation of Schumpeter’s Theory of Economic Develop
ment (1935).
19.4 Increasing Returns: From Antonio Serra

(1613) to Alfred Marshall (1890) and Paul
Krugman (1981)
I apprehend (the elimination of Diminishing Returns) to be not only an error, but
the most serious one, to be found in the whole field of political economy. The ques
tion is more important and fundamental than any other; it involves the whole sub
ject of the causes of poverty;…and unless this matter be thoroughly understood, it
is to no purpose proceeding any further in our inquiry.
John Stuart Mill, Principles of Political Economy (1848)
Joseph Schumpeter gave Antonio Serra the honour of having been ‘the first to compose a
scientific treatise…on economic principles and policy’ (Schumpeter, 1954: 195).
Schumpeter’s succinct description of Serra’s work confirms the author’s anti-bullionist
bias, the normal criticism against mercantilists:
Page 13 of 53
guments
Its chief merit does not consist in his having explained the outflow of gold and sil
ver from the Neapolitan Kingdom by the state of the balance of payments but in
the fact that he did not stop there but went on to explain the latter by a general
analysis of the conditions that determine the state of an economic organism. Es
sentially, the treatise is about the factors on which depend the abundance not of
money but of commodities—natural resources, quality of the people, the develop
ment of industry and trade, the efficiency of government—the implication being
that if the economic process as a whole functions properly, the monetary element
will take care of itself and not require any specific therapy.23
Regardless whether this long theoretical tradition which dominated Europe until the late
eighteenth century be labelled mercantilist, Colbertist, or cameralist, Botero’s narratives
and Serra’s theories in a sense laid the foundations for all three schools by establishing
two crucial dichotomies in economics. The taxonomies Serra established are important
for understanding the wealth and poverty of nations, and indeed provide a continuing key
to what his contemporaries called buon governo, or good government (Patalano and S.
Reinert, 2016).24
(p. 522) Serra’s two dichotomies, I will argue, were in the recent past still part and parcel
of all three dominant ideologies and their economic policies in the 1930s,25 but were sub
sequently lost with the formalization of modern neoclassical economics, and are conspicu
ously absent in the rather superficial discussion of good governance presented by the
Washington Institutions today. The first is a dichotomy separating economic activities sub
ject to increasing returns from those subject to diminishing returns. Putting ‘manufactur
ing’ in a different category from ‘raw materials’ from the point of view of policymaking
had already been the core element of the English Tudor strategy from 1485, promoting
woollen manufactures at the expense of the export of raw wool by slowly building up the
export duties on raw wool. (For a thorough discussion see Reinert, 2007.)
There had been scattered references to the wisdom of such practices,26 but what Giovan
ni Botero did in his volume was to elaborate the vision of the role of manufacturing, the
insight that civilization was based on adding knowledge and value to nature’s raw materi
al, into a full-fledged theory of economic development.
In Botero’s volume the degree of economic and societal development manifested itself as
the ability of a city to hold the maximum number of inhabitants in satisfactory conditions.
This again was the result of the number of different professions that were exercised in
the city: in other words, the degree of division of labour—the degree of complexity—
would determine the wealth of a city. This argument has recently been convincingly recre
ated and proven (Hausmann, Hidalgo et al., 2011). Botero explained the mechanisms, but
Serra’s big contribution was to explain why. He did so by highlighting the key difference
between the production of raw materials and manufactured goods, that is, what happens
to the development of costs as production is increased. In manufacturing there were in
creasing returns, and the synergies of the multitude of artisanal and manufacturing activ
ities, each of them subject to increasing returns, produced the synergies, linkages, and
Page 14 of 53
guments
cumulative causations that Botero and Serra saw as the main factor attracting so many
people to the city-states that had specialized in manufacturing.
In the first edition of his Principles of Economics, Alfred Marshall, the founder of neoclas
sical economics, emphasizes the crucial importance of diminishing returns: ‘The tendency
to a Diminishing Return was the cause of Abraham’s parting from Lot,27 and of most of
the migrations of which history tells’ (Marshall, 1890: 201). Today the migration experi
enced in Europe is from nations dominated by diminishing-returns activities (for example,
Eritrea) to nations where increasing-returns activities dominate (for example, Holland).
The second dichotomy is that separating the financial sector from the real economy. As al
ready mentioned, this dichotomy is of course much older than Serra’s work. (p. 523) An
academic expression of the problems that may arise when the financial and monetary
spheres decouple hails back at least to Nicolaus Oresme (c.1320–82)28 and to the Bible.
This dichotomy is not there in Botero’s Greatness of Cities—which concentrates on the re
al economy—but it is very much there in Antonio Serra’s discussion with his contempo
rary MarcAntonio de Santis on how to deal with the outflow of money from the
kingdom.29 De Santis was of the opinion that the lack of money in the kingdom was due to
the excessively high level of the exchange rate. On the basis of his theory, several mea
sures had been introduced to manage the rate of exchange and limit the export of metal
lic money, without positive results. Serra, on the other hand, starts by noting that there
are countries with no natural supplies of metals from domestic mines that nevertheless
manage to have an abundance of money. In other words, Serra asks: why on earth do the
gold and silver which flow into Spain from the New World end up accumulating in places
like Venice, which have no mines and raw materials at all?
Serra’s reply was based on Botero’s analysis of what attracted people and resources to
some cities and not to others, above all the abundance of different manufacturing indus
tries. In other words, the solution to the problems posed by dichotomy two—the conflict
between the financial and the real economy—lies in observing the insights emanating
from dichotomy one: money will leave the cities and countries with no increasing-returns
activities, being attracted to cities with manufacturing and increasing returns. In
Schumpeter’s quote above, he emphasizes Serra putting the real economy centre stage:
‘if the economic process as a whole functions properly, the monetary element will take
care of itself and not require any specific therapy.’30
In fact, digging deeper into Serra’s arguments, we can argue with him that de Santis’s
fiddling with monetary variables—as long as these monetary variables did not positively
affect the health of the real economy—were not only completely in vain, but potentially
destructive to the real economy. The present tragedy of Greece inside the European
Union carries with it the same type of discussion as that between Messrs de Santis and
Serra more than 400 years ago.
The jury is still out on whether the policies carried out from the start of the financial cri
sis until the present (2020) by the Federal Reserve—and even more so those of Mario
Draghi and the European Central Bank—again will justify Antonio Serra’s warning: fid
Page 15 of 53
guments
dling around with financial variables, which in reality do not improve conditions in the re
al economy, will not help, but will probably worsen the situation. Schumpeter saw the
need for economic ‘cold showers’ provided by financial crises, because unproductive capi
tal lost its value and the system was reset with a clean slate. From that point of view, we
can ask whether Draghi, by providing more liquidity and more debt, is currently prevent
ing Europe from taking the necessary ‘shower’, liberating itself from (p. 524) a huge debt
overhang and kick-starting the real economy. Increasing debt and demand contraction in
vicious circles—as a result of austerity policies—seem to prevent the virtuous circles that
originate in Serra’s increasing returns to scale (that is, falling unit costs as the volume of
production increases).31
The key factor being put back into trade theory is, again, increasing returns, and the key
person in the process of rediscovery is MIT’s Paul Krugman. Krugman correctly observes
that economic theory ‘has followed the perceived line of least mathematical resistance’.32
His explanation is that the reason scale effects were excluded was that the profession
was unable to express these mathematically.
Starting in 1979, Krugman published a series of articles introducing increasing returns in

international trade theory. His 1981 article (Krugman, 1981) models a world where an ini
tial discrepancy in capital/labour ratio exists between two countries or groups of coun
tries. A period of increasing international trade follows, where only the industrial sector
works under increasing returns to scale. The result of this is a world divided into two
groups, a rich industrialized centre and a poor underdeveloped periphery. In this article,
one of the very few in which he includes both increasing and diminishing returns, Krug
man, a later recipient of the Nobel Memorial Prize in Economic Sciences, picks up a mod
el of the increasing/diminishing returns dichotomy from US economist Frank Graham
(1923) and also recognizes the result of this mechanism: ‘This might mean that in addi
tion to exporting capital, the industrial region might, in the second stage of growth, begin
importing labour—a point also noted both by John Hobson and by the same Vladimir
Lenin.’ Krugman’s references here place the increasing-vs.-diminishing-returns dichoto
my in the context of works on imperialism by Hobson (1902) and Lenin (1939). In the
same article, Krugman also refers to development economics and to Myrdal, Frank,
Baran, and Wallerstein.
This breakthrough in international trade theory was the result of using models originat
ing in the study of imperfectly competitive markets in the field of industrial economics.
Krugman inadvertently opened a Pandora’s box, where international markets are no
longer fully competitive, and where countries may grow poorer in the presence of free
trade than under autarky. Paradoxically, the wave of Reaganomics free-market policies
which hit the developing countries in the early 1980s coincided with the first proof of
neoclassical trade theorists that government intervention really could improve the free-
trade situation of a poor country. After the early 1980s, however, Krugman seems not to
have used models with both increasing and diminishing returns. Schumpeter had re
ferred to using overly abstract models with limited practical relevance such as Ricardian
trade theory as ‘the Ricardian Vice’. To this I added the (p. 525) concept of ‘the Krugman
Page 16 of 53
guments
ian Vice’: having much more relevant theory but refusing to apply it in practical economic
policy.
The core of nineteenth-century protectionism is exactly what Krugman points out: by pro
tecting the national market for national industries the market was extended, because the
increasing returns which accrued to new industries more than outweighed the initial in
crease in price caused by the protection. A higher initial price for industrial goods was
traded off for an even higher increase in real wages and profits in the protecting nation—
a phenomenon which is inexplicable without the existence of imperfect competition and/
or increasing returns. The worker as a wage earner would be more than compensated for
the initially higher prices that the same person would have to pay as a consumer.
19.5 How Economic Activities Differ, Learning

Curves, and the Sequencing of Technological
Change
Figure 19.6 ranks the development of fifty-one industrial sectors in the United States
from 1899 to 1939 according to three factors: (a) the increase in production (output), (b)
the number of employees (wage earners) in the industry, and (c) the increase in produc
tivity (measured as the lowered use of manpower behind each unit of production). The
sector with the highest increase in output, almost 200,000 per cent, was the automotive
industry. In this industry, however, the number of wage earners only increased by just
over 20,000 per cent. The number of wage earners per unit of product (the third column)
was reduced by more than 85 per cent. Figure 19.6 is also an indication of Verdoorn’s
Law: that industries with the largest increase in volume of production also tend to
present the largest growth in productivity.33
Figure 19.6 How industries differ
Source: Fabricant (1942).
Page 17 of 53
guments
Assuming perfect competition, the differences in Figure 19.6 would not have had any im
pact on the rate of economic development in countries that had an automotive industry
vs. countries that did not. In reality, however, the automotive industry became a wage
leader, pulling up the general wage level in the United States. In March 1914, Henry Ford
doubled the wages of his workers, from US$2.50 a day to $5 a day. The Marshall Plan pol
icy to distribute important high-growth/high-productivity-growth industries among all
large nations dominated until the end of the 1980s. Italy’s car industry was protected
from foreign competition by an annual import ceiling of 40,000 engines for assembling
the Italian version of the Morris Mini, the Innocenti. With the integration of Spain into the
EEC during the 1980s, tariffs were lowered gradually, (p. 526) (p. 527) while support was
given to local industries, particularly to the important Spanish automotive industries with
many subcontractors. The last 10 per cent duty on Japanese cars imported into the Euro
pean Union was abolished in December 2017. Although the theoretical reasons for these
policies were gone from the Washington Institutions paradigm, in the rich countries the
policies themselves lingered on in practice.
During what we could call the nation-based development paradigm, most countries had
the whole range of production represented in Figure 19.6. There was of course much
trade, but mostly symmetrical trade: France and Germany exporting cars to each other.
Smaller countries also had car production: Holland had their DAF and Yugoslavia had a
car called the Yugo, rudimentary and inexpensive. But with globalization came Daewoo
and Hyundai from a military-dominated South Korea, where wages were lower than in Yu
goslavia. Korea’s strategic significance for the United States—especially after the defeat
in Vietnam—allowed Korea to use protectionist rules, contrary to what was prescribed by
the Washington Consensus. An independent communist Yugoslavia could not, and the Yu
go disappeared.
This is but one very brief example of what happened when the global economy took over
from the nation-based economy. Many countries lost the high value-added activities on
the left side in Figure 19.6, and world trade became much more of a winner-takes-all
game (see Table 19.1).
Page 18 of 53
guments
Table 19.1 How the game changed
Nation-based capitalism Global capitalism
High-tech, high-growth in High-tech, high-growth industries

dustries present in all coun disappear in peripheral countries
tries of any size (from Greece to Mexico)
Move advanced economic ac Move human beings from one coun
tivities to lagging countries try to the other.
Adjust exchange rates (Eu Freeze exchange rates (Europe)

rope)
Create jobs in order to solve Create money in order to solve

crises (Keynes) crises (EU Central Bank President
Mario Draghi)
Strong government and Gradual power shift to the financial

labour unions (balance of sector (the rule of the one per cent)
countervailing powers)
Harmonization (harmony cre Polarization (spontaneous chaos)*

ated through economic poli
cy)
Note: My book Spontaneous Chaos was published in Norwegian in

2009 and has been translated into Russian and Serbian.
Source: Author.
Figure 19.6 shows a virtual ‘productivity explosion’ in the automotive industry to the left
in the graph. Figure 19.7 shows the First Industrial Revolution as another such ‘produc
tivity explosion’. All European countries attempted to get their share of this by attracting
such activities to their own states: they would bring higher wages, higher profits, and
higher tax incomes to the state treasury.
Page 19 of 53
guments
Figure 19.7 Productivity explosion: cotton spinning
Source: Carlota Pérez/Reinert (2008).
(p. 528) Figure 19.8 shows similar dynamics in the form of a learning curve, plot
(p. 529)
ting the number of man-hours per unit of production. A similar tool—the experience
curve34—plots the lowering of total costs. These are useful tools for any industrial policy
research. The natural dynamics of global competition tends to farm out mature products
with flat learning curves to poor countries: the lack of technical change means that such
products appear ‘labour intensive’ and as such suitable for low-wage countries. The
mechanism—that technological dead-ends tend to be farmed out to poor countries—is a
key tool automatically reinforcing the vicious circles of national poverty.
Figure 19.8 US learning curve in men’s shoes, 1850–

1936
Note: An important aim for industrial policy is to pro

duce where the learning curves are steep.
Source: Reinert (1980).
Page 20 of 53
guments
19.6 On the ‘Quality’ of Economic Activities:

Barriers to Entry, Hierarchies of Skills, and Dy
namic Imperfect Competition
Salomon Fabricant’s chart—reproduced as Figure 19.6—shows the vast differences in vol
ume of production, productivity growth and employment between various US industries
in the early twentieth century. However, the chart does not distinguish between qualita
tively different types of competition.
By the time the third edition of his Principles of Political Economy and Taxation (1817) ap
peared in 1821, David Ricardo had probably heard that the extensive use of machinery
might make his theory of comparative advantage invalid. In this edition Ricardo therefore
explicitly assumed that the use of machinery would simply lower the price of the goods.
In that case, Fabricant’s taxonomy of economic activities would not have any conse
quences. Especially so because David Ricardo’s trade theory—the essence of modern cap
italist trade theory—does not at all consider capital as a factor of production. The as
sumption behind this theory—one which was later to be picked up by Karl Marx—is the
labour theory of value.
On the basis of this I distinguish between two types of competition. One is classical com
petition, based on the classical economist David Ricardo, where the only thing technologi
cal change does is to lower the prices of production. Its counterpoint, collusive competi
tion, is collusive in the sense that, as a result of dynamic imperfect competition and high
barriers to entry, capitalists (profits), workers (wages), and government (taxes) are able
to ‘collude’ and prevent prices from falling at the same rate as the productivity increases
(Reinert, 1994). This is a main reason why industrial policy is important. Table 19.2 sets
out the mechanisms which create, respectively, collusive and classical modes of diffusion
of productivity improvements.
Page 21 of 53
guments
Table 19.2 Characteristics of the collusive and classical modes of diffu

sion of productivity improvements
Collusive Classical
Characteristics of mode
Divisibility of in Indivisible, comes in Divisible

vestments ‘chunks’ Perfect (competitive
Degree of perfect Imperfect (e.g. market for technolo
information patents, internal R&D) gy itself)
Source of technol Internal, or external in External

ogy from user big chunks = high de
company point of gree of economies of
view scale
Barriers to entry Increase No change

Industry structure Increases concentra Neutral
Economies of scale tion No change
Market shares Increase Unimportant
Very important
How benefits spread
GNP as measured Highly visible (at pro Tends not to appear

ducer level) (Solow- paradoxes)
Profits level Increases stakes: pos No change

sibilities for larger
profits or losses
Monetary wages Increase No change

Real wages (na Increase Increase
tionally) No change Decreases
Price level No change Turns against indus
Terms of trade tries experiencing
technological
progress
Page 22 of 53
guments
Examples of inno New pharmaceuticals, Electricity, online

vations in the two automotive paint pro sales of hotel book
groups duction, Microsoft, ings and used
Google, Facebook books, use of PCs,
dispersion paint
production, contain
ers
Where found Traditionally mainly in In primary and ter

industry, in recent tiary industry, use of
products and process new basic technolo
es, in IT-related mo gies, mature indus
nopolies try
Source: Reinert (1994), modified.
(p. 530) In Fabricant’s graph (Figure 19.6) we find that industry 4, beet sugar, appears to
have a very high score in output and productivity increase. However, we can safely as
sume that a commodity like sugar will operate under near-perfect competition, so we can
not expect profits or wages to increase as they do in the automotive industry. In fact we
find that in all developed countries beet sugar—which in productivity is inferior to tropi
cal cane sugar—is, like so many agricultural products, subsidized by governments.
The conceptual quality index of economic activities (Figure 19.9) adds the dynamics of
profits, wages, and taxes from oligopolistic competition and the lack of such (p. 531)
(p. 532) (p. 533) dynamics from perfect competition (commodity competition) to
Fabricant’s graph. New inventions and innovations enter at the top—initially under near-
monopoly conditions—but fall, with very different gravity, towards perfect competition.
Economic theory generally only defines well monopolies (black, at the top of the quality
index) and perfect competition (white, at the bottom).
Page 23 of 53
guments
Figure 19.9 The quality index of economic activities
Source: Reinert (1994: 184).
Most economic activities are in between, and in the absence of new innovations tend to
fall towards perfect competition. For example, when a patent expires on a medicine, this
product would experience a sharp drop on the quality index (towards perfect competi
tion). As a product matures and the learning curves flatten out (Figure 19.8), the same
thing will happen. The fact that only continuous innovations will secure continuous prof
its is the main source of dynamism in an economy. Rich countries produce goods with a
high score on the quality index, poor countries tend to produce goods with a low score
(see Reinert, 2007, for more details).
These dynamics make it possible to theorize about industrial policy around ‘good’ and
bad’ economic activities, and around the recent EU concept of ‘smart specialization’ and
its counterpart, ‘unsmart’ specialization (Reinert, 2018). Table 19.3 sets out their con
trasting characteristics.
Page 24 of 53
guments
Table 19.3 ‘Good’ and ‘bad’ economic activities
Characteristics of economic ac Characteristics of bad/un

tivities that are good/smart to smart economic activities
specialize in
Increasing returns to scale (higher Diminishing returns (higher

volume = lower costs) volume = higher costs, after
a point)
Rapid technological development Slow technological change

(steep learning curves) (flat learning curves)
Technical change leads to higher Technical change tends to

wages to the producers (Fordist lower prices to the con
wage regime) sumers
Dynamic imperfect competition Perfect competition (com

modity competition)
Have stable prices Show strong price fluctua

tions
Generally skilled labour Generally unskilled labour
Create a middle class Create ‘feudal’ class struc

ture
Irreversible wages (‘stickiness’ of Reversible wages

wages)
Create large synergies (linkages, Create few synergies

clusters)
Source: Reinert (2007).
Empirically the effect of these activities can be measured as in Table 19.4.
Page 25 of 53
guments
Table 19.4 Average wage per cluster category in Europe, based on

255 European regions
Cluster catego Average Cluster category Average

ry wage, € wage, €
Aerospace 44,718 Paper products 24,995
Financial ser 43,930 Sporting and 23,498

vices children’s goods
Biotech 42,384 Building fixtures 22,827
Pharmaceuticals 40,735 Stone quarries 21,183
Analytical instru 39,519 Processed food 20,993

ments
Chemical prod 38,381 Construction 20,894

ucts
Information 37,360 Construction ma 20,063

technology terials
Oil and gas 36,073 Textiles 17,902
Telecommunica 35,960 Jewellery and pre 16,303

tion equipment cious metals
Production tech 32,371 Furniture 16,131

nology
Automotive 29,399 Leather products 15,594
Plastics 29,066 Maritime 14,274
Medical devices 28,927 Tourism and hos 13,961

pitality
Power genera 28,927 Agricultural prod 13,852

tion and trans ucts
mission
Page 26 of 53
guments
Lighting and 28,767 Tobacco 13,567

electrical equip
ment
Transportation 27,462 Education and 13,132

and logistics knowledge cre
ation
Heavy machin 26,393 Apparel 11,885

ery
Metal manufac 26,269 Footwear 11,238

turing
Business ser 25,964 Entertainment 11,034

vices
Distribution 25,888 Farming and ani 3,859

mal husbandry
Media and pub 25,556

lishing
Source: http://www.foreurope.eu/fileadmin/documents/pdf/Workingpa
pers/WWWforEurope_WPS_no014_MS47.pdf. Accessed September
2019.
19.7 Colonialism and Industrial Policy

Two recent works on early economics bestsellers—defined as economics books appearing
in ten editions or more before 1850 (Reinert et al., 2017)35—reveal that many works in
economics, which were very influential in their time, are today virtually forgotten. And
forgotten to the extent that there are not even entries on them in Wikipedia.36
In this context we shall only mention one work, published by Joshua Gee in 1729, which is
typical of English industrial policy before Adam Smith. It is also typical of much of Eng
lish colonial practice until long after Smith and Ricardo.
The massive title, in the tradition of the day, reads:
The trade and navigation of Great-Britain considered: shewing that the surest way
for a nation to increase in riches, is to prevent the importation of such foreign
commodities as may be rais’d at home. That this Kingdom is capable of raising
Page 27 of 53
guments
within itself, and its colonies, materials for employing all our poor in those manu
factures, which we now import from such of our neighbours who refuse the admis
sion of ours. Some account of the commodities each country we trade with take
from us, and what we take from them; with observations on the Balance.
London: Printed by Sam. Buckley, in Amen-Corner. MDCCXXIX
There were at least twenty editions of Gee’s work between 1729 and 1780, and
(p. 534)
the issues are unusually widely spread geographically. There are English editions pub
lished in London, Glasgow, and Dublin, French translations (the first in 1749), published
in London, Amsterdam, and Geneva, Dutch (1750), Spanish (1753), and German (in
Copenhagen, 1757).
One factor leading both to the geographical spread of this book, and to its later oblivion is
probably that Gee was not only very straightforward when he described England’s inter
est in protecting its manufacturing industry, he was also unusually honest about the in
tention of colonialism being the opposite, to hinder manufacturing in the colonies:
That all Negroes shall be prohibited from weaving or spinning or combing of Wool,
or manufacturing hats…Indeed, if they set up manufactures, and the Government
afterwards shall be under a Necessity of stopping their progress, we must not ex
pect that it will be done with the same ease that now it may.
At the time it must have occurred to those who published the 1730 edition of Gee’s vol
ume that not only were blacks subject to this policy, so was Ireland. In 1779, John Hely-
Hutchison—then Provost of Trinity College, Dublin—anonymously published The Commer
cial Restraints of Ireland considered in a series of letters addressed to a Noble Lord, con
taining an historical account of the affairs of that Kingdom, so far as they relate to this
subject (Hely-Hutchinson, 1779).37 The English authorities thought Hely-Hutchinson’s
book protesting against the prohibition on exporting woollen manufactures from Ireland
so insidious that it became the last book in the United Kingdom to be publicly burned by
the hangman.
Joshua Gee was a contributor to the journal The British Merchant which opposed a com
mercial treaty that would have established free trade with France. The polemical articles
from this journal were published in 1721 in three volumes as The British Merchant; or,
Commerce preserv’d, with Charles King as the author/compiler, and became another best
seller (King, 1721).
Together with Charles King and John Cary (Cary, 1695: see also S. Reinert, 2011), Joshua
Gee’s volume probably scores higher than any other book on this list on what we could
call the ‘fame to oblivion axis’: compared to their popularity at the time these volumes
seem to be the least remembered today. These were the three authors who probably were
the most honest in explaining the policies that were actually carried out by the English.
They show, without modesty, that the industrial policy of colonialism was preventing man
ufacturing from taking place in the colonies.
Page 28 of 53
guments
This was, of course, an important reason for the United States to wish independence from
England, an event which caused another economics bestseller—that of Alexander Hamil
ton (1791)—to clearly spell out the reasons why the United States would not be a wealthy
country without a manufacturing industry.38 (p. 535)
19.8 The Marshall Plan and the Need to Redis

cover the Economics of Industry, Trade, and
Population Density
There is a phase of this matter which is both interesting and serious. The farmer
has always produced the foodstuffs to exchange with the city dweller for the other
necessities of life.
This division of labor is the basis of modern civilization. At the present time it is
threatened with breakdown.
George Marshall, announcing the future Marshall Plan, Harvard University, 5 June
1947 (emphasis added)
Curiously the immediate post-Second World War era saw two contradictory types of eco
nomic theory grow simultaneously. On the one hand, with the 1947 Marshall Plan, we
saw, at the practical level, a repeat of the principles that this chapter has traced back to
England in the 1400s: the only way to create widespread national wealth is through in
dustrialization. Indeed, the Marshall Plan re-industrialized Europe, but also created a
‘sanitary belt’ of wealthy industrial countries around the communist block from Norway
in the North-west via Italy, Greece, and Turkey to South Korea and Japan in the North-
east. On the other hand, Paul Samuelson’s theoretical papers in The Economic Journal in
1948 and 1949 (cited in section 19.1), based on Ricardian trade theory, argued almost the
exact opposite: whatever you produced, international trade tended to create ‘factor price
equalization’.
The two opposite theories—that manufacturing was needed for wealth creation and that it
was not—lived side by side, but slowly, as the Cold War developed, Samuelson’s theory
got the upper hand over 500 years of experience. The Washington Consensus ideology tri
umphed over the Marshall Plan ideology, which had also been hatched in Washington.
While the Marshall Plan ideology successfully stopped the spread of communism, the
Washington Consensus—especially after the death of communism—in effect put back the
old type of colonial policy: surprisingly, with the thinly veiled excuse of David Ricardo—
which hardly anyone had listened to until the Cold War—as its main tool.
The UN institutions, including UNCTAD and initially UNIDO, defended the old Marshall
Plan order, while the Washington Institutions—the International Monetary Fund (IMF)
and the World Bank (WB)—started basing their recommendations on the Ricardo/Samuel
Page 29 of 53
guments
son theories. As the communist threat waned, a Washington-based type of neo-colonial
ism was initiated, often with disastrous economic results.39
At the very core of the Marshall Plan was a profound understanding of the rela
(p. 536)
tionship between a nation’s economic structure and its carrying capacity in terms of pop
ulation density. We argue that it is necessary to rediscover this theoretical understanding
—which has profound implications for trade and industrial policy—in the mutual interest
of poor and rich countries.
In early 1947, worries grew in Washington that an impoverished Germany, where manu
facturing industry had been forbidden under the Morgenthau Plan, would fall an easy
prey to the Soviet Union. US President Truman therefore sent former president Herbert
Hoover on a fact-finding mission to Germany. One powerful sentence in Hoover’s report
of 18 March zeroed in on the basic problem:
There is the illusion that the New Germany left after the annexations can be re
duced to a ‘pastoral state’ [i.e. a country without industry]. It cannot be done un
less we exterminate or move 25,000,000 out of it.40
Hoover understood that the population density of a country is determined by its economic
structure. Industrialization makes it possible to dramatically increase the population-car
rying capacity of a nation. ‘Exterminate’ was a very strong word to use after the horrors
of the Second World War, and everyone understood that there was no place where 25 mil
lion Germans could be sent. Re-industrialization was the only option.
The lesson from the Marshall Plan is that only extreme danger, in this case a communist
takeover of Germany, will convince the West temporarily to give up what has been called
‘free-trade imperialism’. Temporarily, we argue, two events come together that may en
able a rediscovery of the relationship between the economic structure and population
densities of nations, and consequently benefit Africa.
At the moment—facing a situation similar to that of England after the 1929 crisis—the
United States under Donald Trump is withdrawing from the ideology of free trade. ‘Don
ald Trump can embed a single visceral truth in a welter of falsehoods,’ wrote Rana
Foroohar in The Financial Times in 2018. The ‘visceral truth’ is that David Ricardo’s 1817
trade theory is being marginalized. The last time that Ricardian trade theory collapsed—
in the 1930s—marked the start of a process of industrialization in Latin America that last
ed for decades. We argue that the current situation similarly presents a major opportuni
ty for Africa and other poor countries.
A second event is migration. In 1947, Herbert Hoover stated the facts regarding industri
alization and population density. However, Alfred Marshall, the founder of neoclassical
economics, in his 1890 textbook Principles of Economics, had already provided a frame
work of understanding: activities subject to diminishing returns (agriculture, mining, fish
eries) must after a point shed population, while activities subject to increasing returns at
tract population. As mentioned before, Marshall emphasized the huge impacts of dimin
Page 30 of 53
guments
ishing returns: ‘This tendency to Diminishing Returns was the cause of Abraham’s part
ing from Lot, and of most of the migrations of which (p. 537) history tells.’ This includes
the present migration from Africa, we argue. In an attempt to show us the age of this fun
damental insight—as we have seen in the footnote in section 19.4—Marshall refers to
Genesis xiii: 6: ‘And the land was not able to bear them that they might dwell together;
for their substance was great so they could not dwell together.’41
Alfred Marshall essentially rediscovered what was already old knowledge. All over Eu
rope, development economics of the 1600s and well into the 1700s was dominated by the
insights of Giovanni Botero’s On the Greatness of Cities (1591), a work that appeared in
more than forty editions in all the main European languages. Botero explained why the
only ‘islands’ of wealth in Europe were a few cities, such as Venice, Amsterdam, and Flo
rence, where adding value to raw materials, producing manufactures, was the key to
wealth. In 1613, Antonio Serra42 added the basic theoretical foundation to this: the pro
duction of raw materials was subject to diminishing returns, while manufacturing was
subject to increasing returns. Consequent productivity increases and barriers to entry
made it possible for manufacturing cities simultaneously to raise wages and lower the
cost of their goods.
Centuries of trade policy followed the principles of Botero and Serra all over Europe and
in the United States. Former World Bank Chief Economist Justin Yifu Lin put it very suc
cinctly: ‘Except for a few oil-exporting countries, no countries have ever gotten rich with
out industrialization first.’43
In line with this analysis, we suggest it is time for Africa and poor countries elsewhere to
follow Alfred Marshall’s recommendation: ‘One simple plan would be the levying of a tax
by the community on their own incomes, or on the production of those goods which obey
the Law of Diminishing Returns, and devoting the tax to a bounty on the production of
those goods with regard to which the Law of Increasing Returns acts sharply.’44 Here
Marshall describes what all presently wealthy countries have done, mostly through the
protection of increasing returns activities through tariffs, ever since England in the 1400s
started to tax the export of raw wool, while at the same time subsidizing the local produc
tion of woollen cloth. This was the essence of import-substitution industrialization that
took some non-Western countries out of economic colonialism. For centuries, colonies
were essentially areas where the production of most industrial products was prohibited,
as in the United States until 1776.
The United States under Donald Trump is now ideologically and indirectly paving the way
for the industrialization of Africa. This must be an industrialization not primarily focused
on the nation state, as Latin America’s industrialization was. Nor can it be based primari
ly on supplying global markets, as East Asia’s industrialization was. It must be focused on
the African continent, producing industrial goods that rich countries take for granted, but
whose production has not reached Africa to any extent.
Page 31 of 53
guments
An unintended consequence of the apartheid boycott of Zimbabwe was the rapid growth
of the country’s industrial sector, which reached more than 30 per cent of (p. 538) GDP.
Recently, de-industrialization there has rapidly increased outward migration, proving the
principle Herbert Hoover explained in 1947 still true.
A consequence of Africa’s industrialization could very well be reduced migration because

more Africans will be able to find jobs in Africa. In 1947, a possible communist takeover
of Germany was a sufficient threat for the West to temporarily abandon free-trade imperi
alism. The threat now is the 821 million people worldwide who were undernourished in
2017, according to the FAO. Migration cannot solve their problems. Industrialization can.
We can only hope the West sees the light after the 2020 Coronavirus crisis as it did in
1947.
19.9 ‘Institutions’: A Failed Attempt to Reverse

the Arrows of Causality of Economic Develop
ment
Early economic theory saw strong arrows of causality between the modes of production of
a society and its social and economic structures. Arab economist Ibn Khaldun (1332–
1406) concluded that ‘the differences between different peoples arise out of the differ
ences in their occupations’.45 Francis Bacon, a scholar who was heavily influenced by Gio
vanni Botero,46 wrote in his Novum Organum (1620) that ‘there is a startling difference
between the life of men in the most civilized province of Europe, and in the wildest and
most barbarous districts of New India. This difference comes not from the soil, not from
climate, not from race, but from the arts.’ The same point is strongly emphasized by Ger
man economist Karl Bücher (1847–1930) in his bestseller Die Entstehung der Volk
swirtschaft (1893) in the English translations rendered as Industrial Evolution.
This way of thinking represents an attempt to systematize the understanding of different

categories of human societies—much in the same way as Linnaeus did with plants—free
ing economics from what Nobel Laureate James Buchanan called the equality assumption
in economic theory. In 1721 English economist Charles King made a classification of in
ternational trade in the same spirit as the quality index presented in this chapter. Import
ing manufactured goods and exporting raw material was ‘bad trade’ for a country, while
importing raw materials and exporting manufactured goods was ‘good trade’. Interesting
ly, exchanging manufactures for other manufactures was considered ‘good trade’ for a na
tion. The principle expressed by King was based on the same observations as those of
Giovanni Botero (1589) and Antonio Serra (1613). UNCTAD’s idea of symmetrical trade
as being good for all trading partners 252 years later recalls King’s ideas.
The craving for taxonomies also created the so-called stage theories,47 to which
(p. 539)
Adam Smith also subscribed. In German these theories have recently been referred to as
Page 32 of 53
guments
Wirtschaftsstile,48 which implicitly emphasizes the fact that different economic styles—or
stages—may coexist at the same time in different places.
At the core of stage theories is that the mode of production—that is, whether you are in
the Stone Age or the computer age—will determine your institutional structure. In this
stage tradition, the structure of production tends to influence the institutional structure
more than the other way around.49
On a personal note, having worked with pastoralists in the high Andes and with Saami
reindeer herders in Northern Fennoscandia, I can testify to the striking similarities in the
social organization of pastoralists under extreme climatic conditions in such different ar
eas and cultures. Here, as under other extreme climatic conditions, the market economy
has not penetrated (other than in the extraction of minerals). These societies have se
quential usufruct of land over the years, not private property.
Clearly some institutional innovations are crucial to economic development. Primogeni

ture—the right of the firstborn legitimate son (or child) to inherit his parents’ entire es
tate—has created stability in European kingdoms compared, for example, to the Arab
world. In agriculture, primogeniture prevented farm sizes from diminishing into or be
yond self-sufficiency. Werner Sombart, the great historian of capitalism, sees the birth of
two institutions—double-entry book-keeping and bankruptcy—as the two key institutional
ingredients making the system possible. Historically these institutions bring us back to
Venice in the twelfth and thirteenth centuries.50
However, it is generally most useful to see institutions being born out of the mode of pro
duction itself, as in the quotes from Ibn Khaldun and Francis Bacon above. With their
2012 book Why Nations Fail: The Origins of Power, Prosperity, and Poverty 51 Daron Ace
moğlu and James Robinson in practice come to the defence and salvation of neoclassical
trade theory by blaming former European colonies for not ‘getting the institutions right’.
They seem to disregard the key point that the ‘extractive institutions’ they blame for the
lack of development represent the very essence of Western colonialism. When explaining
that ‘North America became more prosperous [than Peru and Mexico] precisely because
it enthusiastically adopted the technologies and the advances of the Industrial Revolu
tion’ (2012: 53), Acemoğlu and Robinson leave out that Peru and Mexico for a long time
were colonies, and that a key element in colonial policies was precisely to prohibit manu
facturing there (see Gee, 1729). When Peru and Mexico later gained formal indepen
dence, they were still de facto colonies, as power just shifted from Spaniards in Spain to
Spaniards residing locally,52 with the same vested interests (p. 540) in exporting raw mate
rials.53 In this way Acemoğlu and Robinson, far from attacking the colonial policies we
have emphasized in this chapter, appear to be blaming the victims of colonialism for their
own poverty.
I would argue, in the Bücher tradition, that the Venetians did not invent an official prop
erty register (catasto), around 1150, later to create capitalism, but rather because the
capitalist growth of the city created a need for the property register. The problem ap
peared before the solution. Likewise, one could argue that the Venetians did not invent in
Page 33 of 53
guments
surance, and then—based on this—start long-distance trading. Rather, one would argue
that the previous system of spreading risk through ever smaller percentages of owner
ship of ships and cargo became impracticable because of the many owners, and that the
impracticability of this fractionalized ownership is the origin of a system of insurance: in
stead of diluting ownership, risk was externalized. In practice, economic activities and
their institutions co-evolve, and the first geographical area where this process of co-evo
lution created capitalism was in the twelfth-century Italian city-states.54
In short, historical observations tend to reverse the arrows of causality in economic devel
opment compared to the formal theorizing of modern institutional economics (as opposed
to the classical institutional economic of Thorstein Veblen and his contemporaries).
19.10 When Industrial Policy Intuition Clashes

with Ricardian Trade Theory
Economists are unlikely to say to their children: ‘My son (or my daughter), I have ob
served your efficiency in washing the dishes. It is clear to me that you have a comparative
advantage in this activity, and I would recommend a career washing dishes in restau
rants.’ As a parent, the economist would react according to the pre-Ricardian logic we
have described in this chapter. As an economist—advising the children of Africa—their ad
vice would be based on David Ricardo.
This is a clear example of what US economist Thorsten Veblen warned against: that for
mal education might contaminate healthy human instincts (Reinert and Viano, 2012). Exo
teric knowledge—practical and intuitive knowledge—could with higher education be lost
to much more prestigious but of little practical use—esoteric (p. 541) knowledge, such as
Ricardian trade theory. As already referred to, Schumpeter referred to this as ‘the Ricar
dian Vice’; bringing the theory to such a high level of abstraction that it became irrele
vant.
It is also intuitively obvious that, in spite of the theories of David Ricardo, free trade be
tween a Stone Age society and a computer-age society will not tend to produce factor
price equalization. A similar intuitive reasoning made presidential candidate Bill Clinton
in 1992 advocate high-tech industries. The reply from George H. W. Bush’s economic ad
viser Michael Boskin came straight out of the neoclassical economic textbook: ‘computer
chips, potato chips, what’s the difference?’ With time, Boskin seems to be on the losing
side when it comes to US industrial policy, but not when it comes to US trade policy to
wards Africa.55
Page 34 of 53
guments
19.11 Nichification: An ‘Industrial’ Policy

Strategy for Agriculture
The twentieth century was dominated by standardized mass production. Henry Ford’s
statement in 1909 that ‘any customer can have a car painted any color that he wants so
long as it is black’ expresses the need to standardize in order to keep costs down. Gradu
ally, and especially with the introduction of information technology, it was possible to pro
duce smaller runs. The need for standardization diminished.
In agricultural production, more so in Northern than in Southern Europe, standardization

increased as a by-product of the economic crisis of the 1930s. Some agricultural econo
mists claim, probably correctly, that agriculture is the first economic activity to enter into
an economic crisis and the last to leave it. Due both to market power and to strong
unions, during the crisis of the 1930s the industrial workers who kept their jobs tended to
keep their wages. The crisis had a completely different effect on agriculture: farmers’
sales prices and their incomes fell precipitously. John Steinbeck’s 1939 The Grapes of
Wrath captures the drama of the situation.
After the Second World War it was understood that farmers could not produce their way
out of their problems; this would only cause overproduction and falling prices. Agricul
ture was seen as needing more market power, and in that sense agriculture ought to be
more like industry. For this reason, national farmers’ cooperatives were given monopoly
powers, and in the United States agriculture was (and still is) exempt from anti-trust and
often heavily subsidized.
This brought agricultural production, previously locally based, into the logic of
(p. 542)
Fordist mass production. While previously in Europe every farm, or every region or valley,
had its own cheese, cheese production became more and more industrialized and more
and more standardized. This coincided with the rise of big supermarket chains that came
to dominate the retail food market. Farm products became bulk products, and when com
petition slowly opened up the farmers found themselves in the clearly inferior position of
being specialized in bulk products, basically left to compete on price alone. A very ‘bad
specialization’.
In Southern Europe the local and regional pattern survived much longer, and big super
markets also came to dominate later there than in Northern Europe. People wanted their
local cheese and their local salame, so price competition between bulk producers was
much less dominant. The local niche products, and with them decentralized production,
survived.
General de Gaulle once rhetorically asked: ‘How can you govern a country which has 246
varieties of cheese?’ According to a book on Italian cheeses, Italy beats that number by
more than 200 varieties, registering 451 different varieties of cheese.56 Having avoided
Page 35 of 53
guments
the bulk- and mass-production paradigm, French and Italian cheese, as well as some
cheeses from Spain and Switzerland, became a ‘smart specialization’.
The organizational principle of Fordist mass production in bulk was economies of scale in
hierarchies, while ‘smart specialization’ depended on economies of scope among small
players in networks. Competition here is based on quality and product differentiation, not
on price as in the mass-production paradigm.
In agriculture and food production there is today ever-increasing diversity, more so in

Southern Europe than in Northern Europe and the United States. Italy has of course hun
dreds of different types of pasta, and this diversity multiplies because regional differ
ences between pasta types—often with the same names—are enormous. The casoncelli of
Lombardy—a kind of ravioli—are very different in Cremona from those in Bergamo or
Brescia. In many ways this Italian diversity is a remnant of pre-Fordism. More than most
countries, France and Italy have managed to preserve variety in food and agriculture,
while at the same time utilizing the industrial economies of scale. At the other extreme of
the scale, Norway, with only about 5 million people, was probably the country where
Fordist mass production—killing off previous niches—most came to dominate agricultural
production, both meat and milk.
The development and importance of diversity is illustrated by figures from modern biolog
ical research. Figure 19.10 is from Harvard biologist Stephen Jay Gould’s (1996) Full
House: The Spread of Excellence from Plato to Darwin. The illustration shows the evolu
tion of the diversity of biological species from a common ‘ancestor’. In the case of horses,
it would be a kind of Urpferd or Sifrhippus. Each end point further to the right represents
a new biological variety descending from the same ‘ancestor’ (to the left in the drawing),
like Shetland ponies, Peruvian paso horses, (p. 543) (p. 544) zebras, and donkeys. In
Gould’s scheme a small number of varieties grow much larger than the rest as a result of
random evolution. This is represented by the larger varieties at the bottom of the time ax
is (the varieties to the right, seen from the point of view of the ‘ancestor’).
Page 36 of 53
guments
Figure 19.10 Increasing diversity and specialization

over time (= ‘tid’)
Source: Gould (1996).
If we transfer this illustration to economic diversity, we create a graph that corresponds

to Botero’s idea of increasing value added the further one moves away from the raw ma
terial. In this graph the end points represent a product. For example, let this common ‘an
cestor’ be milk (the single starting point to the left). As the biological ‘ancestor’, the start
ing point is generic and non-specialized. Milk can come from a variety of animals, from
cows to sheep, reindeer, and moose. The first more specialized branch could be the prod
uct cheese. The product cheese is again divided into new and ever more specialized prod
ucts as we move towards the right of the time axis. Other products could be yoghurt, but
termilk, whole milk, cream, sour cream, and so on. Far out to the right on the diversity
tree of cows’ milk, we find, for example, Appenzell cheese, which is only produced in two
small cantons in Switzerland, or—as an extreme example—650 Parmesan cheeses coming
from 650 different cheese factories which all produce technically slightly different
cheeses. (On the biological axis far out to the right we find the koala, which is so special
ized that it only eats the leaves from one specific kind of eucalyptus.)
Wine is an example of extreme nichification. If we look at Gould’s starting point at the left
(bottom) of Figure 19.10, the single starting point would be that by fermenting grapes
you can produce wine. If we add that there are green and red grapes, and that red grapes
may be left with the skin for a while to create rosé wine, you have the next stage of diver
sification in Gould’s graph: white, red, and rosé wine. Then, further to the right, a huge
variety of grapes and climates produce a never-ending variety of niche wines. These nich
es—from Barolo in Piedmont to Zinfandel in California—make it possible to compete
along other aspects than price: more value is added as in Botero’s theory. The wine indus
try was the first to use terroir—clusters of environmental factors affecting quality—as a
marketing tool. Reportedly, the first such geographical protection was established in 1716
by Cosimo III de’Medici, the Grand Duke of Tuscany, for the Chianti wine.
Page 37 of 53
guments
With the end of Fordist mass production and the introduction of information technology,
the potential for decentralization increased: on Gould’s axis many production processes
moved towards the right, towards a far greater diversity. The possibilities not only vary
from industry to industry, but also from product to product. In the last instance it is also
the human will—no invisible hand—deciding to what extent the decentralizing element in
the present economic paradigm shift should be used to strengthen the economic periph
ery. Also, in the new organizational paradigm we have large industries—like Boeing and
Microsoft in Seattle—representing the larger varieties at the bottom right of the time ax
is. When it comes to both large and small industries, it is the increasing amount of human
knowledge that advances the process. (p. 545) One of Gould’s main points in the book is
that over time the small units, in spite of the many visible large units (read ‘firms’), domi
nate ever more. We see the same development in the economy during the transition from
the Fordist to the future techno-economic paradigm. Gould’s second important point from
this worldview is that to utilize average values becomes more and more meaningless as
development advances.
In the economic world there are different degrees of demand for the original generic
product (the ‘ancestor’ and the basis for the illustration)—commodities, for example
generic ‘milk’. It is only natural that different business strategies make some firms spe
cialize in production of the generic product, where the demand is for low prices rather
than high quality. Here the margins are very small, and this strategy needs an enormous
turnover (and/or low wage rates) to survive (a result of economies of scale). Here we find
giants like Cargill in the world grain markets. It is worth noting that the strategy in this
volume market essentially implies a fight for market shares because high volume = low
unit costs.
Emilia-Romagna in Italy is an interesting area from the point of view of nichification. In

Emilia-Romagna the high-volume/low-cost strategy was represented by the production of
ultra-pasteurized milk by the giant firm Parmalat, which built on the importance of global
ization and economies of scale in this market by, for example, buying up thirty-six dairies
on the East Coast of South America. However, at the time operating in more than thirty
countries, Parmalat came close to bankruptcy in the midst of a financial scandal.
Page 38 of 53
guments
The high-volume/low-cost strategy bulk production failed Emilia-Romagna’s agriculture.
What makes Emilia-Romagna agriculture so special is the fact that in many agricultural
products—milk, ham, vinegar, olive oil—local raw materials are used. Producers in this re
gion receive higher prices than the producers of the same raw materials do in the rest of
Italy. The explanation is that Emilia-Romagna delivers very high-quality niche products
that we find far to the right in Figure 19.10. Industrial giant Parmalat mass-produced its
standard-quality milk based on milk imported from Bavaria in Germany. When this author
researched this issue in 1996, the producers delivering milk for Parmesan cheese
achieved 40 per cent higher prices than did the producers of normal consumer milk in
nearby regions. When it comes to milk production this region has managed to get the
best of all worlds:
1) High prices for local raw materials for niche products, higher prices than for the
same products in many parts of Europe.
2) A decentralized production of niche products utilizes the partly rugged geographi
cal territory in the Apennines very well (1996: 650 dairies producing milk for Parme
san cheese).
3) To the extent this still lasts, economies of scale in high-tech mass production of
bulk milk based on the import of cheap milk imported from Germany (Parmalat etc.).
(p. 546)
19.12 What’s New in the Present Industrial Pol

icy Game?
As regards industrial policy, the US Republican Party platform in 1884 sounded as if it
had been based on the writings of Giovanni Botero almost 300 years earlier:
The largest diversity of industry is most productive of general prosperity, and of

the comfort and independence of the people. We, therefore, demand that the im
position of duties on foreign imports shall be made, not ‘for revenue only’, but that
in raising the requisite revenues for the government, such duties shall be so levied
as to afford security to our diversified industries and protection to the rights and
wages of the laborer; to the end that active and intelligent labor, as well as capi
tal, may have its just reward, and the laboring man his full share in the national
prosperity.
Against the so-called economic system of the Democratic party, which would de
grade our labor to the foreign standard, we enter our earnest protest.57
Policies to this effect could be achieved by allowing all raw materials free of duty into the
country, and by increasing the customs duty in proportion to the added value to the raw
materials. The principle of industrial policy was to apply the highest duty to the techno
logically most sophisticated product.
Page 39 of 53
guments
Since then, many developments have complicated the issue. The minimum efficient size of
industrial firms—and therefore also of nations—has increased considerably.58 In the
1930s a country like Estonia, at the time richer than Finland, with a population of only 1
million people, could have internal competition in many industries. This is no longer pos
sible. The 1988 Cecchini Report59 arguing for a single market in Europe had a good point
when predicting that most of the benefits from the single market would occur because of
increasing returns to scale in manufacturing industry. What Cecchini did not calculate
was that, particularly with the help of a frozen exchange rate in the form of the euro,
manufacturing industry would tend to die out in the countries in the European periphery,
thus creating a win‒lose situation where Germany in the end appears to be the biggest,
or sole, winner.
From a perceived hierarchy of increased value added based on raw materials, the appear
ance of global value chains has complicated the issue. Often global value chains can be
analysed using the quality index in this chapter, but sometimes the categories, such as
high-tech goods, contain a large diversity of products of different levels of sophistication.
The trade tensions between the United States and China illustrate the issue. One thing is
clear, though, that by putting high tariffs on important products that are already subsi
dized—like soy beans—China hit a weak point in US trade policy. (p. 547)
19.13 Conclusion: Unrealistic Utopias that

Boomerang as the Curse of Europe and the
West
The naive optimism of ‘laissez-faire’ and the childish and frivolous appeal to revo
lution, the naive hope that the tyranny of the proletariat would lead to world hap
piness, increasingly showed their real nature, they were twins of an ahistorical ra
tionalism…The period 1870–1890 led to the theoretical and practical bankruptcy
of both the old schools.
Gustav Schmoller, German economist, Inaugural speech as Rector of the Universi

ty of Berlin, 1897
Again we appear to stand at the intellectual bankruptcy of a school of economics, of a big

vision of how the world was supposed to work. This time it is neo-liberalism, the modern
version of Manchester liberalism, one of the two ahistorical twins Schmoller refers to
above. Since one irrational twin, communism, died in 1989, in the ideological triumphal
ism that followed we allowed the other twin, neo-liberalism, to rule the field virtually
alone for thirty years. One curious aspect of this is that in the balance of countervailing
power that had existed between capitalism and communism the industrial policy of
Friedrich List had ruled on both sides: West Germany (Deutsche Bundesrepublik) and
East Germany (Deutsche Demokratische Republik) both issued stamps with the same por
trait of Friedrich List. List’s 1841 textbook on industrialism had been ruling on both
Page 40 of 53
guments
sides. The disappearance of this ideology was probably the most important defect when
neo-liberalism went from theory to practical policy following the 1989 fall of the Berlin
Wall.
Austrian-Swiss economist Felix Somary (1881–1956) made the perceptive observation

that all big universalist projects of Europe have boomeranged, and caused the opposite of
what was intended:60
1. The Crusades to the Holy Land should bring the infidels to Christianity and
strengthen the union of the Catholic Church. Instead the Crusades led to the fall of
Constantinople—of the Eastern Roman Empire and its Church—to the Muslims.
2. The basic rule of religious tolerance under the Reformation—cuius regio, eius reli
gio (‘whose realm, his religion’)—meaning that the religion of the ruler was to dictate
the religion of the country, suffered the indignity of centuries of devastating religious
wars and more intolerance (including anti-Semitism). After the 1648 Peace of West
phalia—which brought an end to this period—there were around 400 small states in
Germany alone. (p. 548)
3. A little more than 100 years later the French bourgeoisie started its revolution for
political and economic freedom, which led to a bloodbath, despotism, and four gener
ations of dictatorship. Adam Smith had taught us that: ‘It is not from the benevo
lence of the…baker that we expect our dinner, but from [his] regard to [his] own self-
interest.’ However, what Smith had not envisioned was that in the decades before
the French Revolution much more money could be made from withholding flour and
grain from the market—causing prices to rise—than by baking bread.61 To most econ
omistes of the time—Physiocrats—it was inconceivable that money made from pro
ducing goods and services could have different effects in the economy than money
made from increasing the prices of things already produced (i.e. from speculation).62
French minister of finance Jacques Necker (1732‒1804), born in the Republic of Geneva,
was in ardent opposition to the Physiocratic laissez-faire doctrine which was mainly re
sponsible for the shortage of bread in Paris. When the French Revolution broke out with
the storming of the Bastille on 14 July 1789, the fact that Necker had been dismissed as
minister of finance three days earlier was a main reason for these disturbances. Necker’s
popularity as an economist is proven by the fact that he is the only author—of more than
eighty—represented with four different works on the list of economics bestsellers before
1850.63
Today we can observe—in the spirit of Somary—that the lofty ideals of the European
Union project resulted in countries divided by a common currency with serious economic
problems in the periphery; that the United States is faced with falling real wages, a dwin
dling middle class, and increased mortality of white males; and that the vision of global
ization as an exercise in increasing harmony—propelled by the Ricardian Cold War vi
sions of Paul Samuelson—often proves to be more an exercise in factor price polarization
than in factor price equalization (with China as the great exception).
Page 41 of 53
guments
During the French Revolution, the supposedly ‘natural’ forces of automatic harmony, from
Franz Anton Mesmer’s quack science of Mesmerism,64 and its supposed creation of a So
ciety of Universal Harmony, to the Physiocrats’ idea of laissez-faire provided the opposite:
famine and revolution. This shows that the nature of economic systems obeys Charles
Darwin more than it obeys any unrealistically abstract theories creating illusions of auto
matic harmony: economic harmony is man-made. A point which has become very clear
now, when decades of neo-liberal policies have created (p. 549) massive problems of in
equality. We therefore need meso-level theories adapted to the specific contexts of each
nation. From this point of view neoclassical economics and neo-liberal theory have func
tioned as snake oil, a cure-all for most ailments. Googling the term ‘austerity’ together
with ‘snake oil’ in fact gives a surprisingly large number of hits.
The wealth and poverty of a country are to such a degree a result of its economic struc
ture, that industrial policy, in its widest sense, is necessary in order to create peace and
economic justice. The present ecological crises only reinforce the need for the under
standing of the mechanisms of diminishing returns (which created Malthus’ ‘dismal sci
ence’) and increasing returns under which it is possible to harvest a generous nature
rather than to unrenewably extract the products of nature.65 Utopian visions of automatic
harmony resulting from free trade are main drivers of human migration away from un-in
dustrialized or de-industrialized countries in Africa and Latin America to Europe and
North America.
In 1997, the WTO director-general, Renato Ruggiero, declared—in the spirit of Paul
Samuelson’s interpretation of David Ricardo—that we should unleash ‘the borderless
economy’s potential to equalize relations among countries and regions’.66 This is the
foundation upon which the whole world economic order came to rest, and also the ideo
logical marching order for the Washington Institutions—the International Monetary Fund
and the World Bank—in their policies towards the poor countries of the world. It is again
important to emphasize that, in terms of industrial policy, this theoretical understanding
was exactly the opposite of the theory on which the extremely successful 1947 Marshall
Plan had been built.
‘The worse the situation, the less laissez-faire works,’ said Keynes. Historical develop
ments proved this to be true: the negative effects of laissez-faire first showed up in the
Third World, then in the Second (former communist) World, and in the end in the core of
the First World, in the United States and the European Union (Reinert, 2012).
One important side-effect of the policies of the Washington Institutions is that dictator
ships obtain a considerable advantage over countries which try to be democratic. China
and Singapore are doing extremely well under their system of well-managed free trade—
opening up to free trade where it suits them as Europe used to do—while the policies of
the Washington Institutions de-industrialize and impoverish many democratic countries
(Figure 19.11). The virtual ‘prohibition’ of industrial policy through the conditionalities of
the Washington Institutions has, in many places, been very costly in terms of failed devel
opment and economic retrogression.67
Page 42 of 53
guments
In 1989, real wages in Ukraine were considerably higher than those in Belarus. The rea
son real wages in Belarus now are about double those in Ukraine is not because (p. 550)
the statue of Lenin still stands on the main square in Minsk, it is probably not that Be
larus is less corrupt than its neighbours, it is mainly because, as a dictatorship, the coun
try does not have to follow the foreign dictates of the Washington Institutions and can ac
tually pursue a national industrial policy.
We are in a period when the attitude towards industrial policy is slowly changing, but in
the reverse order of what should happen if we follow Keynes’ insight above. The clearest
changes in favour of industrial policy are taking place in Germany and in the United
States. The process is much slower in poor countries where there has been a prohibition
of industrial policy and where it is most needed. In 2018, asked to contribute to an annual
report on the development of the OECD, I was able to observe how alternative ideas are
only extremely slowly filtering into this powerful global institution. It is difficult for a
whole generation of experts to admit that they were wrong, and the virtual monopoly of
neoclassical economics at the university level makes it difficult to recruit professionals
with alternative views. In a sense the world faces the same kind of intellectual monocul
ture that faced the universities in former East Germany at the 1990 German unification:
there were twenty-three professors of Marxist economics at the universities in former
East Germany, and little else.
Figure 19.11 Economic growth since 1989 (fall of

the Berlin Wall), selected countries: percentiles of
population with income growth above/below the
1989 level/the G7 average level
Source: European Bank for Construction and Devel

opment, Transition Report 2016–2017, London, 2016.
References
Acemoğlu, Daron and James Robinson (2012) Why Nations Fail: The Origins of Power,
Prosperity, and Poverty. London: Profile Books.
Baade, Fritz (1955) ‘Gruß und Dank an Herbert Hoover’, Weltwirtschaftliches Archiv
74(1): 1–6.
Page 43 of 53
guments
Bacon, Francis (1620) Novum Organum. London: J. Billum.
Bacon, Francis (1625) ‘An Essay on Innovation’, in The essayes or counsels, civill and
morall, of Francis Lo. Verulam, Viscount St. Alban. Newly written. London: printed by
John Haviland for Hanna Barret.
Bollinger, Armin (1970) Widersprüche zwischen Unabhängigkeit und Dekolonisation in

Lateinamerika. Zürich: Orell Füssli.
Boston Consulting Group (1972) Perspectives on Experience. Boston, MA: Boston Consult
ing Group.
Botero, Giovanni (1589) Della Ragion di Stato: Libri dieci, con Tre Libri delle Cause della
Grandezza, e Magnificenza delle Città. Venice: Appresso i Gioliti.
Botero, Giovanni (1622) Relazioni Universali. Venice: Alessandro Vecchi.
Botero, Giovanni (1635) The Cause of the Greatnesse of Cities. Three Bookes, With Cer
taine Observations concerning the Sea. Written in Italian by John Botero…London: Print
ed by E.P. for Henry Seile.
Bücher, Karl (1893) Die Entstehung der Volkswirtschaft. Tübingen: Laupp. (English trans
lation: Industrial Evolution 1901. London: G. Bell & Sons. New York: H. Holt.)
Cary, John (1695) An Essay on the State of England. Bristol: W. Bonny.
Cecchini, Paulo (1988) The European Challenge 1992: The Benefits of a Single Market.
Aldershot: Wildwood House.
Childs, Marquis William (1936) Sweden: The Middle Way. New Haven, CT: Yale University
Press.
Darnton, Robert (1968) Mesmerism and the End of Enlightenment in France. Cambridge,
Defoe, Daniel (1728) Plan of English Commerce. London: Rivington.
di Corato, Riccardo (1977) 451 formaggi d’Italia. Catalogo storico-gastronomico delle va
rietà regionali. Milan: Sonzogno.
Fabricant, Solomon (1942) Employment in Manufacturing, 1899–1939. New York: Nation

al Bureau of Economic Research.
Firpo, Luigi (1971) ‘Botero, Giovanni’, Dizionario Biografico degli Italiani, 13. Available at
http://www.treccani.it/enciclopedia/giovanni-botero_(Dizionario-Biografico).
Firpo, Luigi (1975) ‘La “Ragion di Stato” di Giovanni Botero: redazione, rifacimenti, fortu
na’, in Gianrenzo Clivio and Riccardo Massano (eds) Civiltà del Piemonte. Studi in onore
Page 44 of 53
guments
di Renzo Gandalfo nel suo settantacinquesimo compleanno, Vol. I. Turin: Centro Studi
Piemontesi, pp. 139–64.
Gee, Joshua (1729) The Trade and Navigation of Great-Britain Considered… London:
Printed by Sam Buckley.
Goldthwaite, Richard A. (2009) The Economy of Renaissance Florence. Baltimore, MD:

Johns Hopkins University Press.
Gould, Stephen Jay (1996) Full House: The Spread of Excellence from Plato to Darwin.
New York: Harmony Books. (p. 552)
Graham, Frank (1923) ‘Some Aspects of Protection Further Considered’, Quarterly Jour
nal of Economics 1(37): 199–227.
Hamilton, Alexander (1791) Report of the Secretary of the Treasury of the United States
on the subject of manufactures: Presented to the House of Representatives, December 5,
1791. Philadelphia, PA: Childs and Swaine.
Hausmann, Ricardo, César Hidalgo et al. (2011) The Atlas of Economic Complexity.
Cambridge, MA: Center for International Development, Harvard University, Harvard
Kennedy School.
Hely-Hutchison, John (1779) The Commercial Restraints of Ireland considered in a series

of letters addressed to a Noble Lord, containing an historical account of the affairs of that
Kingdom, so far as they relate to this subject. Dublin: William Hallhead. For a reproduc
tion of a second edition: Dublin: M. H. Gill & Son, 1882. Available at http://
www.gutenberg.org/files/38841/38841-h/38841-h.htm.
Hobson, John A. (1902) Imperialism: A Study. London: Nisbet.
Hume, David (1767) History of England, Vol. 3. London: Cadell.
Issawi, Charles (1987) An Arab Philosophy of History: Selections from the Prolegomena of
Ibn Khaldun of Tunis (1332–1406). Princeton, NJ: Darwin Press.
Johnson, Donald Bruce and Kirk H. Porter (eds) (1973) National Party Platforms 1840–
1972. Urbana, IL: University of Illinois Press, pp. 72–3.
Kaplan, Steven (2015) Bread, Politics and Political Economy in the Reign of Louis XV. 2nd
edition. London: Anthem.
King, Charles (1721) The British Merchant; or, Commerce Preserv’d. 3 vols. London: John
Darby.
Krugman, Paul (1981) ‘Trade, Accumulation and Uneven Development’, Journal of Devel
opment Economics 8: 149–61.
Krugman, Paul (1990) Rethinking International Trade. Cambridge, MA: MIT Press.
Page 45 of 53
guments
Lawson, Tony (1989) ‘Abstraction, Tendencies and Stylised Facts: A Realist Approach to
Economic Analysis’, in Tony Lawson, J. Gabriel Palma, and John Sender (eds) Kaldor’s Po
litical Economy. London: Academic Press.
Lenin, Vladimir (1939) Imperialism, the Highest Stage of Capitalism. New York: Interna
tional Publishers.
Lin, Justin Yifu (2012) New Structural Economics: A Framework for Rethinking Develop
ment and Policy. Washington, DC: World Bank.
Lösch, August. 1954. The Economics of Location: A Pioneer Book in the Relations be
tween Economic Goods and Geography. New Haven, CT: Princeton University Press.
Marshall, Alfred (1890) Principles of Economics. London: Macmillan.
Marshall, Alfred, and Mary Paley Marshall (1879) The Economics of Industry. London:
Macmillan.
Mathews, John and Erik Reinert (2014) ‘Renewables, Manufacturing and Green Growth:
An Energy Strategy Based on Capturing Increasing Returns’, Futures 61: 13–22.
Meek, Ronald (1976) Social Science and the Ignoble Savage. Cambridge: Cambridge Uni
versity Press.
Meinecke, Friedrich (1925) Die Idee der Staatsräson in der neueren Geschichte. Berlin:
Oldenbourg.
Mervin, Leonard (1981) Vitamin C. Willingborough: Thorsons Publishers.
Mill, John Stuart (1848) Principles of Political Economy. London: Parker.
Misselden, Edward (1623) The Circle of Commerce or the Balance of Trade. London:
Dawson for Nicholas Bourne. (p. 553)
Mueller, Reinhold and Frederic Lane (1997) The Venetian Money Market: Banks, Panics,
and the Public Debt: 1200–1500. Baltimore, MD: Johns Hopkins University Press.
Müller-Armack, Alfred (1947) Wirtschaftslenkung und Marktwirtschaft. Hamburg: Verlag

für Wirtschaft und Sozialpolitik.
Mun, Thomas (1664) England’s Treasure by Forraign Trade. London: Thomas Clark.
Oresme, Nicolas (1356) De Moneta. Manuscript reprinted with Schefold, Bertram (1995)
Vademecum zu einem Klassiker der Mittelaterlichen Geldlehre. Düsseldorf: Verlag
Wirtschaft und Finanzen.
Ortiz de Zúñiga, Iñigo (1967) Visita de la provincia de León de Huánuco en 1565,

Huánuco, Perú: Universidad Nacional Hermilio Valdizán, Facultad de Letras y Educación.
Page 46 of 53
guments
Patalano, Rosario and Sophus Reinert (eds) (2016) Antonio Serra and the Economics of
Good Government. London: Palgrave.
Perroux, François (1955) ‘Note sur la notion de pôle de croissance’, Economie Appliquée
8, pp. 307–20.
Pierce, Charles S. (1867) in Collected Papers of Charles Sanders Pierce, Vol. V, ed. C.
Hartshorne and P. Weiss. Cambridge, MA: Harvard University Press.
Porter, Michael (1990) The Competitive Advantage of Nations, New York: Free Press.
Reinert, Erik S. (1980) ‘International Trade and the Economic Mechanisms of Underde
velopment’, PhD thesis, Cornell University.
Reinert, Erik S. (1994) ‘Catching-up from Way Behind: A Third World Perspective on First
World History’, in Jan Fagerberg, Bart Verspagen, and Nick von Tunzelmann (eds) The
Dynamics of Technology, Trade, and Growth. Aldershot: Edward Elgar, pp. 168–97.
Reinert, Erik S. (ed.) (2004) Globalization, Economic Development and Inequality: An Al
ternative Perspective. Cheltenham: Edward Elgar.
Reinert, Erik S. (2005) ‘A Brief Introduction to Veit Ludwig von Seckendorff (1626–1692)’,
European Journal of Law and Economics 19: 221–30.
London: Constable.
Reinert, Erik S. (2009a) ‘Emulating Success: Contemporary Views of the Dutch Economy
before 1800’, in Oscar Gelderblom (ed.) The Political Economy of the Dutch Republic.
Aldershot: Ashgate, pp. 19–40.
Reinert, Erik S. (2009b) ‘Emulation vs. Comparative Advantage: Competing and Comple
mentary Principles in the History of Economic Policy’, in Mario Cimoli, Giovanni Dosi, and
Joseph E. Stiglitz (eds) Industrial Policy and Development: The Political Economy of Capa
bilities Accumulation. Oxford: Oxford University Press, pp. 79–106.
Reinert, Erik S. (2012) ‘Neo-classical Economics: A Trail of Economic Destruction since

the 1970s’, Real-world Economics Review 60: 2–17. Available at http://www.paecon.net/
PAEReview/issue60/Reinert60.pdf.
Reinert, Erik S. (2018) ‘Smart Specialization: Theory and Brief Case Studies’. A Report to
the European Commission, Directorate General JRC, JRC Seville. The Other Canon Foun
dation and Tallinn University of Technology Working Papers in Technology Governance
and Economic Dynamics, No. 81. Available at http://technologygovernance.eu/eng/
the_core_faculty/working_papers/.
Page 47 of 53
guments
Reinert, Erik S. and Arno Daastøl (1997) ‘Exploring the Genesis of Economic Innovations:
The Religious Gestalt-switch and the Duty to Invent as Preconditions for Economic
Growth’, European Journal of Law and Economics 4: 233–83.
Reinert, Erik S. and Fernanda Reinert (2018) ‘33 Economic Bestsellers Published before
1750’, The European Journal of the History of Economic Thought 25: 1–58. (p. 554)
Reinert, Erik S. and Francesca Viano (2012) Thorstein Veblen: Economist for an Age of
Crises. London: Anthem.
Reinert, Erik S., Kenneth Carpenter, Fernanda A. Reinert, and Sophus A. Reinert (2017)
‘80 Economic Bestsellers before 1850: A Fresh Look at the History of Economic Thought’.
The Other Canon Foundation and Tallinn University of Technology Working Papers in
Technology Governance and Economic Dynamics No. 74.
Reinert, Sophus A. (ed.) (2011) Antonio Serra, A Short Treatise on the Wealth and Poverty
of Nations (1613). London: Anthem.
Ricardo, David (1817) Principles of Political Economy and Taxation. London: John Murray.
Rubies, Joan-Pau (2000) Travel and Ethnology in the Renaissance. Cambridge: Cambridge
University Press.
Ruggiero, Renato (1997) Speech in Berlin, 23 October. Available at https://www.wto.org/

english/news_e/sprr_e/berlin_e.htm.
Samuelson, Paul A. (1948) ‘International Trade and the Equalisation of Factor Prices’,
Economic Journal 58: 163–84.
Samuelson, Paul A. (1949) ‘International Factor-Price Equalisation Once Again’, Economic

Journal 59: 181–97.
Schefold, Bertram (1994/1995) Wirtschaftsstile. 2 vols. Frankfurt: Fischer.
Schmoller, Gustav (1897) Wechselnde Theorien und feststehende Wahrheiten im Gebiete

der Staats- und Socialwissenschaften und die heutige deutsche Volkswirthschaftslehre:
Rede bei Antritt des Rectorats gehalten in der Aula der Königlichen Friedrich-Wilhelms-
Universität am 15. October 1897. Berlin: W. Büxenstein.
Schumpeter, Joseph (1935) Théorie de l’Evolution Economique. Paris: Dalloz.
Schumpeter, Joseph (1954) History of Economic Analysis. New York: Oxford University
Press.
Seckendorff, Veit Ludwig von (1656) Teutscher Fürsten-Stat (sic). Frankfurt: Götz.
Seligman, Edwin (1920) Curiosities of Early Economic Literature. San Francisco, CA: John
Henry Nash.
Page 48 of 53
guments
Smith, Adam (1776) The Wealth of Nations.
Smith, Adam (1812 [1759]) ‘The Theory of Moral Sentiments’, in Collected Works, Vol. 1.
London: Cadell and Davies.
Somary, Felix (2010 [1952]) Krise und Zukunft der Demokratie. Jena: TvR Medienverlag.
Sombart, Werner (1916) Der Moderne Kapitalismus, 4 vols. Munich and Leipzig: Duncker
& Humblot. (English translation: Modern Capitalism, ed. Erik S. Reinert. Berlin/Heidel
berg: Springer, forthcoming 2021).
Steinbeck, John (1939) The Grapes of Wrath. New York: Viking Press.
Symcox, Geoffrey (ed.) (2012) Giovanni Botero, On the Causes of the Greatness and Mag
nificence of Cities (1589). Toronto: University of Toronto Press.
Trace, Jamie (2018) ‘Giovanni Botero and English Political Thought’, PhD thesis, Cam
bridge University.
Verdoorn, J. P. (1949) ‘Fattori che regolano lo sviluppo della produttività del lavoro’,
L’industria 1: 3–10.
Villner, Katarina (1986) Blod, Kryddor och Sott. Stockholm: Carlsson Bokförlag.
von Thünen, Johann Heinrich (1826) Der Isolirte Staat. Hamburg: Perthes.
Weber, Alfred (1909) Über den Standort der Industrie (Theory of the Location of Indus
tries). Tübingen: Mohr.
Notes:
(1) A more complete quote from Adam Smith (1759):
The same principle, the same love of system, the same regard to the beauty of or
der…frequently serves to recommend those institutions which tend to promote the
public welfare…When the legislature establishes premiums and other encourage
ments to advance the linen or woollen manufactures, its conduct seldom proceeds
from pure sympathy with the wearer of cheap or fine cloth, and much less from
that with the manufacturer or merchant. The perfection of police [i.e. policy], the
extension of trade and manufactures, are noble and magnificent objects. The con
templation of them pleases us, and we are interested in whatever can tend to ad
vance them. They make part of the great system of government, and the wheels of
the political machine seem to move with more harmony and ease by means of
them. We take pleasure in beholding the perfection of so beautiful and grand a
system, and we are uneasy till we remove any obstruction that can in the least dis
turb or encumber the regularity of its motions. (Smith, 1812 [1759], vol. 1: 320;
my emphasis)
Page 49 of 53
guments
(2) Economist Alfred Müller-Armack used the term in a 1947 book, also defining it as a
Third Way, but it became popular later.
(3) However, a modern version is found in Symcox (2012).
(4) The only exception to this appears to be the Dutch Republic. On this see Reinert
(2009a).
(5) For a more general discussion of these issues, see Reinert and Daastøl (1997).
(6) Pierce (1867: 146). Quoted in Lawson (1989: 68).
(7) Reinert et al. (2017) and Reinert and Reinert (2018).
(8) Reinert (2005).
(9) Trace (2018) documents Botero’s influence on Bacon and other English authors.
(10) See Firpo (1971, 1975).
(11) The Fürstenspiegel (‘Kings’ Mirror’) literature, bringing advice to the rulers on how
to govern, can be seen as part of this broader tradition. The Sachsenspiegel, from Ger
man Saxony, about 1230, is the best known, but even in peripheral Norway, this tradition
goes back to a text from around 1250: Konungs skuggsjá (Old Norse for ‘King’s mirror’).
(12) The tradition of accurate country surveys and descriptions, dating back to De magnal
ibus urbis mediolani of Bonvesin de la Riva (1288) and later works also on the Florentine
state. Such descriptive surveys were the purpose of costly and extensive visitas in the
Spanish provinces of the New World, some of which have been republished (Ortiz de
Zúñiga, 1565/1967).
(13) Of course, Botero made mistakes, such as when his sources were not correct. The re
markable thing, however, is the acuteness of his analysis of generalized wealth and policy
prescriptions that came to typify the centuries of economics that go under the name of
cameralism and mercantilism. Botero was involved in the process of making the Vatican’s
list of prohibited books, which in his day also comprised the works of Jean Bodin (1530–
96). This position clearly gave him access to much new material.
(14) Bacon’s (1625) Essays represent the transition from innovations being a threat to the
status quo and therefore doubtful, as when Roger Bacon was arrested in Oxford around
1277 for ‘suspicious innovations’, into something desirable. Before 1850 Bacon’s Essays
had been translated into Dutch, French, German, Spanish, and Swedish.
(15) ‘Wahre Katakomben von vergessener Literatur’ (Meinecke, 1925: 83n).
(16) Symcox (2012).
(17) The second English translation is clearer on this and is used here (Botero, 1635: 85‒
6).
Page 50 of 53
guments
(18) That is, greater diversity of products.
(19) Presumably this indicates that these products command a higher price and therefore
a higher profit for the producer. Today we could say that manufactured goods are pro
duced under higher barriers to entry than most raw materials, and under increasing
rather than diminishing returns. Both these factors would produce a higher profit margin
for manufactured goods than for raw materials.
(20) The second English translation is clearer on this and is used here (Botero, 1635: 85‒
6).
(21) Botero (1635: 88‒9).
(22) von Thünen (1826).
(23) Schumpeter (1954: 195). Note the term ‘economic organism’, which indicates a type
of economic theory based on biological metaphors, rather than on metaphors from
physics as is present-day theory.
(24) See Patalano and S. Reinert (2016).
(25) I am here referring to communism, fascism, and Roosevelt’s New Deal.
(26) Luis Ortiz’s 1558 memorandum to King Philip II of Spain is a famous example.
(27) Marshall’s footnote here: ‘The land was not able to bear them that they might dwell
together; for their substance was great so that they could not dwell together.’ Genesis xi
ii.
(28) Oresme (1356). For a recent discussion see Schefold (1994/1995).
(29) This was a problem of the whole of Spain, not only in the Spanish Viceroyalty of
Naples where the discussions between Serra and de Santis took place.
(30) Schumpeter (1954: 195).
(31) It should be noted that empirically it may be difficult to distinguish the effects of in
creasing returns to scale from the effects of technological change, simply because ad
vanced techniques only exist in large-scale production (Henry Ford’s technology did not
exist in small-scale production). Schumpeter therefore suggests using the term ‘historical
increasing returns’ to cover the combined effects of scale and technological change
(Schumpeter, 1954: 263).
(32) Krugman (1990: 4).
(33) Verdoorn (1949).
(34) Boston Consulting Group (1972).
(35) Reinert et al. (2017) and Reinert and Reinert (2018).

Page 51 of 53
guments
(36) This has recently changed as my students at Tallinn University of Technology have
been writing such entries as partial fulfilment of their course obligations.
(37) Dublin, William Hallhead, 1779. For the reproduction of a second edition (Dublin, M.
H. Gill & Son, 1882), see http://www.gutenberg.org/files/38841/38841-h/38841-h.htm.
(38) Hamilton (1791).
(39) There are several examples in Reinert (2007). On Mongolia as a particularly ugly ex
ample, see Reinert (2004: 157‒214).
(40) Quoted in Baade (1955).
(41) Marshall (1890: 201).
(42) S. Reinert (2011).
(43) Lin (2012: 350).
(44) Marshall (1890: 452).
(45) Ibn Khaldun quoted in Issawi (1987: 17).
(46) See Trace (2018).
(47) Meek (1976).
(48) Schefold (1994/1995).
(49) Some anthropologists argue that the institution of cannibalism tends to appear in en
vironments with limited sources of animal protein, for example in Central America and
the Caribbean. The conditions create the institutions, not the other way around.
(50) Mueller and Lane (1997).
(51) Acemoğlu and Robinson (2012).
(52) Bollinger (1970).
(53) There were examples of local industrial policies in Latin America, in Peru when the
exports of silver had died out and before guano exports started, and in Brazil before gold
was found in the province of Minas Gerais. But the local elites fell back on traditional raw
material exports whenever a new commodity became available. In that way the industrial
mentality hardly had a chance to be established.
(54) The massive work on Der Moderne Kapitalismus by Werner Sombart (1863–1941) ar
gues for the origins of capitalism in the Italian city states. A first English translation of
the four-volume 1916 edition of this work is scheduled to be published as Modern Capital
ism (Springer, 2021).
Page 52 of 53
guments
(55) As an early confrontation between neoclassical economics and common sense this
quote is precious. But potato chips unfortunately are not a good example to use in inter
national trade. This product is very intensive in transportation costs and is normally not
carried over long distances.
(56) di Corato (1977).
(57) Johnson and Porter (1973: 72‒3).
(58) Minimum efficient size is discussed in Reinert (1980: 143–6).
(59) Cecchini (1988).
(60) Somary (2010 [1952]).
(61) Kaplan (2015).
(62) In Florence, the practice—apparently dating from the 1200s—of prohibiting the
movement of food out of the city, shows that this problem had once been understood. It is
worth noting that while the French economists at the time, the Physiocrats, argued that
all economic activities other than agriculture were sterile, and therefore rendered much
power to the feudal lords, the practice of Florence—one of our earliest democracies—was
to exclude landowners from the political process.
(63) See Reinert et al. (2017).
(64) Darnton (1968).
(65) Mathews and Reinert (2014).
(66) Ruggiero (1997).
(67) Reinert (2004) presents two particularly ugly case studies, Mongolia and Peru.
Erik S. Reinert
Erik S. Reinert a Norwegian citizen, is a professor at Tallinn University of Technology

and guest researcher at the Centre for Sciences and the Humanities at the Universi
ty of Bergen. His research area is the theory of uneven growth, that is, the factors
which—contrary to the predictions of standard economic theory—cause world eco
nomic development to be such an uneven process. He holds a BA from Hochschule
St. Gallen, Switzerland, an MBA from Harvard University and a PhD in economics
from Cornell University. His book How Rich Countries Got Rich and Why Poor Coun
tries Stay Poor has been published in more than twenty languages.
Page 53 of 53
Post-war American Industrial Policy: Market Myths and Production Reali
ties
Post-war American Industrial Policy: Market Myths and

Production Realities
Michael H. Best
Why did the post-war US industrial heartland suffer severe de-industrialization despite its
pivotal and recent contribution to the victory of the Allies in the Second World War? This
is the question that motivates my chapter. The argument is that the US manufacturing
system that won the war came under threat from two existential challenges. The first was
the emergence of more advanced production systems abroad, exemplified by Japan and
Germany. But the second and more important challenge was a historical shift in the
nation’s economic policy framework. With the 1980’s, triumph of neo-liberalism, the regu
latory framework and manufacturing infrastructures that had engineered the nation’s in
dustrial heartland, won the war, and fostered innovation, were eroded and ultimately dis
mantled. Neo-liberalism is a policy framework with roots in the extractive industries of
the South. Plantation owners and employers who use unskilled labour in mines, lumber,
and textiles are always opposed to anything that would raise wages or taxes. What was
new in the 1980s was the triumph at a national level of a neo-liberal economic discourse
and anti-government policy framework that had hitherto been limited to the South.
Keywords: American system, extractive industries, manufacturing belt, Arsenal of Democracy, rust belt, de-indus
trialization, production economics, manufacturing infrastructure, technology management, neo-liberalism, busi
ness system, production capability
20.1 Introduction to the Economics of Produc

tion and Strategic Policy Frameworks
THE 1970s marked a dramatic turning point in the performance of the American econo
my. Manufacturing employment’s share of total US employment declined steadily from
near 25 per cent in 1970 to 9 per cent in 2011 (Baily and Bosworth, 2014: figure 1).1 Real
weekly wages in 2019 were around 9 per cent below their 1973 peak for private-sector
production and non-supervisory workers (Blanchflower, 2019: 6). Described as the ‘great
reversal’, wages that had risen year on year in the post-war era for those with no more
Page 1 of 42
ties
than a high-school education began to fall in the 1980s and continued to do so for three
consecutive decades (Acemoğlu and Autor, 2011; Frey, 2019: 225). From 1980 to 2014,
Piketty, Saez, and Zucman (2018) find that none of the growth in per-adult national in
come went to the bottom 50 per cent, while 32 per cent went to the 50th to the 90th per
centile, 68 per cent to the top 10 per cent, and 36 (p. 556) per cent to the top 1 per cent.
Productivity growth and the nation’s trade balance both declined as well.2
What has gone wrong? In this chapter, I examine the role of industrial policy not as an ex
ternal fine-tuner to a self-organizing, autonomous economy but as integral to the produc
tive structures and core processes that constitute an economy and how it operates. The
industrial policy of a nation is more than a technical issue in economics: it is at the core
of competing visions of a nation’s ideals.
The two major approaches guiding national economic policymaking frameworks in the
world today can be traced back to the economics of Adam Smith (Best, 2018). For simplic
ity one can be called market-centric and the other production-centric.
Given the logic of market-centric economics, industrial policy can only fail as the market
knows best.3 But real-world governments are known to craft production-centric policy
frameworks that foster rapid growth and industrial transformation. For example, ar
guably the most successful industrial policy experience in history was conducted by the
US government during the Second World War. New industries were created, others trans
formed, and output nearly doubled in half a decade. Policymakers of late developers like
Japan, Korea, and China learned about the primacy of productive structures to economic
performance from successful growth experiences, crafted strategic production-centric
policy frameworks, and galvanized business enterprises to drive them. These experiences
open a broader context in which to consider the role of industrial policy and state agen
cies/activities in shaping how economies are organized and that address the defining
questions of textbook economics: what gets produced, how it gets produced, and for
whom.
Industrial policy calls attention to the powers and responsibility of government to ad
vance the manufacturing base of a region or nation, because the manufacturing sector
has unique characteristics in determining prosperity. Its products are tradable, improve
ments in its processes are the source of productivity gains, and it is the last link in an in
novation process chain that includes the translation of scientific and technological innova
tions into products. Industrial policy is not about correcting for market failure; it is about
shaping a nation’s and region’s productive structures.
The question that motivates this chapter is: why did post-war American manufacturing
and the nation’s industrial heartland undergo severe de-industrialization so soon after its
pivotal contribution to the victory of the Allies in the Second World (p. 557) War? I apply
an economic analysis of comparative policy frameworks and production systems to better
understand what went wrong. It takes us inside a set of interconnected processes linking
Page 2 of 42
ties
political power, policy frameworks, production capabilities, business organization, and
skill formation.
The argument is that the American mass-production system that won the war came under
threat from two existential challenges: the emergence of more advanced production sys
tems abroad, exemplified by Japan and Germany, but more importantly an abrupt and his
torical shift in the nation’s economic policy framework. With the 1980s triumph of neo-lib
eralism, the regulations, progressive tax system, labour legislation, and manufacturing
support infrastructures around which the nation’s industrial heartland had been built and
the war had been won were eroded and dismantled.
The neo-liberal policy perspective and its economic and social consequences are not new
to the 1980s; they are as old as the Republic. It is a policy framework and a political econ
omy perspective with roots in the plantation economy of the American South. Ironically,
the idea that free markets and democracy are interconnected has long been used to legit
imize repressive labour regimes, beginning with defenders of slavery (Einhorn, 2006a).
What is new is that the 1980s marked a historic shift in political power and direction of
national economic policymaking undertaken by both major parties. Neo-liberalism res
onates with, and is a triumph of, the economic philosophy, policy frameworks, extractive
industries, and social hierarchy upon which the production system and autocratic busi
ness model of the South were built and have endured for four centuries.
The power of wealth and corporate elites to influence politics and dictate policymaking is
not the whole story, important as it is. It is also a story of the power of economic ideas to
persist, whatever evidence to the contrary may be readily observable. The market-centric
economics axiom that industrial policy cannot lead to improved economic performance
has long outlived its contradiction by post-war Japan and Germany as well as the US Se
cond World War economic experience. Nevertheless, since the 1970s, US economic poli
cymakers have continued to act on the myth and deny the reality.
The policymaking regime ‘assumed’ that strategic Japanese industrial policy could not be
a threat to America’s manufacturing leadership. The post-war economic theory and policy
debates centred around monetarist versus Keynesian macroeconomic stabilization poli
cies. Lacking an economics of production, the debates were blind to the success of gov
ernments abroad in crafting strategic policy frameworks based on advancing their
nation’s business and production capabilities to target the vulnerabilities of post-war
American industry. The goal of policymakers abroad was not to attack the American work
ing class; it was a consequence of statecraft lessons drawn from the Second World War
experience and elsewhere.
American economic policymaking turned in a different direction. Neo-liberalism is an eco

nomic growth agenda that appeals to the ‘invisible hand’ parable to justify ‘market’
deregulation and tax reductions on corporations and the wealthy. Its corollaries include
that deregulation improves business performance, as better managed firms win in free
markets; that deregulated financial markets ensure that the resources (p. 558) of poorly
Page 3 of 42
ties
managed firms get reallocated to superior firms; and that increased productivity gener
ates profits that trickle down to workers and reduce prices to consumers.
As noted, neo-liberalism is not a new agenda to America. The South has always practised
a version of neo-liberal policymaking, beginning with the plantation economy. What
changed in the post-war era was the migration of neo-liberal policymaking from the South
to the national level. A centuries-old civil war between policy frameworks was won by the
South. The historic production-informed policymaking regime of the North succumbed in
the 1980s to the South’s low-tax, deregulation, anti-union, ‘rentier capitalism’ policy
regime which for the first time became the federal government’s economic policymaking
regime. Apart from the rentier elite, it never worked for the South and for the last four
decades it has severely eroded the production capabilities and infrastructural foundations
of the nation’s industrial heartland.
The nation’s policymaking leadership, of both political parties, did not foresee the de-in
dustrialization consequences of neo-liberalism. The economics they ‘knew’ assumed that
the power of the profit motive, if liberated from taxes and regulations, would incentivize
business managers to pursue activities that would maximize national prosperity. It left
them blind to the potential for a neo-liberal policy agenda to shift the managerial incen
tive structure from value creation to wealth-extraction strategies with devastating conse
quences.
The economics they ‘knew’ ignored the historic public-sector role in shaping the organiza
tional and extra-firm infrastructural foundations of the nation’s manufacturing system
and their dependence on public investment on the one hand and their vulnerability to
competition from foreign production systems on the other. The production engineering
expertise and managerial culture by which manufacturing enterprises had invested heavi
ly in innovative production and technological capabilities and skills to the benefit of the
nation gave way to shareholder pressures and innovations in financial engineering prac
tices to the benefit of wealthy and shareholding elites. The reduction in taxes in the
deregulated environment created a new incentive structure for corporate managers to
join the shareholding elites and amass fortunes. It unleashed the drivers of a rapid in
crease in inequality. Even as the nation’s manufacturing belt became the ‘Rust Belt’, the
financialization of America’s industrial enterprises created huge profit opportunities for
hedge fund owners, asset management companies, and investment banking. The rentier
landowning class of the South was joined by a rentier financial elite.
Policymakers, too, fell victim to the power of free-market economics as a political weapon
to shape popular economic discourse and influence economic policy. Economic policymak
ers did not reckon on the power of special interests, if unchecked by government regula
tion, to ‘financialize’ a nation’s manufacturing enterprises, toxify their workplace environ
ments, and undermine their productive structures. Absent the strategic direction from po
litical leadership and policymakers’ awareness of the cost in loss of production capabili
ties and manufacturing employment to America’s industrial heartland, the neo-liberal
Page 4 of 42
ties
agenda became national policy with devastating consequences on the families and com
munities on which they depended.
The policy framework of the South has always been, in effect, anti-skill-intensive
(p. 559)
manufacturing. The South has, with rare exception, never developed learning firms, man
ufacturing clusters or innovative city dynamics and, except for foreign car companies, has
always imported rather than produced manufactured products. This, tragically, is the an
ti-industrial policy framework that has become national policy over recent decades. It is
the story of how Dixie won a centuries-old, North–South civil war of production systems.
In defending and expanding on these claims, this chapter is organized as follows. First,
we bring America’s Second World War production ‘miracle’ down to earth with an exami
nation of the policy framework, implementation agencies, and organizational change
methodologies through which policymakers organized a transformative advance in the
nation’s industrial performance. It takes us deep into the internal productive structures of
business enterprises and change programmes by which government and business joined
forces to upgrade production methods, educate a workforce with the requisite skills, and
advance productivity on such a scale as to make a step change in national economic per
formance. Successful implementation drew upon the active participation of management,
production engineers, and skilled labour to make changes in organizational practices.
While the US Second World War experience is a special case, it is consistent with histori
cal accounts of the role of government leadership in crafting strategic policy frameworks
and organizing enterprise change methods and infrastructural agencies to advance the
productivity of a region’s or nation’s business enterprises (Best, 2018).
Next we examine the post-war policy legacy. The Second World War experience perma
nently transformed the interconnectedness of government and the economy in major
ways. The US transitioned from a virtually non-existent pre-war weapons industry to a de
fence budget larger than that of all other nations combined. A huge government-contrac
tor system was created with an estimated 15,000 defence research contracts in 1950 ex
panding to perhaps 80,000 in 1960.
Wartime organizational innovations in economic governance and industrial policy were

selectively adapted and permanently institutionalized in the post-war era. For example,
the wartime creation of administrative agencies to design and produce technologically
advanced weapons systems became the world’s biggest and most advanced science and
technology infrastructure. In contrast, wartime successes in designing management and
manpower development methodologies to foster workforce participation in decision-mak
ing and continuous innovation were, unfortunately, not institutionalized. In other coun
tries, such as Japan and West Germany, they became the manufacturing cells of rival pro
duction systems with superior new product development and technology management ca
pabilities.
Post-war macroeconomic policy was profoundly changed by the transition from the era of
Pax Britannica to Pax America. Pax America offered the defeated nations of Japan and
Germany open access to the vast American market. It involved a free-trade policy and an
Page 5 of 42
ties
overvalued exchange-rate policy that created opportunities for the defeated nations to
craft export-driven industry policy frameworks organized around (p. 560) innovation-led
manufacturing systems and Schumpeterian business strategies. US macroeconomic poli
cy, in contrast, focused on stabilization policy issues based on the belief that US techno
logical leadership would ensure US manufacturing leadership.
The strategic policy framework of Japan both built on the strengths of the American
mass-production model and exposed its weaknesses. Japan’s manufacturing advantage
was not based on building a bigger science and technology infrastructure than that of the
United States. Instead, Japanese business leaders and policymakers focused on building
competitive advantage by superior manufacturing capability. They began by applying
lessons from the US Second World War experience about the policy interconnectedness of
production capabilities, business organization, and skill formation. As they progressed
rapidly up the production capability spectrum (Best, 1990, 2001), American mass produc
tion became a victim of its past success.
The war-winning advantage of American mass production was organized around

economies of speed (throughput efficiency) and the principle of single product flow. It
maximized throughput and productivity by hardwiring the factory to a single product and
a single design (Best, 2001). The product’s design was frozen because design changes re
quired the whole production line to be re-engineered. A single product was required be
cause multiple products would mean changing machine set-ups and require specialist ex
pertise to change over from one product to another, which could cause the production
line to be shut down for hours or days, if not more. It was a highly productive system, but
it was inflexible. As Henry Ford famously said, ‘Any customer can have a car painted any
color that he wants so long as it is black.’
Japanese engineers created a flexible mass-production system based on the principle of

multi-product flow. They built factories that were organized to produce multiple products
on the same assembly line and to transition from one to another and back again in min
utes. Participatory and multi-skilled labour practices enabled quick changeovers from
product to product and continuous improvement (kaizen) in production methods to de
sign quality into the production process. It was a labour system in which work teams,
multi-skilled workers, and manufacturing cells replaced the single-task worker as the pri
mary work unit. The result was revolutionary; it set a new global standard for throughput
efficiency, cost, quality, and delivery time (Best, 2001). It was a challenge that demanded
a transformation in national-production industrial policy at a time in which market-cen
tric economics was hegemonic.
However, an internal challenge to America’s industrial heartland was looming in the form
of a rival production system with roots going back four centuries. Manufacturing as a sys
tem of production and industrial policy took root in the north-east roughly a century and
a half after the plantation economy, coincided with the founding of the Republic, and
evolved into a highly diversified manufacturing production system in the cities and states
of what became the nation’s manufacturing belt. The Northern ‘social democratic’ policy
Page 6 of 42
ties
framework involved collecting taxes for the public provision of material and organization
al infrastructures, such as public education and the federal armouries, and a financial and
regulatory environment to support the development of the nation’s nascent manufactur
ing enterprises. Concentrations of wealth were limited (p. 561) by highly progressive in
come and wealth taxes; by anti-trust and regulatory policies; and by restricting the in
volvement of corporations in politics. The historic activist state that sought the advan
tages of corporate economic power while limiting the disadvantages of corporate political
and market powers were leveraged by both New Deal and Second World War industrial
policymaking. The Second World War industrial policy framework went beyond the New
Deal in galvanizing the synergistic power of US business and labour working together to
build a highly productive economy. Tragically, this constructive alliance was immediately
attacked with the passage of the anti-union Taft Hartley Act in 1947, which heralded the
first major neo-liberal victory.
The Cotton Belt of the South began as a plantation economy with a landholding aristocra
cy and slave labour. From its early days the policy framework of the South was crafted to
protect the institution of slavery, based on fears that the North would use national policy
making powers, such as taxation, to suppress and destroy the forced labour regime (Ein
horn, 2006a). The Southern hierarchy defined itself in opposition to the emergent manu
facturing development initiatives of the Northern colonies from their origins in the 1700s,
as it was clear that the latter would depend upon free labour.
Fast forward to the massive New Deal federal transfers, physical infrastructure, and mili
tary spending that turned the ‘Cotton Belt’ into the post-war ‘Sun Belt’. The political
elites, the policy framework, and the dominance of extractive industries did not change.
The oil and gas industry along the southern coast extended westward, and with mining
and forestry northward joined the historic mining industries in the Rocky Mountain
states. The manufacturing strategy of southern states and towns was to compete with one
another to attract Northern companies seeking government subsidies, low taxes, anti-
union laws, and lax regulation. Indigenous manufacturing did not take hold.
The South’s manufacturing strategy worked to the extent that many Northern firms,
faced with a choice between restructuring to meet the new world-class production stan
dards and relocating to the anti-union, low-cost environment of the South, chose the lat
ter. However, by the 2000s, the ‘smokestack-chasing’ manufacturing strategy failed, as
the South’s major labour-intensive industries, led by textile and furniture firms, suc
cumbed not only to low-wage competition but to enterprises with more advanced manu
facturing capabilities from China and elsewhere. Consequently, gas and oil production
and mining, combined with federal defence spending, have become the mainstays of the
region’s economic livelihood.
Given four centuries of failed economic progress it seems incomprehensible that the anti-
labour policy framework of the South would be adopted as national policy. Critically, polit
ical leadership, mainstream economics, and economic policymakers failed to learn from
the policy framework of the American South and all other plantation economies before
Page 7 of 42
ties
dismantling progressive income and wealth taxes, industrial policies, and regulatory
agencies.
In academic economics, macroeconomic modelling absorbed Keynesian demand manage

ment as a special case of market failure, and stabilization policy dominated economics ed
ucation and popular discourse. The stagflation era of the 1970s eroded (p. 562) public con
fidence in the efficacy of stabilization instruments to fine-tune the economy and created
the opening for an anti-industrial policy and a neo-liberal economics agenda to shape
public discourse.
Even though the South benefited most from America’s de facto industrial policy of mili
tary contracts and subsidized corporate clients, its apparent antipathy to industrial policy
was central to its political ideology. In the real world of industrial policymaking, the South
fostered a new American version of ‘state capitalism’ in which political elites leverage
their government insider power to become rentier capitalists (Lind, 2003).4
In the name of free enterprise, the neo-liberal policy framework campaigned for deregula
tion of financial markets, elimination of anti-trust legislation, the passage of anti-union or
ganizing laws such as the Labor Management Relations Act of 1947 (better known as the
Taft Hartley Act), and relaxation of social controls and historic norms of corporate gover
nance. Critically, the historic stakeholder model of business organization and manage
ment norms was overtaken by the ideology propounded by Milton Friedman and Michael
Jensen that the only purpose of business was to make profits and enrich shareholders.
The appointment of Lewis Powell to the Supreme Court by President Nixon in 1971 sig
nalled the elevation of a key player in the crafting of the Southern strategy. Powell, an at
torney for the tobacco industry, authored the Powell Memorandum, a secretive campaign
to mobilize billionaires to fund neo-liberal think tanks to influence public opinion about
the ideals of free markets. The goal of the neo-liberal campaign was the passage of legis
lation to dismantle New Deal institutions but also to promote market fundamentalist lead
ership of the economic policymaking agencies at the national level, including the Trea
sury Department, the Federal Reserve Bank, and the Commerce Department. It was a
campaign that did not depend upon who was elected but aimed to sway public opinion in
dependent of political party. Thus began the age of dark money in American politics, fund
ed by the wealthy beneficiaries of extractive and monopolized industries.
The costs to the North of the neo-liberalism era have run deep not only in fiscal transfers
to the South but more especially in the erosion of the nation’s material and extra-firm in
frastructures (educational, banking, engineering support) that co-evolved with the
nation’s manufacturing belt and upon which the nation’s industrial leadership had been
built. The industrial heartland lost its voice in policymaking at great cost to the nation’s
industrial performance in loss of production capabilities, skills, high-wage employment,
and tax bases upon which the region’s once prosperous cities depended.
Page 8 of 42
ties
In the final part of this chapter, an economics perspective and industrial policy frame
work is introduced to address the task of transforming the American economy to meet the
challenges of climate change and four decades of eroding infrastructural foundations,
both material and organizational. The term ‘capability triad’ is presented as an expository
device that draws upon the common policy dimensions of a series of (p. 563) transforma
tive experiences on the scale of economic ‘miracles’ led by Franklin Roosevelt (Second
World War), Deng Xiaoping (China), Taiichi Ohno/Edwards Deming (Japan), Ludwig Er
hard (West Germany), Ralph Flanders (Massachusetts), and Kenneth Whitaker (Ireland)
(Best, 2018).
The economic ideas that informed the New Deal are contrasted with those that informed
the Second World War policy framework. President Roosevelt drew critically important
lessons from the failure of the New Deal as a transformative policy agenda. The Ten
nessee Valley Authority (TVA) was the centrepiece of the New Deal policy framework. It
was hugely successful in the electrification of the South as the Hoover Dam was in the
South-west and Grand Coulee Dam was in the North-west. As material infrastructure,
they were critical to building the Arsenal of Democracy, but not sufficient.
The TVA experience is instructive from a capability triad perspective. It created a grow
ing demand for electricity which did become a regional export industry, but as new
sources of hydro power ran out the TVA turned to coal as a source of energy (Jacobs,
1984).5 It did not shift the South’s structural dependence on extractive industries or the
interconnnected labour system and deeply anti-democratic skill formation system that ap
plied in both public and business spheres. Dixie has always meant a steadfast commit
ment to not increasing the skills of the region’s workforce. It started with reconstruction-
era fears that a numerate workforce would understand how to counter the debt-peonage
system and instigate worker rebellion. But without advancing skills, business enterprises
cannot successfully pursue Schumpeterian product-led competitive strategies and build
new product development or technology management capabilities.
This was a hard-earned lesson that distinguished the focus of President Roosevelt’s Se
cond World War industrial policy on fostering business capability development and har
nessing the production engineering and human capital expertise of America’s most suc
cessful manufacturing enterprises. It worked. It provided lessons that were taken up by
the defeated nations but unfortunately ignored by the victorious.
20.2 The US Second World War Production-in

formed Policy Framework
The most successful industrial policy experience in history is arguably that conducted by
the US government during the Second World War, in which President Roosevelt’s vision of
an ‘Arsenal of Democracy’ involved a near doubling of national output in a (p. 564) half a
decade. How was it done? The answer was to organize an interdependent innovation-
process chain linking basic research, developmental research, and applied research, com
Page 9 of 42
ties
bined with the diffusion of world-class production capabilities. The US business system
was transformed to harness the driving force of innovating enterprises. In the process,
the US production system was, in effect, re-engineered to accomplish a step change in
performance standards. An inter-organizational technology management capability was
institutionalized as a consequence.
The manufacturing process in virtually every enterprise in the United States was re-engi
neered to achieve the output targets. It required a national translational research capabil
ity to be created that linked ‘top-down’ enabling agencies with ‘bottom-up’ operational
drivers. Implementation involved the combined efforts of a set of unified extra-firm infra
structural agencies led by the Office of Scientific Research and Development (OSRD), the
War Production Board (WPB), the Manpower Development Commission (MDC), and the
Defense Plant Corporation (DPC). Each was dismantled at the end of the war.
The key to implementation was inter-organizational connectedness. The OSRD, led by

Vannevar Bush, was tasked with building a national science and technology ‘extra-firm’
infrastructure that could create an inter-organizational capability to design, develop, and
produce technologically advanced weapons systems. Examples such as microwave or
radar, penicillin, and synthetic rubber required basic, developmental, and applied re
search to be married with design for manufacturability by mass-production facilities. The
latter required the production base of the country to be re-engineered to meet the perfor
mance standards of cheaper, better, faster. When adopted and extended by Japan in the
post-war era, the process innovations became widely known, imitated, and diffused as JIT,
TQM, and SDWTs.6
With 12 million Americans transferred into the armed services, the production system
had to be transformed. Designing the means to undertake the transformation across the
economy to meet the production targets was the task of the WPB and the MDC. For exam
ple, meeting the production target of one B-24 bomber per hour required the supply
chain to deliver 1.5 million parts to the factory gates per hour. The War Manpower Com
mission designed and operationalized the functional equivalent of a fast-track national
training programme to equip both management and labour with the skills and organiza
tional methods to meet the production targets.
Economic statisticians played a key role. The engineering challenge through which the
nation’s production system was restructured was critical to the implementation of the
Victory Program itself written by Simon Kuznets, chief economist at the War Production
Board. He and fellow economic statisticians diligently created statistical tables on the ex
isting and planned output of the nation’s manufacturing enterprises. They identified se
quential bottlenecks that required investments to remove them. Kuznets, as the father of
US national income and production accounts, was the perfect (p. 565) person and perhaps
the only person who could have linked the nation’s production system requirements with
the Allies’ war strategy.
Page 10 of 42
ties
No doubt, the US Second World War policymaking experience was a special case. Never
theless, it is informative with respect to an economics of production, innovation dynam
ics, and enterprise capability development.
It also offers a powerful negative lesson, examined more fully in section 20.3. US post-war
policymaking has been dominated by debates over stabilization instruments and targets
that, in effect, rendered the production side of the economy invisible. The focus of the De
partment of Commerce’s national account measurements of Keynesian expenditures and
income reflects this invisibility, as does the failure to incorporate a production side into
macroeconomic models. The dominance of macro stabilization economics and the omis
sion of production knowledge has prevailed at the Federal Reserve Bank and the Trea
sury, the pinnacle of economic policymaking over recent decades.
Kuznets and his team of statisticians did not have computers or today’s sophisticated da
ta-generating survey techniques, but they did construct tables measuring the production
capacity of the nation’s enterprises with estimates of the links between existing output
and the output that could be achieved by undertaking macroeconomic bottleneck analysis
and diffusing the best human resource change methodologies. These calculations were
necessary inputs into the estimates of the nation’s supply capabilities that lay behind the
strategic military decision to delay the Allied invasion of Europe until 1944. Without
them, the supply lines required for the D-Day landings would have been problematic, and
the decision was made to not land in Europe until the supplies were in place to support
the military drive across Europe of up to 4 million men and 1 million tanks, aircraft, and
vehicles in the year-long effort.
The wartime government did not depend upon or seek centralized authority to plan the
economy as demanded by the military. It meant subordinating military authority over pro
duction planning to the economic statisticians at the WPB and subordinating the technol
ogy priorities of the military authorities to the OSRD. Within the WPB, the economic sta
tisticians, partnering with production engineers, did not seek to supplant operational de
cision making at the enterprise level. Instead, they devised policies, including innovative
planning accounting measures, to galvanize the energies of those with the requisite ex
pertise, skills, and experience to design methods and practices to make advances in pro
duction performance happen.
The government took over the banking system’s role described by Schumpeter as the
‘headquarters of the capitalist system’. During the war, two-thirds of expenditure on in
dustrial facilities was directly financed by the government (White, 1949: 156; Jones and
Angly, 1951). Economic policymakers went outside the market system to create a leasing
mechanism to negotiate the industry–government divide. The Defense Plant Corporation
(DPC) was created in 1940 as a subsidiary of the Depression-era Reconstruction Finance
Corporation (RFC), ‘with such powers as it may deem necessary to aid the Government of
the United States in its national defense program’ (White, 1949: 161). The rules govern
ing the RFC’s Depression-era financing of industry were (p. 566) modified in May and June
of 1940 to enable subsidiaries of the RFC to finance and own defence plants, and to lease
Page 11 of 42
ties
them to private industry (White, 1949: 161). Jessie Jones, the head of the DPC, was de
scribed by Time magazine in the following terms: ‘In all the U.S. today there is only one
man whose power is greater: Franklin Roosevelt’ (13 January 1941).
20.3 Post-War Policymaking Legacy

The Second World War experience permanently transformed government and business in
stitutional relationships in major ways. The neglect of economic analysis of these changes
and consequences stems in part from the triumph of the free-market paradigm in post-
war economic theory and policymaking. In this section, legacies of the Second World War
transformational experience that permanently changed government and industry inter-re
lationships are sketched.
First, while the Second World War mobilization agencies that implemented the transfor
mation and galvanized growth were disbanded, many of their functions were incorporat
ed into rapidly expanded departments of the federal government. The Department of De
fense and the Department of Energy became the lead agencies in building a vast science/
technology infrastructure which linked government agencies, industrial, university, and
government laboratories, and business enterprises. The publicly stated purpose was na
tional security. The Manhattan Project and the atom bomb seared into the body politic the
role that science and scientists could play in defence and war. It was a de facto industrial
policy administered by the Department of Defense. The United States does not have a de
partment of industry, as industrial policy is ostensibly run by the Department of Com
merce, but the title of the department is a giveaway: it is not about industry as the do
main of production.
Second, and closely related, the US government established an unrivalled management

capability in inventing disruptive technologies that to date business enterprises have
found difficult if not impossible to emulate. DARPA (Defense Advanced Research Projects
Agency) has had a major influence on the invention of, for example, the Internet, UNIX,
GPS, stealth fighters, the Windows operating system, the World Wide Web, videoconfer
encing, and Google Maps.
Third, wartime planning, coordination, and accounting practices originally developed to

manage multi-divisional enterprises were adapted to create a national industrial planning
capability. Known as control materials planning (CMP), the performance accounting sys
tem involved the creation of a new tier of inter-organizational management by which offi
cers in government departments and prime contractors worked together, shared offices,
and moved freely between government and contractor employment (Best, 2018). Today
Boeing, for example, has a thousand employees working in the Federal Aviation Authority.
Fourth, the government institutionalized a vast procurement capacity, as it was

(p. 567)
and has remained the major customer of emergent technology-driven sectors. In fact, ma
jor export industries of America in the post-war period received substantial government
R&D and purchasing support in their formative, low-productivity years (Markusen and
Page 12 of 42
ties
Yudken, 1992). This included aircraft, computers, electronics, telecommunications, and
instruments. It has been estimated that the federal government funded 85 per cent of the
R&D for the emerging electronics industry in the 1960s but as late as the mid-1980s the
government purchased over two-thirds of the output of the aircraft industry.
The wartime armaments industry morphed into a huge permanent government-contractor

system and created what H. L. Nieburg (1966) termed the ‘contract state’. In the words
of Daniel Kevles (1978): ‘For almost a quarter century after 1945, defense research ex
penditures rose virtually exponentially, even in constant dollars, accounting through 1960
for 80 percent or more of the entire federal R&D budget. In 1950, it was estimated that
there were 15,000 defense research projects; in the early 1960s, perhaps 80,000’ (1978:
466, cited in Hounshell, 1996: 48).7
Fifth, in macroeconomic policymaking the government combined Keynesian demand man

agement with military expenditures which became a de facto long-term employment poli
cy. While economics employment policy discourse focused on fiscal and monetary instru
ments, real-world high employment was dependent upon military spending.
No other country had a science and technology infrastructure and government participa
tion in the business world approaching the scale of the United States. In this respect, US
defence policy is, in fact, its post-war industrial policy. Within the defence establishment
it is celebrated as dual-use technology policy. Government-funded R&D for defence pur
poses generates new product and process ideas that can be put to commercial use as well
as that of national security. As industrial policy its success record is mixed. On the plus
side, it has fostered world-leading, high-tech regional innovation systems in Silicon Valley
and Greater Boston. Both regions have enjoyed a rapidly expanding technology base, re
plenished by technological breakthroughs at an unprecedented rate by rapidly growing
spin-off firms from world-leading universities.
However, important as they have been in technology development, policy initiatives else
where to establish high-tech clusters have not proven tractable, particularly in localities
suffering from the loss of manufacturing enterprises. This is understandable. The wartime
government’s Office of Scientific Research and Development attracted many thousands of
scientists and engineers to research, develop, and design advanced technology weapons
systems in California and Massachusetts at huge expense. The result was a legacy post-
war science, technology, and skill base which formed the (p. 568) foundation for emergent
industrial ecosystems in which early-stage technology development business systems
could flourish. No single or even multiple state governments can invest more than a small
fraction of what the federal government invested in Greater Boston and California from
San Francisco south to San Diego (Wright, 2020).
In any case, de-industrialization has persisted despite the combination of New Deal infra
structure investment in the South, the creation of the world-leading science and technolo
gy infrastructure, and a huge government procurement programme including for emer
gent sectors. Manufacturing employment declined by one-third between 1969 and 1996
and by a third again in the 2000s. While service and temporary employment offset the
Page 13 of 42
ties
lost manufacturing jobs, they did not provide the US$20/hour-wage plus benefits for non-
college educated workers.
Macroeconomic policies did not come to the rescue. In fact, the wartime legacy turned
what had been an inward-looking to an outward-looking orientation which, unintentional
ly, was a major force in driving American de-industrialization. The Second World War end
ed the long era of the British Empire and Pax Americana filled the vacuum. The United
States and the Pentagon became the administrator of the nation’s 2,900 military bases
around the world. In exchange for becoming allies in the Cold War the new US foreign in
dustrial policy removed all barriers to imports and offered open access to the world’s
biggest market. The defeated nations took full advantage of free entry into the massive
US market to build manufacturing export industries. Moreover, these countries pursued
exchange-rate and interest-rate policies in support of rebuilding their manufacturing
bases. The exchange- and interest-rate policies pursued by the United States, meanwhile,
decimated domestic manufacturing and supported the strategic export-driven policy
frameworks of other nations.
Most prominently, the long, costly, and unpopular war in Southeast Asia was largely fund
ed by monetary creation rather than fiscal policy and the resultant inflation of the 1970s
(Best, 1972). Federal Reserve Chairman Paul Voelker, appointed from Goldman Sachs in
1979 by President Carter, increased the interest rate to 17 per cent, seven points over the
inflation rate, with devastating effects on the cost of credit upon which manufacturing in
vestment is dependent. The high interest rates attracted huge capital inflows and the de
mand for US dollars led to a 40 per cent increase in the exchange rate over the next four
years. The consequences were devastating for the profitability and global competitive ad
vantage of US manufacturing. It marked the end of the Keynesian macroeconomic era
and ushered in the Voelker-Greenspan-Reagan era of neo-liberalism. Clinton acquiesced
and agreed to Japan’s manipulation of the value of the yen in order to support Japanese
manufacturing growth.
In its simplest terms, the United States turned to a stabilization-informed macroeconomic

policy framework at the same time other nations were subordinating stabilization policies
to export-driven, capability-informed macroeconomic frameworks. The high interest rates
attracted short-term capital inflows which drove up the demand for dollars and offset the
growing trade imbalance in manufactured goods. Here America’s macro and trade poli
cies contributed hugely to the conversion of the manufacturing belt into the Rust Belt.
(p. 569)
20.4 The Japanese Economic Miracle: Rival

Technology Management Paradigms
It is now conventional wisdom that, first in Japan and then in Germany, competitors
emerged that built manufacturing export platforms based on more advanced production
systems capable of achieving higher performance standards. In time they gained market
Page 14 of 42
ties
share in the United States, which had been the world’s leading manufacturing nation. But
less appreciated is that competition between the Japanese and American production sys
tems was at the same time competition between two rival models of technology manage
ment at both the enterprise and industrial policy levels. In fact, foreign rivals pursued
policy frameworks that turned America’s massive advantage in science and technology to
their own manufacturing advantage.
The competitive challenge to American industry began to appear in the 1950s when the
United States first began losing market share to Japan in the steel industry. It was attrib
uted to a lack of investment in new continuous casting technologies by lethargic Ameri
can steel companies, long-time members of a cosy cartel. But market share declines in
complex production industries such as cameras, cars, and motorcycles soon followed.
There was little cause for concern as various interpretations of the causes of decline in
‘traditional’ industries all agreed that America’s strength in high-tech industries was
unassailable. Heads finally turned in the early 1980s as the pattern was repeating itself in
semiconductors followed by semiconductor equipment makers and downstream indus
tries such as advanced consumer electronics, computers, and telecommunications. In the
late 1980s, Japanese machine-tool companies mounted a vigorous and successful attack
on America’s machine-tool industry with devastating effect. They did so by ramping up
the production of CNC (computer-controlled) machine tools, a technology that had been
invented in the United States but diffused rapidly into manufacturing in Japan.
The Japanese production system evolved, in stages, into a superior technology manage
ment capability that exposed a weakness in the American production system. During the
wartime production miracle, technological innovations were separated from manufactur
ing. Innovations in radar systems, for example, were not introduced to the aircraft pro
duction lines. The weakness of Germany and the reason it could not expand production of
aircraft even close to the US scale is that they continuously changed aircraft design to im
prove its functioning. The United States did not; the design was frozen. The miracle ad
vances in technologically advanced weapons systems were organized entirely separately.
Japan’s manufacturing advantage was not based on building a bigger science and tech
nology infrastructure than the United States. The Japanese business leaders and policy
makers focused on building competitive advantage by superior manufacturing (p. 570)
performance. They began by applying lessons about the primacy of production capabili
ties, business organization, and skill formation from the US Second World War experi
ence, and progressed rapidly up the production capability spectrum (Best, 1990, 2001).
Japanese enterprises advanced and institutionalized Second World War US production

and management principles in the form of just-in-time, the quality movement, and kaizen.
Unlike US post-war industrial policy, their policy framework was strategic. For example,
they learned from W. Edwards Deming that US manufacturing enterprises did not em
brace or even understand the fundamental organizational principle of system integration
and systems thinking.8 America’s leadership in industry had been based on production
engineering; but production engineering was never a concept in economic policymaking
Page 15 of 42
ties
in the post-war era and without policies based on an accurate conception of production
there could be no production economics. Second, Japanese enterprises moved beyond
competitive advantage in production performance in cost, quality, and throughput effi
ciency to rapid new product development and technology management. Third, based on
superior technology management capability, Japanese manufacturers converted access to
America’s scientific and technology infrastructure into a strategic competitive advantage
against US manufacturing.
Unfortunately, in the post-war era, much of American business had turned to price-led,
low-cost strategies which locked them into productive structures and work organizational
practices and blinded them to the real competitive threat of enterprises elsewhere that
emphasized increasing labour productivity rather than lowering wage costs.
The emergent Japanese manufacturing advantage can also be described as a rival tech
nology policy paradigm anchored not in leadership in science but in organizing an indus
trial ecosystem in which business enterprises could pursue a strategy based on building
world-class manufacturing capabilities in production, new product development, and
technology management. In the United States the presumption was that US technology
policy would prevail, since Japan lacked a vast science and technology infrastructure that
included an unrivalled research-intensive university system producing scientists and engi
neers.
The result was a change in the dominant form of competition from price led to product
led. The defining feature of the mass-production system when it emerged, and the reason
that it put all other production systems on notice in its time was that it drove down the
costs and prices of production. The defining feature of the New Competition emergent in
Japan was rapid new product development created by the marriage of (p. 571) productivi
ty and innovation, and the redefinition of both. The result was a new interactive manufac
turing and innovation dynamic. Japan did it, not by investing heavily in the creation of a
science and technology infrastructure, but by building a production capability that could
turn America’s technology knowledge base into a national competitive advantage for
Japan at the expense of American manufacturing.
The rival technology paradigm was about shifting the technological base from the science
end of the spectrum to the production end. Technological innovation is driven as much by
competitive dynamics as by ‘science push’.9 Research in fundamental science is less im
portant than cultivating the already existing technology base which can be harvested for
new technological combinations and permutations as part of the product development
process.
They all involve the creation of new technological knowledge. Knowledge is certainly cru
cial to new product development. But knowledge is not research; knowledge is the accu
mulated stock of wisdom that can be reinterpreted, combined and recombined in new
ways. Technological knowledge, unlike scientific knowledge, is not always codified or
even explicit. Polanyi, Penrose, and others have stressed the role of technological and ex
periential knowledge as tacit or informal knowledge often built into collective organiza
Page 16 of 42
ties
tional practices, much as craft skills are inseparable from the muscle memory of machin
ists and technicians who have learned by doing.
To summarize, product-led competition engendered new organizational capabilities which

involve the redefinition and integration of three processes:
1. Manufacturing. The cell is the building block of the whole edifice; without cellular
manufacturing the rest of the business system cannot drive product-led competition.
2. Design/manufacturing cycle. Companies need to compete on the basis of rapid
new product development, or they will fall behind in technology adoption.
3. Technology diffusion. This is pulled by the first two processes as distinct from be
ing pushed by autonomous R&D activities.
The US technology leadership paradigm was based on three implicit assumptions. In this
it was a failed industrial policy strategy.
The first assumption was that technological innovation travels along a one-way sequence
of activities proceeding from basic science to applied science to engineering science to
engineering design and development to production. The presupposition was that scientif
ic pre-eminence generated technological pre-eminence which, in turn, assures product
pre-eminence. The second assumption was that technology diffusion follows a 25‒50-year
cycle. By the time that Japan and other countries were able to capitalize on (p. 572) the
commodity production stage of the product life cycle, US firms would have moved on to
leadership in the development of new products. The third assumption was that organiza
tional capabilities are not important in explaining competitive advantage.
Therefore, the threat to US manufacturing competitiveness came as a surprise. It was

witnessed first in an order-of-magnitude advance in both lower-cost and higher-quality
performance previously assumed to be a trade-off. Anchored in a participatory model of
work organization, the Japanese production system established new performance stan
dards for cost, quality, and time to market. The latter was a consequence of superior new
product development and technology management capabilities. The consequences of the
loss of competitiveness of American manufacturing were evident with the unexpected fis
cal crises of US manufacturing belt cities as early as the 1950s. To understand these new
dynamic forces, including the structural roots of de-industrialization, we must go inside
the policy frameworks and core processes that drive modern economies both in the Unit
ed States and other nations.
20.5 Production System Legacies: The North

versus the South
The United States has always been divided between a Northern and a Southern produc
tion system and social structures. While part of a single nation, the two regions evolved
radically distinctive models of capitalism, modes of economic governance, business orga
nization, and labour systems. From the early days of the plantation economy, the South
Page 17 of 42
ties
has been organized around extractive industries. In contrast, from the early days of the
manufacturing economy, the North underwent a historical trajectory of industrial transi
tion and technological innovation.
Economies are complex and terms arise in economics discourse to simplify and enable
discussion. For example, the term ‘miracle’ is commonly used to describe rapid economic
transformations that are outside conventional economic theory and policy frameworks. I
suggest that in the United States a common taxonomy of geographical belts is another ex
ample. Miracles and belts are metaphors that facilitate communication about economic
matters that are deemed important, but which are not anchored within the boundaries of
market-centric economics. Efforts to better understand both require concepts that under
mine the assumptions required by market fundamentalism. In this section I apply the
metaphor of economic ‘belts’ to contrast regionally distinctive systems of production and
economic governance that evolved in the United States.
It has a second purpose. The major argument of the chapter is that the neo-liberal policy
revolution in the 1970s was not new to the nation but has always dominated economic
governance in the South. Its extension to the national policy framework for the first time
was a triumph of the South’s political elite and the production system it governs.
(p. 573) 20.5.1 The Manufacturing Belt
America’s economic governance practices going back to the founding of the Republic fo
cused attention on infrastructural requirements, both physical and organizational, to ad
vance the nation’s productive structures. Educational, technical, and financial infrastruc
tures were organized within a strategic economic governance framework to promote
manufacturing and national independence.
Alexander Hamilton’s attention to manufacturing is well known, but I suggest that

Thomas Jefferson was an equally important developer of the emerging and globally dis
tinctive Northern production system (Best, 2018; Winchester, 2018).10 As President
Washington’s ambassador to France, Jefferson, like Benjamin Franklin, was highly
learned in the advances being made in French science and engineering. Jefferson brought
to America the production principle of interchangeability, a French invention but not
practice, and championed its introduction in the nation’s armouries. Under Colonel
Roswell Lee, the Springfield Armory in Massachusetts was organized to operationalize
the principle of interchangeability of parts which institutionalized product engineering as
a defining characteristic of what the British termed the ‘American system of manufactur
ing’ or just the ‘American System’ (Marshall, 1920 [1890]: 257‒8). This became the strate
gic industrial policy that organized the creation of the first machine-tool industry based
on interchangeability, a precondition for American mass production. It also unleashed the
competitive drive for ever-smaller dimensions and precision engineering, perhaps the
most important and enduring source of American technological leadership. The drive to
ever-smaller critical-size dimensions is at the core of America’s unmatched advantage in
opening new technological domains (see figure 3.1 in Best, 2018).
Page 18 of 42
ties
Industrial policies that are informed by fundamental production principles are integral to
advancing a nation’s productive structures and establishing industrial success and nation
al competitive advantage. But implementation depends upon investment in engineering
education. As president, Jefferson founded the nation’s first engineering college at West
Point, birthing the Army Corps of Engineers. Thus began the institutionalization of the
American civil engineering expertise which enabled physical infrastructural investments
in transportation and energy and the creation of America’s ‘industrial heartland’. Presi
dents Abraham Lincoln, a railroad lawyer, and Franklin Roosevelt also undertook nation
wide policies to advance engineering and science education to upgrade the production
capabilities of the nation’s business enterprises and increase the skills and productivity of
the nation’s labour force. It is but one form of extra-firm infrastructure that impacts on
business system development, opportunities for innovation, and national productivity.
In these ways the skill formation domain of the triangular relations linking gov
(p. 574)
ernment, education, and industry was shaped in sync with the emergence of a machinist
community that was integral to the emergence and growth of many new sectors and the
diffusion of the American practice of product engineering as the catalyst for tool design
and the historical trajectory of driving down critical-size dimensions. Precision engineer
ing opened new technological domains. Cities, including Springfield, MA, Providence, RI,
and greater Boston, grew and extended to push the emergent manufacturing belt west
ward.
Thus began the establishment of America’s industrial heartland, commonly described as

the manufacturing belt. Early in the 1900s, in the words of Paul Krugman, ‘geographers
noted the great bulk of U.S. manufacturing was concentrated in a relatively small part of
the Northeast and the eastern part of the Midwest…within the approximate parallelo
gram Green Bay—St. Louis—Baltimore—Portland’ (Krugman, 1991: 11). It proved remark
ably persistent from its peak in 1900 when it is estimated to have accounted for 74 per
cent of America’s manufacturing employment up until 1957 when it still contained 64 per
cent of the nation’s manufacturing jobs. Equally remarkable, until well into the twentieth
century, ‘most of the manufacturing activities outside [the manufacturing belt] were con
centrated in processing of primary products or manufacturing for a very local
area’ (Krugman, 1991: 12‒13).
This was the heartland of American manufacturing which propelled the nation’s industri
al leadership and was a platform for realizing Roosevelt’s Arsenal of Democracy vision. It
encompassed both the ‘American system of manufacturing’, driven by the world’s first
machine-tool industry based on the principle of interchangeability in the nineteenth cen
tury, and the first mass-production enterprises organized by the principle of flow in the
first half of the twentieth century.11
Wages increased with productivity; the public school system was funded by business and
local property taxes; research-intensive universities established by the Morrill Act of 1862
flourished in every state throughout the region and established an educational pathway to
an engineering curriculum in sync with the needs of emerging technologies, fast-growing
Page 19 of 42
ties
enterprises, sectors, and clusters; the machine-tool industry acted as an extra-firm infra
structure working closely with technology-innovative firms but also provided the tooling,
machines, instruments, and equipment for the mass producers required for continuous in
novation; state-chartered banks provided due diligence and patient capital for growing
enterprises, whole sectors, and cluster dynamic processes; and the federal government
designed, engineered, often funded, and organized the building of transportation, energy,
and communication material infrastructures that turned the wheels of a multi-state re
gion, combining a regional production system with its productivity and wages with a
matching demand side in the form of a large and accessible market.
The post-war mythology has been that free markets created America’s industrial
(p. 575)
heartland; but markets do not create, only people and organizations have the power to
create. Groups of people working together build and organize companies and, with the
support of government policies and infrastructures, mutually create industrial ecosys
tems that foster cluster dynamic processes. The bottom line is that firms, individually and
collectively, are the drivers of innovation and production capability development; markets
are the servants of successful economic policymaking, not the masters. Governments are
organizers of the requisite regulatory systems and infrastructures that can turn markets
into useful servants of policy frameworks. Regulation is a bulwark to prevent corporate
political power from capturing government and fostering a business culture dependent
upon subsidization as distinct from an entrepreneurial business culture that fosters inno
vation.
No other part of the nation had cities populated by symbiotically networked groups of
specialist innovative enterprises which collectively and cumulatively advanced a region’s
production capabilities, skill formation processes, and distinctive technology bases. Man
ufacturing enterprises in the rest of the nation were largely concentrated in the process
ing of primary products or producing for a local market. While some were individually
successful, none were embedded in innovative industrial ecosystems that advanced a
region’s production capabilities and skill base.
Krugman and others explain the geographical concentration of manufacturing in the man
ufacturing belt in terms of declining transportation costs and expanding access to an in
creasing national market, but these were the effect of successful industrial and infra
structural policies. Market fundamentalism ignores how important the ways in which the
interrelations between government and the economy are organized is to development.
Historians debate whether it was the states or the federal government that did more to
lay the foundations of US industrial development. It was both together. Lamoreaux and
Wallis (n.d.) conclude that in the mid-western states an ‘extraordinary penetration of the
state and civil society…was often strongly regulatory in its impetus yet sustained by an
unusual degree of legitimacy’. This is what distinguished the governance institutions up
on which the industrial heartland of the United States was built. It was an extraordinary
example of social democracy.
Page 20 of 42
ties
20.5.2 The Sun Belt
In contrast to the Midwest, the South did not attach itself to the manufacturing system
that began along the north-east coast. Emancipation ended the plantation system in 1865
and freed 4 million slaves. The slaves were liberated, but the production system was not
transformed. The land remained concentrated, and the social and political structure of
the plantation economy persisted. New forms of labour coercion were created, and the in
dustrial structure remained one of resource extraction rather than organizational capabil
ity development. What Eric Foner (2011) calls the unfinished (p. 576) revolution was en
trenched along with race hierarchy. The 13th Amendment to the Constitution prohibited
slavery and involuntary servitude but it had an escape clause. Anyone duly convicted of a
crime was exempted from the prohibition. Rapidly, a vast number of ex-slaves became
subject to newly passed local laws such as vagrancy, which included not being able to
show current employment, to create a post-emancipation, peonage labour system com
posed of various forms of involuntary servitude such as penal labour, convict leasing, and
debt bondage.12 The debt-peonage system was dependent on sharecroppers’ lack of nu
meracy. President Lyndon B. Johnson abolished peonage in 1966, which ‘rapidly de
creased sharecropping in every plantation nationwide’.13
Manufacturing was limited to large processing factories in lumber, tobacco, and chemi
cals or to small craft enterprises serving local markets. The ubiquitous metalworking ma
chine shops supplying parts, tooling, and machinery critical to the emergence of the man
ufacturing belt cities were non-existent in the South. Post-Civil War tenant farmers and
sharecroppers, unlike landowning farmers of the North, did not create enough demand to
foster local village or small-town markets and stimulate city-region growth dynamics.
David Meyer estimates that the South’s share of national value added in manufacturing
remained about 8 per cent from 1860 to 1900 (2003: 289). Textiles were the growth ex
ception, increasing from 3 to 10 per cent over the period. During the period only textile
firms relocated into or started in the South in pursuit of large supplies of low-wage
labour. With this exception, the industrial ecosystems and the markets of the Northern
manufacturing cities provided too many advantages for firms to move outside the manu
facturing belt.
Robin Einhorn’s (2006a) classic comparison of the tax systems of the North and the South
has become highly relevant to the present period. From the beginning the Northern
colonies collected taxes for the public provision of material and organizational infrastruc
tures, such as education and the federal armouries, and legislated for a financial regula
tory environment that would support the development of the nation’s nascent manufac
turing capabilities. It evolved into a three-tier system of local, state, and national-level
government, each with tax-raising power to fund public services and otherwise foster in
dustry‒government partnerships, and develop the businesses, employment opportunities,
and communities that came to constitute the nation’s industrial heartland.
The South had an entirely different agenda. In the words of Einhorn (2006b):
Page 21 of 42
ties
Slaveholders had little need for transportation improvements (since their land was
often already on good transportation links such as rivers) and hardly any interest
in an educated workforce (it was illegal to teach slaves to read and write because
slaveholders thought education would help African Americans seize their free
dom). (p. 577) Slaveholders wanted the military, not least to promote the westward
expansion of slavery, and they also wanted local police forces (‘slave patrols’) to
protect them against rebellious slaves. They wanted all manner of government ac
tion to protect slavery, while they tended to dismiss everything else as wasteful
government spending.
The Southern colonies feared that taxes on wealth (i.e. slaves) could be a Trojan horse
threatening the institution of slavery. In the name of states’ rights and freedom from fed
eral government interference and regulations, the South opposed all forms of public pow
er and public decision-making. Manufacturing cities did not develop, even with huge sub
sidies from TVA, and other national infrastructure and low-cost energy supplied by the
federal government. State and local government were not organized to learn from their
rapidly growing Northern counterparts.
By the early decades of the twentieth century, Southern political leaders began to pursue
a strategy of modernizing state government, improving public facilities, and subsidizing
business enterprises to encourage industrial growth (Cobb, 1993: 5‒6). By the 1920s and
1930s states had sanctioned and supervised the sale of municipal bonds to finance indus
trial plant construction. While the granting of subsidies by towns to encourage access to
railroad companies, for example, had long existed, it became the centrepiece of the
region’s industrial policy. It began a new era of competitive subsidization by state and lo
cal governments attracting and adopting ‘footloose’ branches of enterprises from the
manufacturing belt rather than investing in the infrastructures to grow indigenous manu
facturing enterprises (Cobb, 1993). The labour strategy of the South did not include es
tablishing skill-formation institutions, either within business enterprises or public educa
tion. It did not stem the Great Migration in which 6 million African Americans migrated
out of the South between 1916 and 1970.14
The Depression of the 1930s hit hard. In 1938, President Franklin D. Roosevelt declared
that America’s number one economic problem was the continued impoverishment in the
South. The New Deal imposed the most extravagant regional-cum-industrial policy in the
nation’s history and set the stage for the South’s growth during the Second World War
and the post-war decades. In his account of the impact of the New Deal in South Caroli
na, Jack Irby Hayes, Jr writes that the Public Works Administration (PWA)—created under
the National Industrial Recovery Act during the first wave of New Deal legislation—‘liter
ally changed the face of the Palmetto State. Its visible legacy a half-century later included
hundreds of low-cost housing units to replace urban slums, miles of modern highways, a
host of schools, courthouses, hospitals, post offices, and administrative buildings, a thriv
ing shipyard, a number of new sewage and water systems, and two huge hydroelectric
projects’ (Hayes, 2001: 71).
Page 22 of 42
ties
Gavin Wright (1986, 2010) confirms the size of the New Deal’s infrastructural invest
ments in the South. In 1932, one-third of the nation’s urban population drank untreated
water; by 1940, less than 0.2 per cent. Southern cities were disproportionate beneficia
ries of this remarkable upgrade, but prior to the New Deal, many of them (p. 578) lacked
the fiscal capacity or political will to improve basic public services. Altogether, the New
Deal spent more than US$2 billion in the South, much of which supported ‘the types of
services that southern cities would not have provided even in good times’.
Equally important for regional development was the electrification of farms, homes, and
industries. By far the most visible and controversial New Deal electric power programme
was the Tennessee Valley Authority (TVA), created during the first hundred days of legis
lation in 1933. Between 1933 and 1941, the TVA completed seven major dams. Rapid dif
fusion of electricity usage was encouraged not just by lower rates but by subsidized loans
for rural service, initially from the TVA but after 1935 through the Rural Electrification
Administration (REA), primarily to cooperatives. By 1945, 75 per cent of households in
the valley were electrified, compared to just 2 per cent in 1933.15
The South’s post-war industrial strategy was two-pronged: first, to build a regional com
petitive advantage in high-volume, low-cost, price-led production processes to become
dominant in the national market. The business strategy was to attract enterprises and
particularly branches of labour-intensive enterprises primarily from the manufacturing
belt. The New Deal infrastructural investments in transportation, electric power, and
communications were necessary for the success of the business strategy. The large textile
and furniture factories set a pattern for much of subsequent Southern industrialization.
In contrast to the product-led business strategies required to compete in the rapidly
changing complex production industries, the sectors in which the South specialized did
not entail investing in workforce skills, which precluded transitioning to world-class man
ufacturing practices or participatory forms of work organization and which ultimately
made the South vulnerable to off-shoring for low-wage labour.
The second part of the strategy was to build and exercise political muscle in Washington,
DC to attract federal funding concentrating heavily on military installations and bases
from the Department of Defense, the Atomic Energy Commission and its successor the
Department of Energy, and NASA. During the Second World War the WPB gave special
treatment to the South and invested heavily in military bases and production. The South’s
share of military prime contract awards roughly doubled from under 8 per cent in 1951 to
roughly 16 per cent in the years 1958‒63 to nearly one-quarter in the 1970s (Schulman,
1994: 140, table 6.1). In the words of Schulman, ‘Defense dollars permeated nearly every
town in the region’ (1994: 141), and by 1973 more Southerners worked in defence-relat
ed industries than in textiles, synthetics, and apparel combined.
The South was hugely successful in attracting defence contractors. ‘During the 1970s,
the New York Times reported that defence was the single largest employer in four south
ern states, outpacing agriculture, textiles, lumber and all the others. And of the top ten
US defence contractors, seven, including Lockheed, McDonnell-Douglas, General Dynam
Page 23 of 42
ties
ics, and Rockwell operated large installations in the South’ (Schulman, 1994: 142). Mari
etta, Georgia, the home of Lockheed-Georgia, the single largest (p. 579) business organi
zation in the Southeast in the 1960s and 1970s, employed workers in fifty-five of
Georgia’s 159 counties (Schulman, 1994: 141). But the strategy was not limited to mili
tary contracts. The South long played the role of commissary to the nation’s armed
forces, providing textiles, tobacco, food, and coal.
The South had become fortress Dixie and the Sun Belt became the southern periphery of
the Gun Belt.
20.5.3 The Gun Belt
The United States transitioned from a virtually non-existent pre-war weapons industry to
a defence budget larger than the defence budgets of all other nations combined.15
Defence expenditures remained over 10 per cent of GDP into the 1960s and over 5 per
cent until the end of the Cold War.16 Harry Magdoff and Paul Sweezy (1975) argued that
the success of the New Economics of Keynesian demand-management theory rested heav
ily on military expenditures. Out of the total federal purchases of US$1.5 trillion between
1946 and 1971, US$1.1 trillion were defence-related spending (including debt payments
for previous wars, veterans’ benefits, space exploration, and the Atomic Energy Commis
sion). Without this spending, they argue, unemployment rates in the post-war period
would have approximated those of the Great Depression.
Geographically, the Gun Belt forms a periphery along the east and west coasts and across
the South (Markusen et. al., 1991; Markusen and Yudken, 1992; Wright, 2020). The Gun
Belt is a composite of two types of industrial policy. The first involves large R&D expendi
tures to design and develop technologically advanced weapons systems. It has fostered
high-tech regional innovation systems consisting of research-intensive universities, open-
system business models, and ‘manufactories of new sectors’ (Best, 2015), of which Silicon
Valley and Greater Boston are the leading examples. Other cities or regions that have
leveraged defence department R&D to build research-intensive universities and develop
distinctive technology bases include the North Carolina Research Triangle, Los Angeles,
San Diego, and Seattle.17 Federal funds are complemented by city and state government
infrastructural support to (p. 580) establish industrial ecosystems that foster localized
technology and enterprise capability development.18
The second form of defence-driven industrial policy is purchasing contracts for supply of
the myriad provisions required to support the various arms and activities of the defence
department at home and abroad. The major labour-intensive industries of the South, such
as textiles and furniture, have long-established supply relationships to the government to
complement commercial sales. But the production system is that of the South, which spe
cializes in commodity products for which design changes are rare and investment in
skilled labour is not required.
Page 24 of 42
ties
The South has attracted Gun-Belt federal funding for high-tech programmes, most no
tably NASA in Florida and Houston, and research universities in Atlanta, Austin, and San
Antonio. But the region has not established cities that foster indigenous innovative enter
prises and the cluster growth dynamics that historically populated the North and built the
nation’s industrial heartland.
In terms of employment, the Gun Belt became as important to the US economy and indus
trial policy in the post-war era as the manufacturing belt and industrial policy that fos
tered manufacturing infrastructures and nurtured city growth dynamics had been to the
pre-war and wartime economy. Moreover, the high-tech regions have been hugely suc
cessful in advanced technology weapons systems and creating new spin-off information
technology sectors. Nevertheless, despite the huge defence expenditures on R&D and
high-tech business development, and other military spending, manufacturing’s share of
total US employment dropped from close to 25 per cent in 1970 to 9 per cent in 2011
(Baily and Bosworth, 2014). What the Gun Belt has not done is to create high-paying man
ufacturing employment on the scale of the manufacturing belt in the United States’ indus
trial heartland.
20.5.4 The Rust Belt
The 1970s were a turning point in US industrial leadership and policymaking. The manu
facturing belt became the Rust Belt.19 Factory employment in the nation’s manufacturing
sector declined by one-third between 1969 and 1996.
In an ironic twist with tragic consequences, the industrial heartland in the United States,
where management and labour worked together to meet the production targets to win
the Second World War, became the victim of the nation’s success. At exactly the time
when industrial policy was most necessary to address the challenge of restructuring the
nation’s manufacturing system, US political leadership and economic policymakers chose
to look the other way. The fate of the US industrial heartland was (p. 581) subordinated to
the pursuit of the United States’ new political role in the global economy. It was an oppor
tunity seized upon by the South’s political leaders to pursue a campaign to dismantle tax,
regulatory, and labour policies and the manufacturing infrastructures that had long been
integral to the nation’s industrial heartland.
The political and business leadership of the South has never accepted the production or
ganization challenge of Henry Ford to make changes in the organization of production,
engineering principles, and work practices to advance productivity and deliver the wages
upon which the US middle class was built. The New Deal came up against the hierarchi
cal social structure of the South and its dependence on maintaining the repressive labour
systems and the political power of the beneficiaries of the extractive industries upon
which vast fortunes had been built that had endured for centuries. History offers few if
any examples of the landowning elites of plantation economies establishing transforma
tive policy frameworks. New territories were developed and new types of extractive in
Page 25 of 42
ties
dustries have been organized, but the autocratic, exploitative business model is a con
stant.
The first signs of what became a long-term de-industrialization process were in the 1950s
when the fortunes of the cities that constituted the manufacturing belt began to suffer
from declining tax bases, which came to a head in the fiscal crisis of big-city governments
in the 1970s (O’Connor, 1973; Best and Connolly, 1982 [1976]; Bluestone and Harrison,
1982).
Historian Jon Teaford examined twelve large cities in the manufacturing belt, each of
which had a population of over a half a million in 1950. None of these cities had lost pop
ulation before the decade of the 1930s and most had enjoyed the reputation of a boom
town, doubling in population every ten years. All these cities, except for New York, de
clined in population during every decade from 1950 to 1980 (Teaford, 1990: 2). Only
Boston was later to recover as the region became known as the Rust Belt. All other cities
undertook industrial policies to revitalize, but to little effect. Larger forces were at play.
Teaford states:
It was a gradual slide from a position of economic supremacy, that slide beginning
prior to World War II and persisting throughout the remainder of the century. Mid
western cities no longer were economic miracles as in bygone days…at the close
of the 1970s they all feared an emerging vision of the interior as void… Midwest
ern cities were on the defensive, and their economic and political leaders would
embody every tactic, offer every lure, and unleash every power at their command
to keep a corporate headquarters, revive an aging steel plant, or halt the slow
death of downtown retailing.
(Teaford, 1993)
The de-industrialization dynamics have not been limited to the North. The United States
lost one-third of its manufacturing employment in the decade of the 2000s but this time it
was disproportionately in small towns of the rural South. David Carlton and Peter Cocla
nis (2018) cite the example of North Carolina, the leading manufacturing state of the
South. In the first decade of this century, North Carolina’s manufacturing employment
dropped by 44 per cent, from 761,000 to 429,000. Two of North (p. 582) Carolina’s major
industries suffered the most: textiles lost 70 per cent of its employment between 1998
and 2010, and furniture lost nearly 60 per cent.
20.6 Neo-liberalism and Dismantling the ‘Amer

ican System’
The 1970s marked the beginning of the end of the tradition of the anti-trust and pro-regu
lation agenda based on the distrust of concentration of political and economic power go
ing back to Thomas Jefferson and James Madison, the framers of the Constitution and the
authors of Federalist Paper No. 10. They argued the nation could have democracy or con
Page 26 of 42
ties
centrated wealth, but not both. While corporations barely existed at the time of the ratifi
cation of the United States Constitution and Bill of Rights, stark warnings about the risks
to the survival of democratic institutions of corporations with unaccountable power en
gaging in political activities were articulated by Jefferson and Madison a quarter-century
after their ratification.
In 1816, Jefferson, second president of the United States, urged that we ‘crush in its birth
the aristocracy of our monied corporations which dare already to challenge our govern
ment to a trial of strength, and bid defiance to the laws of our country’. Madison, fourth
president and also hailed as the ‘father of the Constitution’, was equally blunt: ‘There is
an evil which ought to be guarded against in the indefinite accumulation of property from
the capacity of holding it in perpetuity by ecclesiastical corporations…the growing wealth
acquired by them never fails to be a source of abuses’ (Whitehouse and Stinnett, 2017: 4‒
5).
In the time of Jefferson and Madison, the challenge of political dominance of government
by ‘monied corporations’ had only begun. By the Gilded Age of the last decades of the
1800s the power of corporations to foster economic prosperity and to capture govern
ment were both operating in high gear. But the power of the ‘robber barons’ to control
government and subvert democracy did not last. The resulting increase in inequality and
economic hardship provoked the progressive era in US politics.
President Theodore Roosevelt engineered the passage of the Tillman Act in 1907 which
outlawed corporate contributions to politics.20 In 1913, President Woodrow Wilson passed
into law the Federal Reserve Act which was to supervise banks and had the dual mandate
of maximizing employment and keeping inflation low, in 1914 the Clayton Antitrust Act
was passed prohibiting anti-competitive mergers, predatory and discriminatory pricing,
and other forms of unethical corporate behaviour, and the (p. 583) Federal Trade Commis
sion was created also in 1914 to protect consumers. President Franklin Roosevelt signed
the United States Banking Act (Glass‒Steagall) in 1933, which decentralized and restruc
tured the banking industry, including the separation of commercial banks from invest
ment banks and securities dealers; and the National Labor Relations Act (Wagner) in
1935, which guarantees the right of private-sector employees to organize into trade
unions, to engage in collective bargaining, and to take collective action such as strikes.
His administration also launched anti-trust investigations into ‘housing, construction, tire,
newsprint, steel, potash, sulphur, retail, fertilizer, tobacco, shoe, and various agricultural
industries’ (Stoller, 2016).21 The Celler–Kefauver Act was passed in 1950 to reform and
strengthen the Clayton Antitrust Act of 1914, which had amended the Sherman Antitrust
Act of 1890.The anti-trust tradition dating back to Thomas Jefferson lasted for the first
three post-Second World War decades.
Matt Stoller (2016) dates the dismantling of the regulation tradition to the alliance of the
post-Watergate generation of congressional Democrats with the traditional anti-regula
tion Republicans which in 1975 combined to overthrow Wright Patman, the chairman of
Page 27 of 42
ties
the US House Committee on Banking and Currency. A former cotton tenant farmer, he
had served in Congress since 1929. In the words of Stoller:
It was a revolution and signalled the end of an era in which banks and big busi
ness were largely kept under control. In 1936 Patman authored the Robinson–Pat
man Act, followed by the Bank Secrecy Act, successfully defended Glass–Stea
gall…authored the Employment Act of 1946 and initiated the first investigation in
to the Nixon administration over Watergate. With Supreme Court Justice Louis
Brandeis…the New Dealers drafted legislation into legally actionable ideas to for
malize the sentiment that big business and democracy were rivals. ‘We may have
democracy, or we may have wealth concentrated in the hands of the few,’ Brandeis
said, ‘but we can’t have both.’
(Stoller, 2016)
President Jimmy Carter deregulated the trucking, banking, and airline industries with the
appointment of Alfred Kahn as ‘anti-inflation czar’. President Reagan went further; the
dismantling of regulation became his major legacy and heralded the end of regulatory
policies to constrain economic and political power. President Clinton joined the deregula
tion crusade and oversaw rapid increases in monopoly and accompanying political power
in banking, media, oil, and telecommunications.
Not by coincidence, the 1970s marked the beginning of a shift in economic discourse
from the mixed economy of Paul Samuelson’s economics textbooks to the policymaking
dominance of market fundamentalism. Alan Greenspan captured the essence (p. 584) of
market fundamentalism: ‘The ultimate regulator of competition in a free economy is the
capital market.’22 He became the nation’s single most powerful economic policymaker.
From 1974 to 1977 Greenspan was chairman of the Council of Economic Advisors under
President Gerald Ford before becoming chair of the Federal Reserve Bank of the United
States from 1987 to 2006. As observed by Stoller: ‘The following year the financial crisis
exploded the myth that the nation’s deregulated financial institutions were the guaran
tors of economic progress. The financial bailouts also illustrated their political power.’
Perhaps an even greater measure of the myth of de-regulation to foster economic pros
perity has been the explosive growth in inequality and the hollowing out of the nation’s
manufacturing capabilities in recent decades. In the 1960s, a period of widespread pros
perity, the ratio between executive pay and worker pay was 18:1; it reached over 400:1 in
2000. In 1982, the Securities and Exchange Commission issued a new rule—10b-18 of the
Securities Exchange Act. Historically, stock buybacks by corporate managers were illegal.
The new rule enabled corporate managers to purchase large quantities of their
company’s stock without being accused of stock-price manipulation. Buybacks reward in
vestors and CEOs only. The consequences for workers and their communities have been
devastating. In a study of 454 companies listed continuously in the S&P 500 between
2004 and 2013, William Lazonick found stock buybacks consumed 51 per cent and divi
dends 35 per cent of net income. In his words: ‘Excluding the recession years 2001 and
2008, buybacks and dividends taken together have averaged 85 per cent of net earnings
Page 28 of 42
ties
for all corporations since 1998’ (2014). This leaves little for R&D, investing in production
capabilities, new product development, technology management, worker training, worker
benefits, worker salaries.
The corporate business model, by which the nation established global leadership in manu
facturing, was swept away and the engineering culture and worker skills upon which it
depended went with it. Put simply, the value-creating business system gave way to a val
ue-extracting business system organized to drive up CEO salaries and shareholder re
turns. Business enterprises were financialized as activist investors, aka vulture capital
ists, who buy a controlling stake in a corporation, cut jobs and other costs to drive up the
share price, and get out leaving a weakened business behind. They ignore the long-term
work of building a company and all too often destroy it instead.
The destructive consequences of neo-liberalism for the common good have been obscured
by billionaire-funded foundations and thinktanks to shape economic discourse.23 The
American Enterprise Institute, the Cato Foundation, the Heritage (p. 585) Foundation, and
the Olin Foundation are major benefactors and champions of billionaire-funded neo-liber
al economic discourse.
The economics they espouse did not alert the nation to the destructive consequences of
the neo-liberalism. The reality is that the nation’s once powerful manufacturing enterpris
es were left ill equipped to compete against production systems organized around world-
class manufacturing principles and self-directed work teams as in Japan, co-determina
tion models for building innovative business enterprises as in Germany, or the latecomer
advantage successfully implemented by Chinese policymakers starting at the bottom but
undertaking a business development strategy organized to move up the production capa
bility spectrum (Best, 2018).
We turn next to an alternative economics perspective and policy framework that draws
lessons from the building of the nation’s industrial heartland, the building of the Arsenal
of Democracy, as well as successful post-war transformative experiences elsewhere.
20.7 The Capability Triad as Policy Framework

The catastrophic consequences of four decades of neo-liberal policymaking are becoming
increasingly apparent. The once world-leading productive structures of the US economy
have undergone perhaps the sharpest organizational erosion in history. The challenge of
overcoming deficits in production capabilities, worker skills, infrastructures, and industri
al ecosystems to reclaim the American Dream are formidable, even without pandemics
and climate change.
The first requirement is an objective assessment of the state of the nation’s economy. The
second requirement is to learn lessons from historical experiences of successful economic
transformations. These are called economic miracles, and it will take a transformative vi
sion and political leadership on the economic ‘miracle’ scale to address the combined
Page 29 of 42
ties
challenges of pandemics, climate change, and nearly half a century of neo-liberalism. Be
sides Franklin Roosevelt and the Second World War, examples of such leadership include
the work of Deng Xiaoping (China), Taiichi Ohno/Edwards Deming (Japan), Ludwig Er
hard (West Germany), Ralph Flanders (Massachusetts), and Kenneth Whitaker (Ireland)
(Best, 2018). These are all architects of strategic policy frameworks that have trans
formed economies facing serious challenges.
I use the capability triad as an expository device to characterize common policy frame
work lessons. It asserts the following proposition:
Rapid growth can only be achieved through the careful pursuit of interconnected
strategies in each of three domains: production capabilities, business organiza
tion, and skill formation. The three domains are not separable and additive compo
nents of growth, but mutually interdependent sub systems of a single developmen
tal (p. 586) process. No one of the three elements of the capability triad can con
tribute to growth independently of mutual adjustment processes involving all
three elements. (Best, 2001: 56; Best, 2018)
Roosevelt’s Arsenal of Democracy policy framework is informative with respect to the fail
ure of the New Deal. The latter was based on a Keynesian-like deficient demand econom
ics in which increased spending and infrastructure investments would expand demand
and incentivize business to increase production. Like all market-centric economics, the fo
cus did not go inside the organization of production and business organization to focus on
capability development; the focus instead was on taxes and subsidies without regard to
the structures of production and business organization.
President Roosevelt’s vision of creating an Arsenal of Democracy to win the war against
fascism called for a policy framework to achieve an order of magnitude increase in output
without an increase in the civilian labour force. Unlike the New Deal, it was anchored in a
policy framework informed by the critical interdependence of mutual adjustment process
es linking the three domains of production, business, and skills.
The Arsenal of Democracy policy framework drew upon production engineers of the
nation’s enterprises with the most advanced production capabilities. It required process
engineering leaders in organizational innovation like Charles Sorensen of Ford Motor
Company (production principle of flow), William Knudsen of General Motors, and the se
nior leaders of skill training programmes at leading companies, to develop and implement
the organizational changes in management practices needed to transform the nation’s
productive structures and double output in under five years. It also required the creation
of the Defense Plant Corporation as a subsidiary of the New Deal’s Reconstruction Fi
nance Corporation, ‘with the powers as it may deem necessary to aid the Government of
the United States in its national defense program’ (White, 1949: 161).
The capability triad extends New Deal policymaking in a second major way. It broadens
the policy framework from material and social infrastructures such as transportation, en
ergy, water supply, communication, social housing, and public health and sanitation to or
Page 30 of 42
ties
ganizational infrastructures critical to business development, including access to human
capital, long-term finance, science and technical expertise, and a machine tool/software
engineering sector. I refer to the latter as extra-firm infrastructures or more narrowly
manufacturing infrastructures. All have suffered as a consequence of neo-liberalism.24
Roosevelt learned from the New Deal experience that material infrastructure investment
is not enough to foster economic transformation. The TVA programme, the (p. 587) biggest
ever regional development policy in the United States, is an example. It succeeded as ma
terial infrastructure policy but even with massive investments in electrification, trans
portation, clean water, and communication it did not galvanize the emergence of a critical
mass of capability-developing business enterprises and thereby industrial development.
The whole region did not produce a single import-replacing and export-generating city
(Jacobs, 1984: 117). The South continued to import virtually all the manufactured prod
ucts it had historically imported. It failed as industrial development and innovation
policy.25
The United States’ Second World War experience provides examples of extra-firm infra
structures designed and coordinated to target productive structure changes within and
across regional or national populations of enterprises. The experience is an exemplar of
industrial policies that focus on enterprise change methodologies to advance production,
new product development, and technology management capabilities at the macroeconom
ic level.
The German post-war economic miracle is an exemplar of industrial policy in the form of
local and state governments as organizers of industrial ecosystems in which small and
medium-sized business enterprises are supported by a range of capability-enabling infra
structures (Best, 2018). Local governments have the local knowledge to coordinate, fo
cus, and galvanize participation in the historically specific range of infrastructural agen
cies required to drive holistic organizational change cutting across the three domains of
production, business, and skill formation. At the same time, sustained growth depends
upon national governments providing budget support for physical and organizational in
frastructure, ideally by investing in and nurturing regional industrial ecosystems in which
a whole host of innovation dynamics characterized by Babbage, Marshall, Young, Pen
rose, Schumpeter, Richardson, and Jacobs can be found (Best, 2018).
The crafting of strategic policy frameworks has both a capability-informed microeconom

ic dimension with respect to industrial and regulatory policies and a capability-informed
macroeconomic policy dimension including stabilization, exchange rate, and interest-rate
policies. A common feature of rapid growth experiences is that government stabilization
policy is subservient to capability development policy. Manufacturing performance is
highly sensitive to the pursuit of stabilization policies that are subservient to capability
development policies, and the failure to do so is a key feature differentiating post-war pol
icymaking in the United States from that of her major manufacturing rivals. The tragedy
is that while the industrial policy by which the US leadership in production was created
Page 31 of 42
ties
has not been appreciated within the United States, it has provided lessons which latecom
er nations have put to good use.
When an economic system is trapped in a fragmented, low-level capability triad, govern

ment leadership is required to craft a strategic policy framework, outline a roadmap of
the way ahead, invest in the requisite extra-firm infrastructures, and (p. 588) galvanize the
energies of all those required to progress the roadmap, transition to a more advanced ca
pability triad, and thereby transform the production system. This is not what markets do.
It is what government leadership has, in special cases, organized with outstanding suc
cess. It offers the only way to address the climate-change challenge in the time required
to make it happen.
20.8 Appendix: Major Contributors to the Capa

bility and Innovation Perspective
Production as the sphere of technological innovation became a subject of political econo
my with the publication in 1832 of Charles Babbage’s On the Economy of Machinery and
Manufactures.26 He went beyond Smith’s rhetorical pin factory to explore the interiors of
the most technologically innovative enterprises of his day.
Babbage introduced a research methodology of systematic observation of change to polit

ical economy.27 He undertook, in his words, an ‘examination of the workshops… which
contain within them a rich mine of knowledge’ (Babbage, 2005 [1832]: 4) to understand
the ‘processes of manufactures’ and to apprehend their ‘general principles and mutual
relations’ (Babbage, 2005 [1832]: 3‒4). He distinguished ‘mechanical principles’ from the
principles of political economy which ‘seemed to pervade many establishments’. But at
the same time, he found that the ‘interior economy of factories’ and questions of political
economy were ‘so interwoven…that it was deemed inadvisable to separate the two
subjects’ (Babbage, 2005 [1832]: 7). Unfortunately, as with his drawings of the first com
puter, Babbage’s contribution to political economy has long been neglected and only re
discovered in recent times.28
Joseph Schumpeter paid tribute to Babbage’s book more than a century after its publica
tion as follows:
This work which was widely used (also by Marx), is a remarkable performance by
a remarkable man. Babbage…was an economist of note. His chief merit was that
he combined a command of simple but sound economic theory with a thorough
first- (p. 589) hand knowledge of industrial technology and of the business proce
dure relevant thereto. This almost unique combination of acquirements enabled
him to provide not only a large quantity of well-known facts but also, unlike other
writers who did the same thing, interpretations. He excelled, amongst other
things, in conceptualization, his definitions of a machine and his conception of in
vention are deservedly famous.
Page 32 of 42
ties
(Schumpeter, 1954: 541, cited in Hyman, 1982: 121)29, 30
Schumpeter made a distinction between price-led and product-led competition and

stressed the necessary role of credit to invest in capability development. Babbage distin
guished types of manufacturing systems and the role of principles of production, distin
guished technology from science, and stressed the critical role of a policy framework in
promoting innovation capabilities. Both insisted on the interconnectedness of production
and business organization and together characterized the mediating role of extra-firm in
frastructures in fostering the building of successful manufacturing enterprises: in
Babbage’s case, government investment in science and technology, and in Schumpeter’s,
banking institutions for investment for the creation of ‘future values’, innovation, and ca
pability development. Failure to do so, as in the case of the United Kingdom in Babbage’s
day, would cause the prosperity of the nation to suffer from a loss of competitive advan
tage and markets.
The idea of a production system extends economics beyond the firm and the form of com
petition to the dynamics of industrial change. Alfred Marshall’s Principles of Economics,
first published in 1890, extended Babbage into an emergent theory of industrial organiza
tion.31 In a celebrated passage, Marshall described economies arising from an increase in
the scale of production as falling into ‘two classes—those dependent on the general devel
opment of the industry, and those dependent upon the individual houses of business en
gaged in it and the efficiency of their management; that is into external and internal
economies’ (1920 [1890]: 221 and repeated on 262). External economies ‘result from the
growth of correlated branches of industry which mutually assist one another’ (Marshall,
1920 [1890]: 264). With the inclusion of organization and inter-firm dynamics as a vari
able, Marshall’s ‘law of increasing (p. 590) returns’ introduces a growth dynamic of cumu
lative increasing returns, later advanced by Thorstein Veblen, Allyn Young, and Gunnar
Myrdal.
Edith Penrose went inside the business enterprise to distinguish resources that are pur
chased in the market from productive resources that require experience and teamwork to
develop (1959). The growing firm develops distinctive productive resources or capabili
ties to meet emergent market opportunities, but success not only produces new products
but also releases resources that can be put to work identifying a new round of opportuni
ties. This is the learning theory of the growth of the firm.
Jane Jacobs asked the questions: Why do cities grow? Why did today’s major cities under
go a period of explosive growth? Why do some cities continue to grow over a long period
while others go into decline? She examines cities historically, giving special attention to
cases of rapid growth to discover patterns of complex interactions in their most pro
nounced forms. She sees cities as the engines of economic advance, providing markets,
jobs, capital, and technology for themselves, the regions around them, and other cities as
well.
Page 33 of 42
ties
Jane Jacobs’ The Economy of Cities was published in 1969. Rather than celebrating eco
nomic efficiency, Jacobs celebrates a city’s growth dynamics as expressed in the rate of
addition of new goods and services. Sustained city growth is simultaneously a process of
increasing differentiation of skills and an experimental process of new product develop
ment and sector evolution. In her words: ‘Existing divisions of labor multiply into more di
visions of labor by grace of intervening added activities that yield up more sums of work
to be divided’ (Jacobs, 1969: 58). The ‘intervening added activities’ are described in terms
of new ‘work’ combined with multiple trials and errors linking the old to the new divisions
of labour. The increasing differentiation in skills increases the opportunities for innova
tion and for sustained city growth.
Acknowledgements
I wish to thank Eva Paus and Lindsay Whitfield for extensive comments on an early draft
of this chapter. As always, I owe a debt to John Bradley for feedback and advice along the
way. I am most grateful to Rick Best, Carol Heim, and Gavin Wright for sharing their ac
cumulated experience, knowledge, and wisdom in conversations over these matters. They
have offered insights, ideas, references, and corrections that I have incorporated
throughout the chapter. We all share a passion for the issues, but I alone am responsible
for the claims being made and the shortcomings that will inevitably have to be addressed.
References
Acemoğlu, Daron and David Autor (2011) ‘Skills, Tasks, and Technologies: Implications
for Employment and Earnings’, in Orley Ashenfelter and David Card (eds) Handbook of
Labor Economics, Vol. 4B. Amsterdam: Elsevier, pp. 1043‒171. (p. 591)
Atkinson, Robert D., Luke Stewart, Scott Andes, and Stephen Ezell (2012) ‘Worse Than
the Great Depression, What Experts Are Missing about American Manufacturing Decline’.
Washington, DC: The Information Technology & Innovation Foundation.
Babbage, Charles (2005 [1832]) On the Economy of Machinery and Manufactures.

Cirencester: Echo Library.
Baily, Martin and Barry Bosworth (2014) ‘US Manufacturing: Understanding its Past and
its Potential Future’, Journal of Economic Perspectives 28(1): 3–26.
Best, Michael H. (1972) ‘Notes on Inflation’, The Review of Radical Political Economics
IV(4): 85‒112.
Best, Michael H. (1990) The New Competition: Institutions of Industrial Restructuring.

Best, Michael H. (2001) The New Competitive Advantage: The Renewal of American In
dustry. Oxford: Oxford University Press.
Page 34 of 42
ties
Best, Michael H. (2011) ‘Review of Internet Alley: High Tech in Tysons Corner, 1945–2005,
Cambridge: MIT Press, by Paul Ceruzzi’, in Business History 53(6): 976–80.
Best, Michael H. (2015) ‘Greater Boston’s Industrial Ecosystem: A Manufactory of Sec

tors’, Technovation 39–40: 4–13.
Best, Michael H. (2020) ‘The Economics of Innovation behind Cluster Dynamic Process
es’, in Justin Yifu Lin and Arkebe Oqubay (eds) Oxford Handbook of Industrial Hubs and
Best, Michael H. and William E. Connolly (1982 [1976]) The Politicized Economy. 2nd edi
tion. Lexington, MA: D. C. Heath.
Bipartisan Policy Center (2016) ‘America’s Aging Water Structure’. Washington, DC. Avail
able at https://bipartisanpolicy.org/wp-content/uploads/2019/03/BPC-Aging-Water-
Infrastructure.pdf.
Blanchflower, David (2019) Not Working: Where Have All the Good Jobs Gone? Princeton,
NJ: Princeton University Press.
Bluestone, Barry and Bennett Harrison (1982) The Deindustrialization of America: Plant
Closings, Community Abandonment, and the Dismantling of Basic Industry. New York: Ba
sic Books.
Carlton, David and Peter Coclanis (2018) ‘The Roots of Southern Deindustrialization’,
Challenge 61(5–6): 418–26.
Caro, Robert (1982) The Path to Power: The Years of Lyndon Johnson. New York: Knopf.
Chandler, Alfred (1977) The Visible Hand. Cambridge, MA: Harvard University Press.
Cobb, James (1993) The Selling of the South: The Southern Crusade for Industrial Devel
opment 1936–1990. 2nd edition. Urbana, IL: University of Illinois Press.
Einhorn, Robin (2006a) American Taxation, American Slavery. Chicago, IL: University of
Chicago Press.
Einhorn, Robin (2006b) ‘Tax Aversion and the Legacy of Slavery’. Available at https://
www.press.uchicago.edu/Misc/Chicago/194876.html p18.
Foner, Eric (2011) The Fiery Trial: Abraham Lincoln and American Slavery. New York: W.
W. Norton.
Frey, Carl (2019) The Technology Trap. Princeton, NJ: Princeton University Press.
Page 35 of 42
ties
Hayes, Jack I. (2001) South Carolina and the New Deal. Columbia, SC: University of South
Carolina Press. (p. 592)
Hounshell, David A. (1996) ‘The Evolution of Industrial Research in the United States’, in
Richard S. Rosenbloom and William J. Spencer (eds) Engines of Innovation: U.S. Industrial
Research at the End of an Era. Boston, MA: Harvard Business School Press, pp. 13–86.
Hyman, Anthony (1982) Charles Babbage: Pioneer of the Computer. Oxford: Oxford Uni
versity Press.
Jacobs, Jane (1969) The Economy of Cities. New York: Random House.
Jacobs, Jane (1984) Cities and the Wealth of Nations. New York: Vintage.
Jones, Jesse H. and Edward Angly (1951) Fifty Billion Dollars: My Thirteen Years with the
RFC (1932–1945). New York: The Macmillan Company.
Kevles, Daniel J. (1978) The Physicists: The History of a Scientific Community in Modern
America. New York: Knopf.
Kline, Stephen (1985) ‘Innovation is Not a Linear Process’, Research Management 28

(Jul–Aug): 36–45.
Krugman, Paul (1991) Geography and Trade. Cambridge, MA: MIT Press.
Lamoreaux, Nancy R. and John J. Wallis (n.d.) ‘General Laws and the Mid-nineteenth-cen
tury Transformation of American Political Economy: Massachusetts, New York, Indiana,
and Beyond’, National Bureau of Economic Research.
Lazonick, William (1991) Business Organization and the Myth of the Market Economy,
Lazonick, William (2014) ‘Profits without Prosperity’, Harvard Business Review, Septem
ber: 47‒55.
Lind, Michael (2003) Made in Texas: George W. Bush and the Southern Takeover of Amer
ican Politics. New York: Basic Books.
Madrigal, Alexis C. (2019) ‘The Toxic Bubble of Technical Debt Threatening America’, At
lantic, 29 October.
Magdoff, Harry and Paul Sweezy (1975) ‘Capitalism and Unemployment’, Monthly Review
June: 1–14.
Markusen, Ann and Joel Yudken (1992) Dismantling the Cold War Economy. New York:
Basic Books.
Page 36 of 42
ties
Markusen, Ann, Peter Hall, Scott Campbell, and Sabina Deitrick (1991) The Rise of the
Gunbelt: The Military Remapping of Industrial America. New York: Oxford University
Press.
Marshall, Alfred (1920 [1890]) Principles of Economics. 8th edition. London and New
York: Macmillan.
Marx, Leo (1998) ‘The Machine’, in Merritt Roe Smith and Gregory Clancey (eds) Major
Problems in the History of American Technology: Documents and Essays. Boston, MA:
Houghton Mifflin, pp. 120‒30.
Mathews, Jessica T. (2019) ‘America’s Indefensible Defense Budget’, The New York Re
view of Books, 18 July.
Mayer, Jane (2016) Dark Money: How a Secretive Group of Billionaires is Trying to Buy
Political Control in the US. London: Scribe Press.
Meyer, David R. (2003) The Roots of American Industrialization. Baltimore, MD: Johns
Hopkins University Press.
Nieburg, H. L. (1966) In the Name of Science. Chicago, IL: Quadrangle Books.
O’Connor, James (1973) The Fiscal Crisis of the State. New York: St. Martin’s Press.
Penrose, Edith (1959) The Theory of the Growth of the Firm. Oxford: Basil Blackwell and
New York: John Wiley and Sons. Revised edition 1995. Oxford: Oxford University Press.
(p. 593)
Phillips-Fein, Kim (2009) Invisible Hands: The Businessmen’s Crusade against the New
Deal. New York: W.W. Norton & Company.
Piketty, Thomas, Emmanuel Saez, and Gabriel Zucman (2018) ‘Distributional National Ac
counts: Methods and Estimates for the United States’, Quarterly Journal of Economics
133(2): 553‒609.
Pisano, Gary and Willy Shih (2012) Producing Prosperity: Why America Needs a Manufac
turing Renaissance. Boston, MA: Harvard Business Review Press.
Schulman, Bruce (1994) From Cotton Belt to Sunbelt. Durham, NC: Duke University
Press.
Schumpeter, Joseph (1954) History of Economic Analysis. London: Allen and Unwin.
Shermer, Elizabeth (2013) Sunbelt Capitalism: Phoenix and the Transformation of Ameri
can Politics. Philadelphia, PA: University of Pennsylvania Press.
Smith, Merritt Roe and Gregory Clancey (eds) (1998) Major Problems in the History of
American Technology: Documents and Essays. Boston, MA: Houghton Mifflin.
Page 37 of 42
ties
Stoller, Matt (2016) ‘How Democrats Killed their Populist Soul’, The Atlantic, 24 October.
Teaford, Jon C. (1990) The Rough Road to Renaissance: Urban Revitalization in America,
1940–1985. Baltimore, MD: Johns Hopkins University Press.
Teaford, Jon C. (1993) Cities of the Heartland: The Rise and Fall of the Industrial
Midwest. Bloomington, IN: Indiana University Press.
White, Gerald T. (1949) ‘Financing Industrial Expansion for War: The Origin of the De
fense Plant Corporation Leases’, The Journal of Economic History 9(2): 156–83.
Whitehouse, Sheldon and Melanie Wachtell Stinnett (2017) Captured: The Corporate Infil
tration of American Democracy. New York: The New Press.
Winchester, Simon (2018) Exactly: How Precision Engineers Created the Modern World.
London: William Collins.
Wright, Gavin (1986) Old South, New South: Revolutions in the Southern Economy since
the Civil War. New York: Basic Books.
Wright, Gavin (2010) ‘The New Deal and the Modernization of the South’, Federal
History, Jan. Issue 2.
Wright, Gavin (2020) ‘World War II, the Cold War, and the Knowledge Economies of the
Pacific Coast’, in Mark Brilliant and David Kennedy (eds) World War II and the West It
Wrought. Stanford, CA: Stanford University Press, pp. 74–99.
Notes:
(1) After 2000, the absolute level of US manufacturing employment fell by one-third to un
der 12 million in 2012 (Atkinson et al., 2012: figure 2). In the 2000s, US manufacturing
suffered its worst performance in US history in terms of employment. Not only did the
United States lose 5.7 million manufacturing jobs, but the decline as a share of total man
ufacturing employment (33 per cent) exceeded the rate of loss in the Great Depression
(Atkinson et al., 2012).
(2) Labour productivity grew at an average rate of 2.1 per cent over decades but fell to
1.2 per cent in the mid-2000s and 0.6 per cent since 2011. In the mid-1970s, the US over
all trade balance began a decline from a surplus of approximately 1 per cent gross do
mestic product to a deficit of –3 per cent in 1986, followed by a recovery before plunging
again to a deficit of between 3 and 5 per cent in the 2000s (Pisano and Shih, 2012: 4).
(3) The market failure approach offers policy guidelines in the form of taxes and subsidies
to correct for deviations of the prices and outputs from an idealized perfectly competitive
market system. For economic modelling the attraction is that the theory is context- and
history-blind and assumes that production is subject to the same optimality rules and
Page 38 of 42
ties
equilibrium conditions as the world of consumption. If only real-world production systems
cooperated.
(4) Elizabeth Shermer’s Sunbelt Capitalism: Phoenix and the Transformation of American
Politics offers a similar account of the civil war of policy frameworks (2013).
(5) The TVA became a World Bank development model around the world long after its
flaws were well known (Jacobs, 1984).
(6) Just-in-time, total quality management, and self-directed work teams.
(7) The extended global reach of the Pentagon was part of the process. In the 1920s, the
US armed forces were stationed in only three countries abroad. During the Second World
War the number grew to thirty-nine. By the 1960s the United States had over 2,900 mili
tary bases spread out over sixty-four countries (Best and Connolly, 1982 [1976]: 175–6).
(8) Deming remarked that what he took to Japan was the ‘theory of the system’ (personal
conversations, 1992). He meant, in part, that much of US business enterprise was func
tionally departmentalized into profit centres to achieve local optimization without regard
to interdependencies, and that with Japanese management came managing interrelation
ship across business activities, hence the organizational principle of process integration,
later adopted by management consultants as process re-engineering. Instead of optimiz
ing business functions A, B, and C independently as in US business, the Japanese focused
on AB, AC, and ABC, thereby filling in white spots in the organizational charts of US firms
(Best, 2001).
(9) Stephen Kline (1985) presented a long list of industries that were formed without di
rect linkage to science, at least new science developed in R&D laboratories: the jet en
gine, sewing machines, weaving machinery, machine tools, most construction methods,
space shuttles, turbomachinery, combined-cycle power plants, vertical and slant take-off
and landing aircraft, and integrated circuits.
(10) See Smith and Clancey (1998: 119) for Jefferson’s letters on manufacturing, for exam
ple, ‘experience has taught me that manufactures are now as necessary to our indepen
dence as to our comfort’, 9 January 1816; and Marx (1998: 122) for Jefferson’s views on
machines and ‘technology’, before the term was coined in 1829 by Jacob Bigelow.
(11) Flow lines had existed but none before were organized around the engineering princi
ple of equalizing cycle times for all activities. This is a requirement for manufacturing
without inventory.
(12) https://en.wikipedia.org/wiki/Peon; https://en.wikipedia.org/wiki/

Slavery_in_the_United_States.
(13) https://en.wikipedia.org/wiki/History_of_unfree_labor_in_the_United_States.
(14) (https://en.wikipedia.org/wiki/Great_Migration_(African_American).)
Page 39 of 42
ties
(15) See Robert Caro (1982: 502‒13) for a description of rural life without electricity.
(15) To quote Jessica Mathews (2019): ‘Today U.S. defense spending is exceptional in two
ways. First, it spends more than the next eight largest spenders combined—China, Saudi
Arabia, India, France, Russia, Britain, Germany, and Japan. Second, defense accounts for
almost three-fifths of the federal government’s discretionary budget. This is a measure of
all federal spending other than mandatory allotments to entitlements and interest on the
national debt. Discretionary spending is everything else the government does.’
(16) In the Second World War defence spending peaked at 41 per cent of GDP, and then
declined to about 10 per cent at the height of the Cold War. Thereafter it declined to 3–5
per cent of GDP, with surges during the 1980s and the 2000s (https://
www.usgovernmentspending.com/defense_spending).
(17) Northern Virginia is a special case; most of its research activity is classified. See Best
(2011).
(18) See Gavin Wright (2020) for a history of the ‘handful of “knowledge economy” clus
ters that dot the country’s western coastline’. Wright maintains that ‘World War II was in
deed the triggering shock that set in motion chains of events whose outcome was the Pa
cific coast economy as we know it’.
(19) https://en.wikipedia.org/wiki/Rust_Belt.
(20) The Tillman Act did not require proof of corrupt intent by the corporation, mere con
tribution of corporate funds to politicians sufficed to establish ‘corruption’. Political
spending was not potentially corrupting, it was corruption as the term was used in the
Act (Whitehouse and Stinnett, 2017: 22).
(21) Franklin D. Roosevelt described the powerful forces of ‘organized money’ as ‘this res
olute enemy within our gates’: ‘Never before in all our history have these forces been so
united against one candidate as they stand against me—and I welcome their hatred. I
should like to have it said of my first administration that in it the forces of selfishness and
of lust for power met their match. I should like to have it said of my second administra
tion that in it these forces met their master’ (Whitehouse and Stinnett, 2017: 10; see also
Phillips-Fein, 2009).
(22) From a paper presented at the Antitrust Seminar of the National Association of Busi
ness Economists in Cleveland, 25 September 1961. Available at https://www.aei.org/
carpe-diem/alan-greenspan-on-monopoly-and-antitrust-policy-in-1961/.
(23) President Richard Nixon appointed Lewis Powell, a long-time attorney for the tobacco
industry, to the Supreme Court in 1971. He was the author of the Powell Memorandum, a
secretive call to action for billionaires to overthrow the tyranny of regulation by spread
ing the doctrine of market fundamentalism (Mayer, 2016; Phillips-Fein, 2009).
Page 40 of 42
ties
(24) By one estimate, with a precipitous drop in water infrastructure spending per capita
since the 1970s, the United States has built up a debt of US$1 trillion that must be paid
over the next twenty-five years (Madrigal, 2019). The American Society of Civil Engineers
estimates it will cost US$3.6 trillion to get the United States back to an acceptable level
of debt in the nation’s material infrastructures (Madrigal, 2019; Bipartisan Policy Center,
2016). To these must be added the erosion of public health infrastructural investments ex
posed by the Covid-19 pandemic.
(25) The South, with rare exceptions, did not promote industrial districts or cluster dy
namic processes that are integral to industrial innovation. See Best (2020).
(26) For an extended treatment from which this is drawn, see Best (2018).
(27) Babbage’s pursuit of principles of change was an application of the systemic observa
tion approach to scientific progress being advanced by his fellow natural philosophers at
Cambridge University. The pursuit of systemic-observational principles of change united
the emerging sciences of evolutionary biology, geology, and astronomy. In the case of po
litical economy, scientific investigation started with observation of production in work
shops and factories in which engineering practices were most innovative and change was
most dramatic.
(28) The Babbage ‘new system of manufacturing’ did not, however, disappear in the real
world; instead the integration of technology development within production and scientific
research reappeared in the form of real-world, vertically integrated industrial enterprises
that drove the Second Industrial Revolution in the United States (Chandler, 1977; Lazon
ick, 1991).
(29) For the modern reader, Babbage’s On the Economy of Machinery and Manufactures
can be read as an early version of Vannevar Bush’s highly influential Science the Endless
Frontier, published in 1945. Both were pioneers in designing complex machines that pre
ceded the computer. Both combined deep practical and scientific knowledge surrounding
the leading-edge engineering innovations and ‘mechanical principles’ of their times. Both
authored, although over a century apart, a visionary text in which productivity and na
tional wealth could be continuously advanced by iterative co-adaptation of science and
production engineering.
(30) Simon Kuznets and Babbage also shared a common theme. Kuznets’ empirical studies
of long-run trends attributed gains in productivity to the marriage of science and produc
tion, and the creation of new industries. But while Vannevar Bush and Kuznets linked eco
nomic progress to scientific advance, Babbage articulated a political economy framework
which brings technology out from the shadow of science.
(31) In Marshall’s words, ‘The law of increasing return may be worded thus: An increase
of labour and capital leads generally to improved organization, which increases the effi
ciency of the work of capital’ (1920 [1890]: 265).
Page 41 of 42
ties
Michael H. Best
Michael H. Best has been actively engaged in industrial restructuring policy in Lon
don, Massachusetts, Slovenia, Cyprus, Jamaica, Honduras, Malaysia, Northern Ire
land, and the Republic of Ireland, and has written production audits based on exten
sive factory visits in many other countries. His three books on industrial policy are
The New Competition: Institution of Industrial Restructuring (HUP, 1990), The New
Competitive Advantage: The Renewal of American Industry (Oxford University Press,
2001), and How Growth Really Happens: The Making of Economic Miracles through
Production, Governance, and Skills (Princeton University Press, 2018) which won the
2018 Schumpeter Prize.
Page 42 of 42
European Industrial Policy: A Comparative Perspective

Patrizio Bianchi and Sandrine Labory
This chapter examines industrial policy implemented in European countries across time,
especially after the Second World War and in recent years. It briefly shows that industrial
ization never occurs without strong supporting industrial policy to provide the enabling
conditions for the deep structural changes involved, such as infrastructure and capabili
ties, especially human capital. Industrialization in the eighteenth and nineteenth cen
turies is associated with the creation and consolidation of nation states, leading to nation
al perspectives on industrial development and industrial policy, at the expense of regional
imbalances. After the Second World War, three phases of industrial policy take place:
strong and selective intervention in the first three decades of the period, market-led ap
proaches in the 1980s and 1990s, and a more pragmatic approach at the turn of the cen
tury, where deep structural changes require industrial and institutional adjustments. With
globalization and the emerging new technological paradigm of Industry 4.0, innovation
and skills, as well as the territorial roots of industrial development, seem to be important
aspects of industrial policy today. The chapter also argues that the multilevel governance
policy framework has to adapt to current disruptions, particularly in Europe.
Keywords: industrial hubs, global value chains, gender, women’s empowerment, working conditions
21.1 Introduction
THIS chapter examines industrial policy implemented in European countries across time,
especially after the Second World War and in recent years. It briefly shows that industrial
ization never occurs without strong supporting industrial policy providing the enabling
conditions for the deep structural changes involved, such as infrastructure, capabilities,
and human capital, and often direct involvement in production, especially in strategic sec
tors. This is true for the pioneer of industrialization, Britain in the eighteenth century, as
well as later industrializers in continental Europe in the nineteenth and twentieth cen
turies. Industrialization is associated with the creation and consolidation of nation states,
Page 1 of 29
and industrial policy has been carried out to strengthen the power of these states, not on
ly because industrial development increased wealth, hence economic and political power,
but also because industrialization contributed to military equipment and capacity. This
led to national perspectives on industrial development and industrial policy, at the ex
pense of regional imbalances.
After the Second World War, industrial policy continued as an instrument of the power of
nation states, especially in the first phase of reconstruction and development where it
was massively, though selectively, implemented. Three phases of industrial policy since
1945 can be distinguished (Bianchi and Labory, 2006, 2011a): strong and selective inter
vention in the first three decades of the period, market-led approaches in the 1980s and
1990s, and what has been called a pragmatic approach at the turn of the century.
The first phase is that of interventionist and selective industrial policy, characterized by
direct intervention in markets, with the government often being producer through
(p. 595) state ownership. Such policies tended to favour ‘national champions’ or consisted
of ‘picking winners’. In Europe, this was also accompanied by a lax implementation of

competition policy and regulation of the ‘command-and-control’ type. This type of policy
started to be questioned in the 1970s, when state-owned firms began to display a number
of inefficiencies and were unable to restructure after crises.
The second period is the 1980s and 1990s, when competition policy was implemented
more stringently, regulation became incentive based, and measures to favour structural
changes were preferably horizontal—implemented across all sectors of the economy
rather than sector or firm specific. Preferred measures included policies for SMEs,
favouring their networking and supporting entrepreneurship by simplifying firm creation
procedures, providing training, better access to finance, innovation policies favouring
R&D collaborative programmes, financing research projects on important—generic—
technologies, and then favouring the interaction between universities and industry for
technological transfer. The aim of industrial policy in that period was to provide the con
ditions for the competitiveness of industry.
The focus on providing the conditions for the competitiveness of firms and the develop
ment of sectors continues in the third period of industrial policy. This essentially starts at
the turn of the new century, when industrial firms, concerned with the intensification of
global competition induced by globalization, increasingly call for industrial policy. Global
ization means the integration of the world market, thanks to new technologies and to po
litical changes in a number of countries that have allowed them to enter world trade
agreements (China entered the WTO in 2001). Competition policy is still stringently ap
plied, regulation is rather incentive based, measures and actions are primarily horizontal
and aim at providing the conditions for competitiveness and development; however, verti
cal measures in the sense of measures specific to sectors are also envisaged if they are
necessary to reach the objectives.
Page 2 of 29
The term ‘industrial policy’ is also used again, not to mean the selective policies of the
past, but to indicate the support necessary for industrial development, namely structural
changes, requiring new infrastructure, adoption of new technologies, developing new ca
pabilities and new institutions.
Industrial policy strategies proposed at the turn of the new century emphasize the impor
tance of knowledge creation, and hence innovation policy, as well as policies on education
and training to raise the skills in the labour force. Bottom-up processes have increasingly
been stressed, with industrial development starting in specific poles or territories where
particular knowledge and competencies are concentrated. Hence, for instance, the policy
of clusters, especially high-tech ones, implemented in many countries. The new century is
characterized by significant and numerous technological changes, and innovations in
many scientific and technological fields to which industries have to adapt by digitalizing,
and changing products, production processes, and distribution channels as digital plat
forms become powerful new market intermediaries (Bianchi and Labory, 2018). These
challenges, which most European countries are addressing in their industrial policies, are
those of the Fourth Industrial Revolution.
(p. 596) This chapter examines the experiences of industrial policy implementation in
some European countries. The aim is not an exhaustive review of all European countries’
policies, nor a systematic comparison of all instruments. Rather, broad policy lines are
compared, and what emerges is that while all European countries have followed broadly
similar policies, some divergence exists regarding particular focuses, such as attention to
regional imbalances and the empowerment of regional or local government levels to con
tribute to industrial development, or the strength of support for particular firm size. For
instance, Germany has paid attention to its SME base, the Mittlestand, while France and
the United Kingdom have not effectively done so, or at least not at the appropriate time:
offshoring by large firms in the latter two countries has caused major difficulties for their
SMEs and perhaps contributed to more acute de-industrialization.
This chapter is structured as follows. We first provide a definition and taxonomy of indus
trial policy, followed by a review of the role of industrial policy in the first three industrial
revolutions. We then analyse countries’ experiences from 1945 before looking at the cur
rent phase of industrial policy implementation, which is characterized by a number of
challenges, such as globalization and the Fourth Industrial Revolution. In section 21.6 we
reflect on the evidence provided in the chapter and draw conclusions.
We largely use previous research (Bianchi and Labory, 2011a, 2011b, 2018, 2019a, 2019b)
to make the review in the first part of this chapter, while the analysis of current industrial
policy essentially confronts official documents on the national strategies.
Page 3 of 29
21.2 Industrial Policy: Definition and Taxonomy

There are numerous definitions of industrial policy in the literature, making it an ‘elusive
concept’ (Vanden Bosch, 2014). Some definitions are restrictive and selective and exclude
horizontal measures that treat all firms and all sectors the same (R&D programmes,
training, SME support, and so on). Thus, Brander (1987: 4) defines industrial policy as a
‘coordinated selection’, namely firm or industry selection through government planning
(coordination). This type of definition is closely linked to industrial policies implemented
in the immediate aftermath of the Second World War, when ‘national champions’ were
strongly supported, leading to a lack of structural adjustment when shocks arose or the
competitive context changed.
Other definitions are broader. For instance, Donges (1980: 189) considers industrial poli
cy as comprising all government measures that influence industry, such as national in
vestment policy, regional and labour policy. For Curzon-Price (1981: 17), industrial policy
consists in sets of measures implemented with the objective of impeding or favouring
structural change.
Recent literature (Cimoli et al., 2009; O’Sullivan et al., 2013; Chang et al., 2013;
(p. 597)
Bailey and Tomlinson, 2017; Andreoni and Chang, 2019; Bianchi and Labory, 2011a, 2018)
can be synthesized to provide a definition that seems to be widely accepted today. First,
industry does not mean exclusively manufacturing but includes all productive activities,
including services. This makes sense because manufacturing activities are increasingly
bundled with services (see Chang and Hauge, 2019, for an interesting discussion on this).
Industry, then, is the capacity to organize production by mobilizing both tangible and in
tangible resources. Second, industrial policy comprises a variety of instruments and ac
tions that are combined in order to set competitive rules and accompany structural
change. The competitive conditions in which firms operate frequently change for various
reasons, such as variations in the extent of the market or technological innovations. In
dustrial policy must adjust the rules of the game as a result of these changes; it must also
adopt a long-term perspective to favour the structural adjustment of enterprises.
Industrial policy can be defensive—protecting existing firms in the face of structural

changes, or proactive—anticipating changes and implementing measures that favour
structural changes. The former type of industrial policy often impedes or delays in
evitable changes, ultimately raising the costs of transformations.
Many policy actions affect industries: fiscal, monetary, environmental, trade, innovation,
and labour policy instruments have a direct or indirect impact on industries. It can there
fore be argued that all countries and territories implement industrial policies in one way
or another. However, what is important to stress is that industrial policy is a deliberate at
tempt by the government (at the local, regional, or national level) to orientate industrial
development towards specific paths. Such orientation can take various forms, from build
ing infrastructure to favour particular sectors to promoting the adoption of new technolo
gies in existing sectors in order to induce their upgrading. It can be ‘interventionist’, or
Page 4 of 29
‘picking the winner’, in the sense of choosing specific industries that are developed from
scratch, with for instance high investments, creation of appropriate infrastructure and
training of human capital, and sometimes state ownership. It can be more inclined to let
ting market forces play, for instance by providing the conditions for industrial develop
ment, and letting entrepreneurs make choices and implement strategies as a result of
which new industries can emerge.
A taxonomy of industrial policy actions can be created from this definition (Labory, 2006:
Pelkmans, 2006), dividing actions into two categories: measures not aimed at industry but
which have an impact on it; and measures aimed at industry. The first group includes
macroeconomic stabilization policies (monetary policies influence inflation and interest
rates, affecting both consumer purchasing power and business investment opportunities),
or social and labour policies. Education policies aim at the social development of the ter
ritory but also influence the availability of human capital for business.
The second group includes all measures directly aimed at industry, which can be divided
into three subgroups: (i) measures defining competitive conditions, such as antitrust poli
cy and product regulation; and measures aimed at enhancing (p. 598) business capabili
ties which can be (ii) horizontal, concerning all firms and industries, such as support to
R&D, development of human capital, access to financing, and so on, or (iii) vertical, spe
cific to particular industries, such as public procurement of specific products, investment
in specific technologies or specific skills, for example, creation of engineering schools.
A particular issue in the definition of industrial policy is coherence, given that the wide
variety of possible measures may have conflicting objectives or unintended combined ef
fects (Bianchi and Labory, 2011a). Instruments are often used for various objectives: for
instance, training policy is useful both for social purposes and for the provision of appro
priately skilled human capital; energy policy often has both social (access to energy for all
at reasonable cost) and industrial goals (ensuring energy costs do not negatively affect
business competitiveness), and sometimes also geopolitical ones (relations with produc
ing countries). The objectives can sometimes conflict, as when geopolitical interests lead
to a preference for certain energy sources although their efficiency might not be optimal.
Similarly, trade policy usually aims at eliminating trade barriers, which can conflict with
industrial policy when the new industries need temporary protection to develop.
21.3 Industrialization and Industrial Policy

Industrial policies have been implemented in most countries since the First Industrial
Revolution at the end of the eighteenth and beginning of the nineteenth century, although
with varying degrees of deliberate planning and monitoring. The main reason is that in
dustrial capacity becomes the main determinant of the wealth of nations, as stressed by
Adam Smith (1776). Countries have to develop this capacity in order to affirm their
strength, both directly because it determines wealth generated in the country, and indi
rectly because industries also produce military equipment.
Page 5 of 29
Industrialization requires complex sets of policy instruments even in the First Industrial
Revolution. They are generally determined by a new technological system (Gille, 1978) or
paradigm (Freeman and Lourça, 2001), but also new needs in society and a changing or
ganizational model which adapts to the new context so that technological progress im
pacts on economic development. In the First Industrial Revolution, the factory system
transforms the organization of production from craftsmanship to the division of labour. Its
diffusion requires new resources and skills, and a new social structure. Entrepreneurs
creating and managing factories constitute a new social class, which increasingly influ
ences the political process. Cities transform with the creation of factories, since the latter
attract a workforce previously hired in agriculture, hence a rural exodus and the growth
of urban areas, which have to re-organize and offer new services, from transport to hous
ing and progressively schooling.
The state must therefore intervene at the local level to develop new services in
(p. 599)
cities (transport, housing, and schooling), but also at the national level since industrializa
tion requires a new institutional framework, including contract law, protection of intellec
tual property rights, financial and insurance systems. Countries industrializing after the
pioneers have to implement particularly strong industrial policy in terms of direct inter
vention in markets, since their industries are developing in a competitive context where
they are immediately challenged by first-industrializer industries. They also have to ab
sorb new knowledge and new technologies rapidly, since technological change proceeds
continuously in the nineteenth and twentieth centuries: knowledge absorption has to be
combined with the development of research competencies if it is to follow and take an ac
tive part in the innovation process.
The first industrial revolutions (IRs) are based on the increasing use of coal as energy
source; iron (First Industrial Revolution) and later steel (Second Industrial Revolution) as
main raw material; and the diffusion of the steam engine in many activities. In the early
nineteenth century, electricity becomes an integral part of continued innovations in the
Second Industrial Revolution. Railways are a key element in nineteenth-century industrial
development because they enlarge the market from regional to national and internation
al. Steamships complement railways to facilitate international trade. The development of
railways also stimulates further innovations, since they require infrastructure such as
bridges, but also a specific organization and planning to avoid incidents and delays. The
organizational innovations introduced in the management of railway companies are often
argued to have been implemented in the large companies that progressively emerged in
the second half of the nineteenth century (Bianchi and Labory, 2018).
21.3.1 The United Kingdom
The United Kingdom was the pioneer of the First Industrial Revolution, in the sense that
key inventions were made in that country and the first factories developed there. Indus
trialization then diffused to other European countries (Belgium, France, Germany), the
United States, and later Japan. Policies to support this process were especially strong in
countries which followed the United Kingdom. Even in the latter part of the twentieth
Page 6 of 29
century, catching-up countries such as Korea, Taiwan, and China did implement industrial
policy (see Lall, 2006; Allen, 2011).
Whether the United Kingdom also implemented industrial policy in the First Industrial
Revolution is debated. Bairoch (1993) claims that the British government implemented in
dustrial policy well before the First Industrial Revolution, using instruments such as tariff
protection, subsidies, and export taxes from as early as the fourteenth century to support
development of the woollen textile industry. This support was later extended to all indus
trial sectors. Complementary measures included massive investment in infrastructure and
development of the financial sector to support it. Measures were taken to protect the new
industry and avoid imitation in other countries, such as a ban on exports of machines in
the late eighteenth and early nineteenth century.
However, a number of scholars argue that the measures taken by the British gov
(p. 600)
ernment were not part of an intentional industrialization programme, but rather aimed at
protecting particular lobbies (Mokyr, 1999; Allen, 2011). According to them, for instance,
the aim of the Calico Act was to protect the lobbies in the cotton industry (Mokyr, 1999),
although it had the secondary effect of promoting the development of that industry. An
other example is the heavy taxation imposed by the central government, which was main
ly used for military purposes and not industrial development; however, this spending al
lowed Britain to gain dominance over sea routes, which was favourable to trade. The de
velopment of education, science, and technology was left to private initiative rather than
being defined in an industrialization programme (Van Neuss, 2015).
Industrialization starts in Great Britain around 1760, with a series of inventions (the spin
ning jenny, the water frame, and the mule) making it possible to use machines in produc
tion, especially in the cotton sector. The rapid pace of innovation contributes to the devel
opment of factories and industries, and especially of the steam engine, continuously im
proved throughout the nineteenth century. Industrialization accelerates thanks to the rail
ways boom from the 1840s. France starts industrializing between 1820 and 1830 with the
diffusion of weaving looms and the building of the first railways. From 1870, Germany,
Canada, and the United States become rivals of France and the United Kingdom, while
Japan and Russia join after 1890.
According to Allen (2011), industrial policy implemented by followers consisted of four

main actions: unification of the internal market; external tariffs to protect domestic indus
try; development of infrastructure such as roads and later railways; and development of
human capital, with better health thanks to improvements in agriculture and food securi
ty, as well as education and training. Sweden and Germany appear to have been pioneers
in the latter, establishing universal basic education in 1842 and 1863 respectively.
21.3.2 Germany
Germany was the most active in this forced industrialization (Landes, 1969). Until the uni
fication of the Reich, King Frederick the Great (1740‒86) actively promoted new indus
try: factories were set up, particularly in the steel and linen industries, together with road
Page 7 of 29
and railway infrastructure. The financial system was developed with the creation of in
vestment banks. The availability of human capital was also ensured with compulsory edu
cation, resulting in a schooling rate of 97 per cent (Landes, 1969). Academic research al
so substantially improved with a focus on industrial applications, both for the state (mili
tary equipment) and for business (for instance, chemistry). The market was unified with
the Zollverein of 1818 and protected, following the recommendations of List who had ob
served trade protection policy in the United States (Bianchi and Labory, 2018).
(p. 601) 21.3.3 France
Industrialization was realized in a similar way in France, thanks to the creation of facto
ries by entrepreneurs, state investment in public goods such as railways and other infra
structure, and the development of a financial system based on banks and shareholders’
associations (known as caisses). Industrial policy in France was based on the Colbertist
doctrine, named after the seventeenth-century minister who created specific companies
belonging to the monarchy that were granted monopoly (Levy-Leboyer, 1996). State inter
vention remained important in France especially after the Second World War.
21.3.4 Italy
Italy started industrializing in the late nineteenth and early twentieth century, with strong
intervention by the state, culminating in the creation of the Institute for Industrial Recon
struction at the beginning of the 1930s, which took shares in numerous companies in
strategic sectors (steel, energy, transport, and communications) after the financial crisis.
Throughout the twentieth century, innovations continue at a sustained pace, based on sci
entific discoveries in all fields generating two new industrial revolutions. The Second In
dustrial Revolution is primarily based on electricity and steel, but also chemicals and the
automobile industry, where mass production begins and is later diffused to all industries.
The Third Industrial Revolution after the Second World War concerns oil as the main en
ergy source, but also nuclear energy, aerospace, automobile, computer, and telecommuni
cations industries (Freeman and Lourça, 2001).
21.4 Industrial Policy from 1945 to the Twenty-

first Century
European countries implemented industrial policies after the Second World War, initially
reconstruction after the devastation of war. Many industries had been turned into war in
dustries and had to re-organize. The United States provided financial support for this ef
fort with the Marshall Plan (European Recovery Programme) of 1948. The political econo
my of industrial policy at that time was to ensure that Western European countries would
recover to become prosperous economies that could face the threat of communism during
the Cold War, besides providing a market for American firms.
Page 8 of 29
Industrial policy was interventionist in the first phase, with the state often the producer
through state-owned enterprises, especially in infrastructural (transport (p. 602) and
telecommunications) and strategic sectors, such as coal and steel. This strong interven
tion also translated into the promotion of national champions that could deal with interna
tional competition: the explicit aim of this policy in France was to have large firms able to
compete against American ones.1 Another aim was to develop the defence industry as a
Cold War deterrent, and this led to important R&D programmes, for instance in the nu
clear and telecommunications industries in France and the United Kingdom in particular;
much less so in Germany and Italy, which were defeated countries.
Other European countries that were more politically neutral or at a different stage of de
velopment made different strategic choices. All European countries benefited from the
economic integration programme, starting with the ECSC (European Coal and Steel Com
munity) in 1951 and the EEC (European Economic Community) in 1957. The Treaty of
Rome set up a common market between the member states following different stages of
integration, from trade union to the single market (see Landesmann and Stöllinger, Chap
ter 22 of this volume). The market therefore enlarged for the member states, allowing
firms to exploit higher economies of scale and scope, since both volume and product vari
ety could increase. The original six member countries—France, Germany, Italy, the
Netherlands, Belgium, and Luxembourg—expanded to include Denmark, the United King
dom, and Ireland in 1973, Greece in 1981, Spain and Portugal in 1986, Austria, Finland,
and Sweden in 1995, and the countries of Eastern and Central Europe from 2004 on
wards.2
The single market is thus often considered as the main industrial policy of the European
Union (Pelkmans, 2006). Other aspects of industrial policy at the European level are com
petition policy and trade policy. Ensuring fair competition on the European market is
therefore a key element, as well as ensuring open trade worldwide.
While the European level has constantly stressed the importance of innovation policy,
with a number of European R&D programmes, such as ESPRIT and EUREKA in the 1980s
(see Chapter 22 by Landesmann and Stöllinger in this volume), the main competence re
garding industrial policy (of which innovation policy is a part) is at the national level. Eu
ropean treaties have included an article on industrial policy since the Maastricht Treaty
of 1993, but it invites member states to support the competitiveness of their industries
with particular attention to SMEs and their collaboration, as well as R&D and innovation,
privileging horizontal measures, that is, measures not specific to firms or sectors. Indus
trial policy (called enterprise or competitiveness policy) is mentioned at the European lev
el only in the second phase of industrial policy implementation, starting with the new ap
proach formulated by Commissioner Bangemann in 1990. Earlier debates at the Euro
pean level had included the Colonna Report of 1970, which stressed that the creation of
the single market and other exogenous factors would (p. 603) make the structural adjust
ment of European firms necessary. However, after the difficulties for European industries
generated by the oil crisis of the 1970s, the reaction of the member states was to pursue
a defensive policy. The best illustration is the Davignon Plan of 1977, which intervened to
Page 9 of 29
support the steel industry in crisis: minimum prices and important firm subsidies were
provided to reduce production capacity. This led to protection of the existing firms (avoid
ance of failures) but without encouragement to realize structural adjustment.
The increasing trade in the European Community was, however, beneficial to businesses
and economy in the member states, which supported their industries during this first pe
riod.
21.4.1 Italy
In Italy, the literature agrees on the important role of the state in its industrial develop
ment (e.g. Bianchi, 2002). After the Second World War the country was still lagging be
hind the United Kingdom, Germany, and France. The government intervened to modern
ize and develop the country, mainly with subsidies to industry and large investment
through the state-holding Institute for Industrial Reconstruction, with large state-owned
companies acting as leader. Trade was progressively liberalized in the 1950s, first in the
ECSC and in the EEC from 1957. Italy then chose a growth strategy based on exports.
The government supported declining industries in the 1970s, while the emphasis shifted
in the 1980s when it realized both the importance of SMEs in the industrial system, espe
cially those grouped in industrial districts, and the need to promote research and innova
tion. Support to industrial districts started being provided at the regional level, using the
European structural funds. IRI was closed at the end of the 1990s, after a big wave of pri
vatizations and the adoption of a national antitrust law in 1990.
Many authors, such as Federico and Foreman-Peck (1998) and Bianco (2003) criticize the
incompleteness of the Italian industrial policy: some measures have been taken, but many
elements are missing, such as an efficient capital market, simple bureaucratic procedures
especially for firm creation, and support for new industries (Malerba, 2000; Rolfo and
Calabrese, 2003). Nonetheless, Italy ranks second in manufacturing in the European
Union today.
21.4.2 France
In France, industrial policy was a national strategy defined by a specific entity, the Com
missariat Général au Plan, created in 1946. In the decades after 1945 this planning office
used state ownership, among other instruments, to support large firms that could com
pete against American ones. The country has experienced two waves of nationalization
when the state acquired private enterprises: in 1944‒5, various firms in the (p. 604) finan
cial, energy, utility, and transport sectors came under state ownership; in the 1980s, the
state acquired firms in declining industries, such as steel and shipbuilding, as well as
strategic sectors such as chemicals, pharmaceuticals, and electronics. Privatizations
started in the 1990s, but many former state-owned companies remained under state con
trol through ‘golden shares’.
Page 10 of 29
As well as state ownership, French industrial policy provided large subsidies to industry,
as well as large R&D programmes (grands programmes) in strategic sectors. These pro
grammes were characterized by state-owned companies and public procurement, with a
strong research focus in science and technology that had important military applications
(aerospace, nuclear, and telecommunications). Industrial policy was definitively re
launched at the beginning of the 2000s.
The United Kingdom also based industrial development on state ownership, subsidies,
knowledge transfer, and fiscal policy in the period 1945 to 1979. Restrictions on interna
tional trade were implemented in the 1950s. The government acquired private enterpris
es in difficulties, such as Rolls Royce in the 1970s. Skills development was strong right up
to the 2000s, but perhaps not enough emphasis was put on intermediate skills: in 1990,
27 per cent of 16-year-olds passed their maths and mother-tongue test, against 62 per
cent of German, 66 per cent of French and 50 per cent of Japanese people of the same
age (Foreman-Peck and Hannah, 1998). Some specific sectors, for instance aerospace and
information technology, and all activities with military applications, were supported. As
early as 1948 the United Kingdom was also the first European country to adopt antitrust
law.
From 1979, with the advent of the Thatcher government, industrial policy was aban
doned. There were privatizations and liberalization of utilities, old industries closed down
with heavy social costs and the only remaining industrial policy measures were innova
tion policy (although this was not effective according to von Tunzelmann, 2010) and FDI
attraction (as a result of which Japanese producers attracted by the European market set
up factories in the United Kingdom in the 1980s).
21.4.4 Germany
According to Feldenkirchen (1998), industrial policy was continuously applied in Ger

many in the twentieth century. While the overall approach was to provide the conditions
for industry to prosper, specific support was also given to industries or firms (Donges,
1980; Feldenkirchen, 1998) in declining or strategic sectors, with military applications in
particular. Despite mergers and demergers experienced due to the political and economic
upheavals before and after the Second World War, German (p. 605) big businesses re
mained stable. They were concentrated in the heavy industries (coal and steel), with, for
instance, Hoesch, Krupp, and Mannesmann, chemicals (IG Farben), electrical engineering
(Siemens, AEG) and automobiles. Germany regained a world-leading position in the
chemical industry after the Second World War, with four main companies created from
the dismantling of the IG Farben company: Bayer, Hoechst, Agfa, and BASF. Only large
banks suffered after 1945, when the Allies forced the dismantling of the three major
banks, the Deutsche Bank, the Dresdner Bank, and the Commerzbank. However, they re
united in the 1950s.
Page 11 of 29
In addition, German industrial policy has been export oriented, especially in the Euro
pean economic integration process. Innovation has also been an element of this policy,
with a strong focus on the industrial application of research: for instance, the Fraunhofer
Institutes were created in 1949 with a focus on industrial application research, while oth
er institutions like the Max Planck Institute focused on basic research. Fraunhofer Insti
tutes establish links with both large firms and especially SMEs (Mittlestand), the back
bone of the German industrial system.
Germany also has a specific banking system focused on long-term lending to industry. The
German government created a number of public or quasi-public banks during the 1950s
to ease industry access to finance. Thus, the Kreditanstalt für Wiederaufbau (KfW) was
created in 1948 in order to manage the Marshall funds allocated to the country for recon
struction. It soon became focused on financing to SMEs. Other activities were progres
sively added, such as export financing in the 1960s, financing of infrastructure on behalf
of the municipalities and other local administrations in the 1970s and 1980s, and financ
ing of the reconstruction of East Germany in the 1990s. Mittlestand companies were sup
ported with access to finance since they also benefited from the Bank for Settlements
(AG) and a decentralized network of savings banks (Sparkassen) and credit cooperatives.
There are about 400 savings banks3 and 875 cooperative banks4 in Germany, which to
gether provide about two-thirds of all lending to Mittelstand companies.
Industrial policy acquired a new impetus in the 1990s with reunification. Alongside a de
cisive effort to modernize and reconvert industry in the former East Germany, measures
for industry in West Germany continued, focused on providing the appropriate conditions
for competitiveness and spurring innovation.
Characteristic of German industrial policy is the coherence of its multilevel governance

framework between the local, the regional (Länder), and national (federal) levels. It is al
so proactive, trying to anticipate changes so that industry can adapt to changing competi
tive and technological conditions, for example, in the recent policy supporting the struc
tural transformations implied by Industry 4.0.
(p. 606) This institutional coherence also prevails in Sweden and Finland. Both countries
have implemented industrial policies with instruments similar to those of other European
countries. Up to the 1980s these policies tended to be selective, directed, for example, at
forest industries which were important in these countries’ economy at the beginning of
industrialization and remain so today (Blomström and Kokko, 2002). A new approach was
adopted in the 1990s, with a higher focus on innovation and technological transfer, as
well as the local roots of industrial development, and important cluster policies (Blom
ström and Kokko, 2002; Ylä-Anttila and Palmberg, 2007). An important feature is the
adoption of a systemic view in which great attention is paid to the relationships between
actors in the innovation process. Policies have been defined in dialogue with stakehold
ers, establishing public‒private partnerships especially at local and regional levels. Indus
trial application of research has thus been rooted locally but favouring the networking of
local clusters and excellence centres at a national level (Ketels, 2009).
Page 12 of 29
21.4.5 Finland
The case of Finland deserves more analysis, because this country achieved very deep
structural changes in only a few decades. More than half the population and 40 per cent
of output were still in the primary sector in the 1950s. However, the country had already
become a mature industrial economy by the 1970s. The evidence suggests that Finland’s
industrialization was strongly backed by the state through industrial policy. From the
1950s an important intervention was made, based on a strong consensus between the
state at all levels of government, businesses, and other stakeholders, that defined a vision
of industrial development as well as measures towards this aim. ‘The Finnish business
elites colluded to distort the markets in a way that sustained high investment
rates’ (Jäntti and Vartiainen, 2009: 9). Actions included large investments, particularly in
forestry and metals, that were enabled by the high rate of savings available in the econo
my at that time. Interest rates were also controlled by credit rationing. Human capital
was developed by investment in education and the development of the welfare state. So
cial cohesion was therefore high, partly for political reasons, since Finland had ‘recently’
become independent from the USSR (1917) and the feeling in the population was that the
country had to grow to justify its independence and workers had to be cared for to avoid
increasing support for communist ideas and the possible influence of its neighbour coun
try. Trade unions were encouraged to take part in the discussions on industrial policy and
unionization was encouraged.
The industrial system developed, favouring economic growth and maintaining social cohe
sion. Moderate wage increases were agreed in the negotiations between the state, busi
nesses, and labour to maintain the international competitiveness of Finnish firms.
Finland was a dynamic knowledge-based economy by the beginning of the 2000s, thanks
to a second industrial transformation in the 1990s, characterized by the rise of (p. 607)
the ICT sector with the dominant firm Nokia, promoted by an industrial policy aimed at
developing new sectors defined at the beginning of the 1990s, and comprising actions
aimed at raising higher education, clusters, and R&D investments.
The flagship company Nokia grew exponentially in the 1990s and became the best-selling
mobile phone brand in the world in 1998. The company represented about 4 per cent of
the country’s GDP, 26 per cent of its total exports and about half of total national R&D in
2000 (Ali-Yrkkö, 2010). However, the company missed the shift to smartphones and went
into deep and rapid trouble from 2007 onwards. The firm represented only 1.6 per cent of
GDP and 10 per cent of the country’s exports in 2009. In just six years (2007 to 2013), the
market value of Nokia declined by 90 per cent.
Nokia was taken over in 2014 by Microsoft, which started a drastic re-organization, lay
ing off employees and closing some factories and research centres in Finland. The
Finnish government made an agreement with Microsoft to avoid excessive lay-offs. More
importantly, the government immediately started a strong policy to support firm creation:
the budget of the Finnish agency for research and innovation, Tekes, was substantially in
creased to support new and innovative businesses; Vigo, a start-up accelerator, was
Page 13 of 29
launched in 2009 by the Finnish Ministry of Employment and the Economy. Since then the
country has remained an innovation leader in the European Union, as witnessed by the
European Innovation scoreboards.5
21.4.6 Ireland
Like Finland, Ireland developed rapidly from the 1990s to 2005. The effects of the 2008
financial crisis were particularly severe in Ireland, seemingly due to structural weakness
es that were not addressed by industrial policy.
Yet the country had experienced average annual GDP growth rates of more than 7 per
cent, leading commentators to speak of the ‘Celtic Tiger’. This development was eased by
industrial policy of a very different kind from that of Finland. While the latter country
based its industrial policy on the development of autonomous capacity, Ireland tended to
import industrial capacity from outside. Irish development in the 1990s was spurred by
the boom of the high-tech sector, especially in electronics and pharmaceuticals, but the
bulk of the sector was and still is foreign owned. The latter represented about 84 per cent
of output in 2006, while employment was almost equally divided between the two types of
firms (Kirby, 2010: 6‒7). The Irish industrial policy model has been very open and based
on the attraction of foreign companies with mainly low tax rates (Bailey et al., 2012). Dif
ferent authors (Kirby, 2010; Cullen, 2004) have shown that this strategy came at a high
social cost, in that the benefits of industrial development have not been shared among the
population; the poverty rate rose during the 1990s at a time of high growth.
The European integration process has deeply influenced the industrial policies of
(p. 608)
European countries. First, it has offered new opportunities for national industries, thanks
to the enlargement of the market, allowing static and dynamic scale and scope
economies. Second, it has induced member countries to coordinate their policies, some
industrial elements of which (competition and trade policy) have become common poli
cies. Other policies such as innovation policy, have included a progressively European di
mension. However, the main competence on industrial policy remains at the national level
in the EU. Third, the European Commission has played an important role in monitoring
and benchmarking industries and economies, highlighting the economic challenges and
helping to focus policy efforts on them. National industrial policies remained focused on
declining industries up to the 1970s, together with R&D programmes. The awareness
that globalization accelerated at the end of the 1980s led to the definition of a new ap
proach to industrial policies by Commissioner Bangemann in 1990. The term ‘industrial
policy’ was no longer used, because of its association with the interventionist and selec
tive policies of the past. Policies supporting structural changes in industries were called
competitiveness or enterprise policies.
The new approach consisted in providing the conditions for the competitiveness of indus
tries, privileging horizontal measures and instruments as well as fair market competition.
The role of the state was defined as a pioneer and catalyst of industrial development,
Page 14 of 29
identifying future technologies and industries and promoting their adoption and expan
sion. It was formalized in Article 130 of the Maastricht Treaty.
The new horizontal approach, however, does not appear to have been successful in reduc
ing the competitiveness gap between the European Union and its main rivals in the
1990s, Japan and the United States. The problem was not lack of innovation, but a limited
capacity to transform innovation into commercial success, which the European Commis
sion called the ‘European paradox’ in 1995 (European Commission, 1995; Dosi et al.,
2006). This is particularly challenging as new rivals have emerged since 2000, especially
from Asia, and China and Korea in particular.
As a consequence, the debate on industrial policy emerged again in the European Union
at the beginning of the new century, especially in 2002 when the heads of state of France
and Germany (respectively Jacques Chirac and Gerhard Schroeder) openly expressed con
cern over de-industrialization and called for industrial policy. The European Commission
has published numerous communications on industrial policy since then, proposing to in
crease R&D effort and industrial applications for research, as well as training and educa
tion, while maintaining a strong emphasis on competition policy and open trade. Some
measures specific to sectors have also been suggested, for instance in the chemicals sec
tor. The numerous communications of the European Commission on the subject in the last
fifteen years (Bianchi and Labory, 2011a; Landesmann and Stöllinger, Chapter 22 in this
volume) are perhaps signs of a search for what industrial policy should look like. The
competitiveness gap has persisted and somehow widened because it now also exists with
respect to China.
Another issue is that although member states are aware of the need to support industries,
most have only been able to do so in a limited manner, especially during (p. 609) crises,
because budget constraints in the eurozone impose limits on spending. In particular, the
golden rule adopted after the financial crisis (the fiscal compact of 2012) has been criti
cized by macroeconomists as severely limiting the ability to use fiscal policies for stabi
lization purposes in a manner consistent with moderate long-run debt levels (Whelan,
2012). The ability of EMU countries to implement fiscal stimulus is extremely limited, and
this has been a problem in the recession that followed the 2008 financial crisis and that
lasted several years (Eyraud et al., 2017; Darvas et al., 2018).
While the horizontal approach has helped, it has the disadvantage of spreading financial
efforts over a large number of actors, which can lead to results that are, at best, diluted.
This is particularly problematic when resources are scarce, as in the case of strict con
straints on public deficit, which is what happened in the first decade of the 2000s, and led
to a new approach, namely smart specialization.
Section 21.5 reviews current industrial policies in European countries, starting from a re
view of the importance and evolution of the manufacturing sector in recent decades.
Page 15 of 29
21.5 New Industrial Policy at the Turn of the

Century
Manufacturing value added as a percentage of GDP tended to fall in all countries over the
period 1991 to 2018 (Figure 21.1). Germany is the European country where the manufac
turing sector remains most important in the national economy: manufacturing value
added represents 21 per cent of GDP in 2018, against about 15 per cent in Finland, Italy,
and Sweden (respectively 15.3, 15.0, and 13.8 per cent). Countries with the lowest shares
are France (10 per cent) and the United Kingdom (8.9 per cent).
Figure 21.1 shows that all countries encountered a long phase of restructuring, whereby
the importance of the manufacturing sector in GDP declined, followed by a stabilization
phase. However, the timing and length of the restructuring process differs across coun
tries: Germany saw a reduction of the share of the manufacturing sector in GDP in the
first five years of the 1990s, followed by a stabilization and slight growth after the finan
cial crisis. In France, Italy, and the United Kingdom the decline carries on up to the finan
cial crisis. In Finland and Sweden, the decline starts later, in 2000, and lasts about ten
years. For all countries except Germany, the decline becomes steeper after 2000, corre
sponding to the entry of China to the WTO (2001) and the period where globalization was
most strongly felt, in the sense of intensification of world competition following the mas
sive entry of new protagonists such as China and the BRICS countries.
FIGURE 21.1 Manufacturing value added as a per

centage of GDP, selected countries, 1991‒2018
Source: World Bank data, https://data.worldbank.org/

indicator/NV.IND.MANF.ZS?locations=CN.
China has become the global manufacturing centre, with manufacturing value added rep
resenting about 30 per cent of GDP against 11 per cent in the United States in 2017.
(p. 610) Chinese production is no longer (or not only) of low quality, since the country has
become a world leader in innovation. For instance, China’s share of world patent applica
tions increased from 4.6 per cent to nearly a third of world patent applications (30.4 per
cent) in the period 2004 to 2017, mainly at the expense of Japan (33.1 to 16.9 per cent),
while the shares of the EU-28 and of the United States slightly reduced.
Page 16 of 29
Overall, European countries invest in R&D, but their size is too small to reach the level of
expenditure (input) but also innovation (output, as shown by patenting activity) of China
and the United States. The innovation capacity of the European Union as a whole is much
higher and comparable to that of China and the United States, although it has slightly de
clined in the last decade: the share of world patent applications of the European Union as
a whole (twenty-eight countries) was more than 20 per cent in 2004 and was about 18 per
cent in 2014. OECD data on triadic patent families6 show that only a few advanced coun
tries increased their share over the period 2005 to 2017; the EU28’s share reduced from
29.7 to 24.5 per cent, the United States experienced the same downward trend from 28
to 22.4 per cent, while China’s share increased from 0.8 to 7.9 per cent (OECD, 2019).
Especially in a period of dramatic transformations, and the continuing Fourth Industrial

Revolution, support for structural changes and upgrading/renewal in industries is funda
mental. European countries are implementing industrial policy to adapt to the industrial
revolution, but their effort is small in size relative to countries such as the United States
and China, and European countries would gain from further integrating their efforts.
The Fourth Industrial Revolution is characterized by the development of scientific

(p. 611)
breakthroughs and disruptive technologies such as the Internet of things (IoT), robotics,
virtual reality (VR), and artificial intelligence (AI), but also nanotechnologies, genomics,
and new materials, which are changing the economy and the society. The Third Industrial
Revolution was characterized by the development of computers and IT (information tech
nology) in the second half of the twentieth century. The Fourth Industrial Revolution
builds on innovations of the third but is considered a new era because it generates tech
nological development that is massive and disruptive (Bianchi and Labory, 2018; Schwab,
2016).
The growing awareness of the deep transformations involved and the importance for in
dustry to adapt and participate in the process have led governments to design and imple
ment industrial policy aimed at supporting these structural changes. European countries
have been among the first countries to do so.
21.5.1 Germany
Germany in particular can be argued to be the first country to have adopted an industrial
policy for the manufacturing revolution.
The term ‘Industry 4.0’ was first publicly introduced in 2011 as Industrie 4.0 by a group
of business, government, and academia representatives who gathered to reflect on how to
enhance German competitiveness in the manufacturing industry. The German federal gov
ernment adopted their proposal in its High-Tech Strategy for 2020. A working group was
created to further advise on the implementation of Industry 4.0, which made recommen
dations in 2013. They stressed the deep changes occurring in manufacturing, since ‘these
cyber-physical systems comprise smart machines, storage systems, and production facili
ties capable of autonomously exchanging information, triggering actions, and controlling
each other independently. This facilitates fundamental improvements to the industrial
Page 17 of 29
processes involved in manufacturing, engineering, material usage, and supply chain and
life-cycle management.’
The term is now widely used to indicate the transformation of productive processes in
duced by the technological changes of the Fourth Industrial Revolution: a new organiza
tion of production is possible in the ‘smart’ or ‘digital’ factory, where machines and robots
interact to adapt and manufacture customized products on a large scale, thanks to cyber-
physical systems that also interact directly with the customer.
Bianchi and Labory (2019a) argue that this implies a change in the manufacturing sys
tem, since the Fourth Industrial Revolution allows the setting up of largely automated
manufacturing processes that enable products to be personalized (economies of scope) on
a large scale (economies of scale). Mass production had high volumes (scale) but low dif
ferentiation (scope). Industry 4.0 (mass customization) allows both high scale and high
scope. These changes allow business to develop new products, launch new markets, and
change existing ones. Disruption in industry is even increased by the (p. 612) emergence
of new market intermediaries in the form of digital platforms (Bianchi and Labory, 2018).
Germany’s industrial policy has therefore been focused on Industry 4.0 in recent years
(together with the earlier focus on green technologies), and the prime importance of tech
nological and innovation policy, since this revolution is primarily about scientific discover
ies and technological breakthroughs in various fields. The governance process is partici
pative, in the sense that it is based on a dialogue with stakeholders—industry, science,
and social partners. The funding model is based on public sources, targeting research,
partnership-building, competence centres, and test-beds, but industry also contributes,
especially larger firms who contribute proportionately more. German industrial policy for
the manufacturing revolution is particularly concerned about SMEs, which may lack ac
cess to finance and other resources for the transition and structural changes involved.
The government strategy requires industry co-financing, but large firms provide propor
tionately more financing than SMEs. The German government has also been increasingly
concerned over the protection of its industrial know-how, especially since Chinese organi
zations have started to acquire German firms. This led the German economics minister
Peter Altmaier to announce the possibility of the German state buying stakes in German
companies in order to maintain its industrial know-how.7
After the period of policies characterized by privatization, deregulation, and the free mar
ket in the 1980s and 1990s, the United Kingdom economy had a weak performance at the
turn of the twenty-first century: increasing trade deficit, decreased manufacturing capaci
ty, and rising regional inequalities (Bailey and Tomlinson, 2017; O’Sullivan et al., 2013;
Chang et al., 2013). The financial crisis made the situation even worse, and the govern
ment realized that countries with stronger manufacturing bases had better capacity to re
cover. Concern over de-industrialization increased and successive governments intro
duced different measures to support a revival of British industries. By the end of the sec
Page 18 of 29
ond decade of the new century, however, the performance of the British economy had not
substantially improved, productivity being particularly low relative to other advanced
countries (Bailey and Tomlinson, 2017).
The United Kingdom has good research capacity, with four universities in the top ten in
the world. However, the manufacturing base is so weakened that it has been unable to
transform this research capacity into upgraded commercial products. It is a question not
only of strengthening the links between research institutions and industry, but also of in
creasing manufacturing capacity, particularly through the development of (p. 613) manu
facturing skills. The United Kingdom particularly lacks the intermediate skills that are so
important to industry.
Catapult centres have been created, following the German model of Fraunhofer Insti
tutes, in order to increase the capacity to transform innovation into industrial products
and processes. They are based on specific technologies and scientific domains and aim at
gathering stakeholders, particularly universities and industry, to promote the industrial
application of research. They are funded by private businesses, the UK government, and
other sources (such as EU funds). According to Bailey and Tomlinson (2017), the public
funds allocated to catapult centres have been limited so far, at least compared to efforts
made in other countries. Another problem is the lack of engagement with SMEs.
As a result of both the fear of de-industrialization and lagging productivity, UK govern

ments have regularly adopted industrial strategies. The measures announced have been
promising, but the problem is that governments have changed and so have strategies, so
the overall picture is confusing and amounts to myriad different measures, some remain
ing from the past and some new, but their coherence is largely uncertain.
Businesses in the United Kingdom complain about the lack of support for SMEs and for
training and education. SMEs do not have the resources to define and implement their
own training and education programmes, so they rely on public programmes, which are
lacking in the United Kingdom.8
A Green Paper on ‘Building our Industrial Strategy’ was adopted in January 2017 to re
launch industrial policy. However, it largely reiterates existing measures (Bailey and Tom
linson, 2017). The government has also committed to a rise in R&D effort in the Green Pa
per, as well as raising skills with the creation of Institutes of Technology that could pro
vide the lacking intermediary skills.
Surprisingly, the Green Paper does not mention Industry 4.0, on which industrial policy in
other European countries, such as Germany and France, is focused. Given the evidence
regarding this important industrial transformation, leading to a new manufacturing para
digm (Bailey and Tomlinson, 2017) or manufacturing regime (Bianchi and Labory, 2019a),
industrial policy should be centred on support to the important structural changes in
volved. Transition to the new manufacturing regime, moreover, could be an opportunity
for the United Kingdom to rebuild British industry following new technologies.
Page 19 of 29
Another important point concerns regional imbalances. We argued above that German in
dustrial policy is extremely coherent in its multilevel governance framework, with the re
gional governments (Länder) tailoring federal policy to regional characteristics, namely
to the regional mix of industries. This is particularly true with respect to R&D and train
ing efforts; for instance, intermediate skills are provided in apprenticeship programmes
that are broadly defined at the federal level but specifically (p. 614) defined together with
industrial actors at the regional level. This favours industrial development, but also bal
anced development across the country. From this perspective, the UK economy has
strong regional inequalities, but recent industrial policy initiatives have not addressed
this issue. The recent abolition of Regional Development Agencies (RDAs) and their re
placement by Local Enterprise Partnerships (LEPs), which are development agencies op
erating at very small scale, namely that of local authorities, does not seem to help region
al rebalancing. LEPs are too small and fragmented to be able to implement effective in
dustrial policy.9
21.5.3 Italy
Like the United Kingdom, Italy has strong regional inequalities, despite a relatively
strong manufacturing sector. In fact, the latter sector is strong mainly because the north
of the country reaches development levels and industry sophistication comparable to the
most advanced regions in the European Union. In contrast, the southern part of the coun
try (the Mezzogiorno) has continuously declined in recent decades. National policies have
failed to address this problem (see Barzotto et al., 2019). Industrial policy at the national
level has been missing, apart from a few initiatives that sometimes could not even start
due to political instability. One initiative that has been effective is Industria 4.0, launched
in 2016 in order to finance the adaptation of Italian industry to the Fourth Industrial Rev
olution. However, the populist coalition that took office in 2018 did not relaunch that poli
cy.
It can be argued that the main industrial policy implemented in Italy is at the regional lev
el, mainly due to a reform of the constitution providing regions with competence over de
velopment policy, particularly industrial development. For instance, Bianchi and Labory
(2011b, 2019b) have analysed the effective industrial policy carried out in the Emilia-Ro
magna region in recent decades. Not all regions, however, have the institutional capacity
to define and implement appropriate industrial policies; southern regions are particularly
disadvantaged in this respect, and decentralization has not helped them.
21.5.4 France
The French government relaunched industrial policy in 2005, with the main objective of
promoting specialization in new, technology-intensive sectors. The two main measures
were the creation of ‘competitiveness poles’ and an Agency for Industrial (p. 615) Innova
tion (AII) that coordinated actions for the new sectors and also intervened in some of the
competitiveness poles. The policy thus constituted an attempt at decentralization, inviting
territories to identify possible synergies between activities so as to create these poles
Page 20 of 29
around specific productions. Each pole gathered together businesses, training and re
search centres, and the local government to reach the necessary critical mass for interna
tional visibility.
The mission of the AII was to promote and support big programmes of industrial innova
tion that would be led by large groups, hopefully leading to commercial success in new
markets. The AII would help with technological foresight and project selection, financing,
and monitoring. The initial budget amounted to €2 billion, mainly financed by privatiza
tions. However, the agency only operated for twenty-seven months. Programmes initiated
before 2008 are now followed by OSEO, a state-controlled company that supports finan
cial access to businesses, especially SMEs. In fact, the absorption of the AII into OSEO is
associated with a shift of industrial policy towards major support for small and medium-
sized firms (1 to 249 employees), as well as enterprises of intermediate size (entreprises
de taille intermédiaire, ETI, 250 to 5,000 employees), which are lacking in the country.
OSEO is characterized by decentralization, in the sense that it has regional offices which
allow it to be closer to territorial entrepreneurs. It favours collaborative projects between
SMEs in strategic sectors that have concrete commercialization prospects at the world
level. The objective is to help the emergence of European champions. A good part of the
financing provided by OSEO goes to competitiveness poles.
In 2012, a state holding, BPI France, was created, bringing together OSEO, CDC Enter
prises (public institutional investor), and the sovereign fund FSI (Fond Stratégique
d’Investissement). In other words, all sovereign funds existing in France at the time were
grouped into a single organization. BPI France supports enterprise growth and interna
tionalization through innovation. It is controlled by the state and the Caisse des Dépôts,
an institution that uses public funds to finance public projects.
The French state has thus returned to stronger intervention in markets. The turning point
was the intervention to save the enterprise Alstom, producer of railway equipment, from
failure in 2004, leading to a return of industrial policy in 2005. The new industrial policy
is pragmatic in that while horizontal measures are favoured, direct and selective inter
ventions are considered when necessary.
However, French industrial policy does not appear to have been consistent across the
years since 2005. While the general approach has continued to be innovation and support
to SMEs and firms of ‘intermediate’ size, measures have been constantly either changed
or re-dimensioned. For instance, in 2004 the objective of the competitiveness poles was to
bring together regional stakeholders to favour the launch of new products and process; in
2009, the objective was to realize ‘fabrics of projects’, while in 2014 it was again to real
ize new products. Each pole brings together a very large number of participants, appar
ently too many to focus projects and reach critical mass for each of them. In addition, the
overall financing of the poles, provided by the Fond Unique d’Innovation (FUI, Single In
novation Fund), has decreased by about 60 per cent over (p. 616) the period 2009 to 2018
(François, 2018).10 Evaluations of this policy do not appear to have assessed the perfor
mance of the poles in terms of product and firm creation, and business competitiveness,
Page 21 of 29
although Hassine and Mathieu (2017) show that the participation of SMEs and ‘interme
diate-sized’ firms has allowed them to invest more in R&D.
The overall objective is to rebuild ‘industrial France’ or develop industries of the future,
but the measures being taken are fragmented and inconsistent, lacking an industrial de
velopment vision for the country and a mobilization of efforts towards this vision. Indus
trial policy requires specific choices and substantial financial support, both of which are
missing. The present government’s project to build a New Industrial France is based on
the plan for industrial reconquest, comprising thirty-four plans, launched by the previous
government in 2013. As a result, French industrial policy has been essentially defensive.
Large firms have been saved from failure, such as Areva (nuclear sector), STX France
(shipbuilding), and Alstom. This contrasts with the German policy, which has been proac
tive.
A number of fiscal reliefs for business have been adopted by President Macron since his
election, but whether industries have used the money saved to invest in long-term growth
plans remains to be seen. What is clear is that the policy defined under the previous gov
ernment and pursued by the new one has not produced positive effects: the condition of
the French industrial sector has not substantially changed, with two million jobs in indus
try lost since the 1980s.
21.6 Conclusions
All European countries have followed similar trends in industrial policy. From a historical
perspective, industrial policy has been an instrument of the power of the state. Thus, con
tinental European countries implemented strong policies to industrialize and reach the
development level of the United Kingdom, pioneer of the First Industrial Revolution. This
was important both to raise the economic performance of the country, and hence its
wealth, but also to enhance its military capacity with new technologies and an increased
power of deterrence in international relations. Especially among the first industrializers,
economic and military, therefore political power, are intimately linked. This was also the
case after the Second World War, when the United States, but also European countries,
invested heavily in new technologies in a race to affirm their status as world powers. This
race in turn fed industrial development, since many innovations made for defence purpos
es had numerous industrial applications in a variety of sectors. Strengthening industry
has been an important strategy for China, which is now a world power.
Industrial policy has also been stronger in times of market disruptions caused
(p. 617)
both by changes in the extent of the market, such as the opening of international trade
and later globalization, and by technological progress, especially in industrial revolutions.
Once markets and production systems are established and incumbent firms defined, the
role of industrial policy reduces. However, when new technologies appear, or new busi
nesses with improved or renewed products enter markets or create new ones, established
Page 22 of 29
firms have to change their structure, and hence whole industries might be threatened or
in need of transformation, calling for government intervention.
Today the focus of industrial policy in European countries is on the adoption of the new
technologies of the Fourth Industrial Revolution to upgrade manufacturing processes, as
well as introducing new products that can encourage new industries. One notable differ
ence concerns regional imbalances: some countries have consistently addressed them
(Germany, Finland), while others have not (France and the United Kingdom). Yet a region
al (or local) focus is particularly required today for two reasons. First, globalization has
made the locational footprint more important for companies today, not less. Firms devel
op or locate in centres of excellence, where institutions are coherently defined and infra
structure well developed, and where human capital with appropriate skills and learning/
innovation capabilities is available. Territories not able to construct these strengths are
left behind (Bianchi and Labory, 2019b). Second, incumbent firms in established sectors
can only be challenged by small, innovative actors that develop from the bottom up. Terri
tories must therefore implement industrial policies in order to develop the capacity to
favour the emergence of these smaller, game-changing actors. This is also important to
maintain social cohesion in times of deep structural changes. Finland is an example, with
the failure of Nokia spurring important government programmes to favour the creation of
new, innovative firms at national and regional levels.
Attention is also required at higher levels of the multilevel governance (MLG) process.
Coherence and action are important between national and international levels of govern
ment, as well as between national and lower levels. In a globalized world, many issues in
troduced by the Fourth Industrial Revolution have to be addressed at an international lev
el, such as regulation of digital platforms and standards for emerging new products and
processes, such as robots and artificial intelligence. Environmental sustainability and cli
mate change are also global issues.
The discussion in this chapter adds another consideration regarding higher levels of the
MLG process: European countries on their own do not have adequate size and scope to
compete against the United States and China. Some truly joint and integrated efforts
should be made in order to favour competitive European businesses such as Airbus.
There are no European firms in a number of key sectors of the manufacturing revolution:
there are no European Googles, Amazons, or Alibabas; the most powerful supercomput
ers produced in the world today are produced in Japan, the United States, and China; and
there is only one European company (SAP, Germany) among the top ten big-data compa
nies in the world in 2019.
References
Ali-Yrkkö, Jyrki (ed.) (2010) ‘Nokia and Finland in a Sea of Change’. Helsinki: ETLA, Re
search Institute of the Economy of Finland, Taloustieto Oy.
Allen, Robert C. (2011) Global Economic History: A Very Short Introduction. Oxford: Ox
Page 23 of 29
Change and Economic Dynamics 48: 136‒50.
Bailey, David and Philip R. Tomlinson (2017) ‘Back to the Future? UK Industrial Policy af
ter the Great Financial Crisis’, in Philip Arestis and Malcolm Sawyer (eds) Economic Poli
cies since the Global Financial Crisis. Basingstoke: Palgrave Macmillan, pp. 221–64.
Bailey, David, Helena Lenihan, and Ajit Singh (2012) ‘Tiger, Tiger, Burning Bright? Indus
trial Policy “Lessons” from Ireland for Small African Economies’, in Akbar Noman, Kwesi
Botchwey, Howard Stein, and Joseph E. Stiglitz (eds) Good Growth and Governance in
Africa: Rethinking Development Strategies. Oxford: Oxford University Press, pp. 406–34.
Bairoch, Paul (1993) Economic and World History. Brighton: Wheatsheaf.
Barzotto, Mariachiara, Carlo Corradini, Felicia M. Fai, Sandrine Labory, and Philip R.
Tomlinson (2019) ‘Enhancing Innovative Capabilities in Lagging Regions: An Extra-re
gional Collaborative Approach to RIS3’, Cambridge Journal of Regions, Economy and So
ciety 12: 213–32.
Bianchi, Patrizio (2002) La rincorsa frenata. L’industria italiana dall’unità nazionale

all’unificazione Europea. Bologna: Il Mulino.
Bianchi, Patrizio and Sandrine Labory (eds) (2006) International Handbook of Industrial
Policy. Cheltenham: Edward Elgar.
Bianchi, Patrizio and Sandrine Labory (2011a) Industrial Policy after the Crisis: Seizing
the Future. Cheltenham: Edward Elgar.
Bianchi, Patrizio and Sandrine Labory (2011b) ‘Industrial Policy after the Crisis: The Case
of the Emilia-Romagna Region in Italy’, Policy Studies 32(4): 429‒46.
Bianchi, Patrizio and Sandrine Labory (2018) Industrial Policy for the Manufacturing Rev
olution. Perspectives on Digital Globalisation. Cheltenham: Edward Elgar.
Bianchi, Patrizio and Sandrine Labory (2019a) ‘Manufacturing Regimes and Transitional
Paths: Lessons for Industrial Policy’, Structural Change and Economic Dynamics 48: 24‒
31.
Bianchi, Patrizio and Sandrine Labory (2019b) ‘Regional Industrial Policy for the Manu
facturing Revolution: Enabling Conditions for Complex Transformations’, Cambridge Jour
nal of Regions, Economy and Society 12(2): 233–49.
Bianco, Magda (2003) L’industria italiana. Bologna: Il Mulino.
Blomström, Magnus and Ari Kokko (2002) ‘From Natural Resources to High-Tech Produc
tion: The Evolution of Industrial Competitiveness in Sweden and Finland’. EIJS Working
Page 24 of 29
Paper Series No. 139, Stockholm School of Economics, The European Institute of Japan
ese Studies.
Brander, James A. (1987) ‘Shaping Comparative Advantage: Trade Policy, Industrial Policy
and Economic Performance’, in Richard G. Lipsey and Wendy Dobson, (eds) Shaping Com
parative Advantages. Toronto: C. D. Howe Institute, pp. 1‒55.
Chang, Ha-Joon and Jostein Hauge (2019) ‘The Role of Manufacturing versus Services in
Economic Development’, in Patrizio Bianchi, Sandrine Labory, and Clemente Ruiz Duran
(p. 619) (eds) Transforming Industrial Policy for the Digital Age: Production, Territories
and Structural Change. Cheltenham: Edward Elgar, pp. 12‒36.
Chang, Ha-Joon, Antonio Andreoni, and Leon Kuan, M. (2013) ‘International Industrial
Policy Experiences and the Lessons for the UK’. Report for the UK Government’s Fore
sight Future of Manufacturing, London.
velopment: The Political Economy of Capabilities Accumulation. The Initiative for Policy
Dialogue Series. Oxford: Oxford University Press.
Cullen, Elizabeth (2004) ‘Unprecedented Growth, But for Whose Benefits?’, FEASTA Re
view 2: 9‒43.
Curzon-Price, Victoria (1981) Industrial Policies in the European Community. London:

Macmillan.
Darvas, Zsolt, Philippe Martin, and Xavier Ragot (2018) ‘European Fiscal Rules Require a
Major Overhaul’. Policy Contribution Issue No. 18, Bruegel Institute. Available at https://
bruegel.org/wp-content/uploads/2018/10/PC-18_2018.pdf.
Donges, Juergen (1980) ‘Industrial Policy in West Germany’s Not So Market-oriented

Economy’, World Economy 3(2): 185–204.
Dosi, Giovanni, Patrick Llerena, and Mauro Sylos Labini (2006) ‘The Relationships be
tween Science, Technologies and their Industrial Exploitation: An Illustration through the
Myths and Realities of the So-called “European Paradox”’, Research Policy 35: 1450‒64.
European Commission (1995) Green Paper on Innovation. Brussels: European Commis

sion.
Eyraud, Luc, Gaspar Vitor, and Tigran Poghosyan (2017) ‘Fiscal Politics in the Euro Area’.
IMF Working Paper No. 17/18. Washington, DC: IMF.
Federico, Giovanni and James Foreman-Peck (eds) (1998) Industrial Policies in Europe.
Page 25 of 29
Feldenkirchen, Wilfried (1998) ‘Germany: The Intervention of Interventionism’, in Giovan

ni Federico and James Foreman-Peck (eds) Industrial Policies in Europe. Oxford: Oxford
Financial Times (2018) ‘Is the UK’s New Industrial Strategy Starting to Work?’, 26 No
vember. Available at https://www.ft.com/content/bb242140-ef2b-11e8-89c8-
d36339d835c0.
Foreman-Peck, James and Leslie Hannah (1998) ‘Britain: From Economic Liberalism to
Socialism—and Back?’, in Giovanni Federico and James Foreman-Peck (eds) Industrial
Policies in Europe. Oxford: Oxford University Press, pp. 18–57.
François, Philippe (2018) ‘Pôles de compétitivité: 2018, Phase IV, le temps des résultats’.
Available at https://ifrap.org/emploi-et-politiques-sociales/poles-de-competi
tivite-2018-phase-iv-le-temps-des-resultats.
Freeman, Chris and Francisco Lourça, (2001) As Time Goes by: From the Industrial Revo
lutions to the Information Revolution. Oxford: Oxford University Press.
Gille, B. (1978) Histoire des techniques: technique et civilisations, technique et science.

Paris: Gallimard, Collection La Pleiade.
Hassine, Haithem-Ben and Claude Mathieu (2017) ‘Évaluation de la politique des pôles de
compétitivité: la fin d’une malédiction?’/France Stratégie, Document de Travail No.
2017/03. Available at https://www.strategie-gouv.fr.
Jäntti, Markus and Juhana Vatiainen (2009) ‘The Finnish Developmental State and its
Growth Regime’. Research Paper No. 2009/35. Helsinki: UNU-WIDER.
Ketels, Christian (2009) ‘Clusters, Cluster Policy, and Swedish Competitiveness in the
Global Economy’. Expert Report No. 30 to Sweden’s Globalisation Council, Stockholm.
Available (p. 620) at https://www.hhs.se/contentassets/
f51b706e1d644e9fa6c4d232abd09e63/
clustersclusterpolicyswedishcompetitivenessd6e53822.pdf.
Kirby, Peadar (2010) Celtic Tiger in Collapse: Explaining the Weaknesses of the Irish Mod
el. 2nd edition. Basingstoke: Palgrave Macmillan.
Labory, Sandrine (2006) ‘La politica industriale in un’economia aperta e basta sulla
conoscenza’, L’Industria 2: 255‒81.
Lall, Sanjaya (2006) ‘Industrial Policy in Developing Countries: What Can We Learn from
East Asia?’, in Patrizio Bianchi and Sandrine Labory (eds) International Handbook of In
dustrial Policy. Cheltenham: Edward Elgar, pp. 79‒97.
Landes, David (1969) The Unbound Prometheus. Cambridge: Cambridge University Press.
Levy-Leboyer, M. (ed.) (1996) Histoire de la France Industrielle. Paris: Larousse.
Page 26 of 29
Malerba, Franco (2000) ‘Il sistema innovativo italiano’, in Franco Malerba (ed.) Economia
dell’innovazione. Roma: Carocci, pp. 461–93.
Markgraf Jonas and Nicolas Véron (2018) ‘Germany’s Savings Banks: Uniquely Inter
twined with Local Politics’. Bruegel Institute. Available at https://bruegel.org/2018/07/
germanys-savings-banks-uniquely-intertwined-with-local-politics/.
Mokyr, Joel (1999) ‘Editor’s Introduction: The New Economic History and the Industrial
Revolution’, in Joel Mokyr (ed.) The British Industrial Revolution: An Economic Perspec
tive. 2nd edition. Boulder, CO: Westview Press, pp. 1‒127.
OECD (2019) Main Science and Technology Indicators. Available at https://read.oecd-

ilibrary.org/science-and-technology/main-science-and-technology-indicators/vol
ume-2019/issue-1_g2g9fb0e-en#page34.
O’Sullivan, Eoin, Antonio Andreoni, Carlos Lopez-Gomez, and Mike Gregory (2013) ‘What
is New in the New Industrial Policy? A Manufacturing System Perspective’, Oxford Re
view of Economic Policy 29(2): 432‒62.
Pelkmans, Jacques (2006) ‘European Industrial Policy’, in Patrizio Bianchi and Sandrine
Labory (eds) International Handbook of Industrial Policy. Cheltenham: Edward Elgar, pp.
45–78.
Rolfo, Secondo and Giuseppe Calabrese (2003) ‘Traditional SMEs and Innovation: The
Role of the Industrial Policy in Italy’, Entrepreneurship and Regional Development 15:
253‒71.
Schwab, Klaus (2016) The Fourth Industrial Revolution. Geneva: World Economic Forum.
Smith, Adam (1776) An Enquiry into the Wealth of Nations, reprinted in 1960 by J. M.
Dent & Sons Ltd, London (2 vols).
Vanden Bosch, Xavier (2014) ‘Industrial Policy in the EU: A Guide to an Elusive Concept’.
Egmont Paper No. 69. Brussels: The Royal Institute for International Relations.
Van Neuss, Leif (2015) ‘Why Did the Industrial Revolution Start in Britain?’, Working Pa
per, University of Liège.
von Tunzelmann, Nick (2010) ‘Technology and Technology Policy in the Postwar UK:
“Market Failure” or “Network Failure”?’, Revue d’économie industrielle 129‒130: 237‒
58.
Whelan, Karl (2012) ‘Golden Rule or Golden Straightjacket?’, VOX, CEPR Policy Portal, 25
February. Available at https://voxeu.org/article/golden-rule-or-golden-straightjack
et-critique-europe-s-fiscal-compact.
Ylä-Anttila, Pekka and Christopher Palmberg (2007) ‘Economic and Industrial Policy
Transformations in Finland’, Journal of Industry, Competition and Trade 7: 169‒87.
Page 27 of 29
Notes:
(1) An influential reflection on this subject was published by Jean-Jacques Servan-

Schreiber in 1968: ‘Le défi américain’ (the American Challenge).
(2) Cyprus, Czech Republic, Poland, Estonia, Lithuania, Latvia, Slovakia, Slovenia, Malta,
and Hungary in 2004; Bulgaria and Romania in 2007; Croatia in 2013.
(3) Markgraf and Véron (2018), Bruegel Institute, https://bruegel.org/2018/07/germanys-

savings-banks-uniquely-intertwined-with-local-politics/.
(4) National Association of German Cooperative Banks (2019), https://www.bvr.de/Press/

Press_releases/
Cooperative_Financial_Network_publishes_its_2018_consolidated_financial_statements_Stable_earnin
(5) https://ec.europa.eu/growth/industry/innovation/facts-figures/scoreboards_en.
(6) A triadic patent family is defined as a set of patents registered in the three major
patent offices: the European Patent Office (EPO), the Japan Patent Office (JPO), and the
United States Patent and Trademark Office (USPTO).
(7) Reuters, 5 February 2019: https://www.reuters.com/article/us-germany-industry/ger

man-state-ready-to-buy-company-stakes-to-protect-core-industry-idUSKCN1PU11D.
(8) For instance, in the Financial Times, ‘Is the UK’s New Industrial Strategy Starting to
Work?’ (2018).
(9) This has been stressed to the UK Parliament by economists of the Regional Studies As
sociation: https://publications.parliament.uk/pa/cm201213/cmselect/cmbis/writev/lep/
m11.htm.
(10) Computation by the authors from data shown in François (2018: 8).
Patrizio Bianchi
Patrizio Bianchi is professor of economics at the University of Ferrara, Italy, and has
been, since 2010, minister for education and research in the regional government of
the Emilia-Romagna region. Besides his academic activity, he has been an adviser to
Italian institutions such as the Prime Minister’s Office, the Minister of Industry’s
Cabinet and the regional government of Emilia-Romagna, and to international institu
tions such as the European Commission, the BID, and UNIDO. Patrizio has more than
200 publications, including books and articles in scientific journals. Publications on
industrial policy include Industrial Policy after the Crisis: Seizing the Future (Edward
Elgar, 2011) and Industrial Policy for the Manufacturing Revolution: Perspectives on
Digital Globalisation (Edward Elgar, 2018) with Sandrine Labory.
Sandrine Labory
Page 28 of 29
Sandrine Labory is associate professor of economics at the University of Ferrara,

Italy. She has a MSc in economics from University College London, and a PhD from
the European University Institute in Florence, Italy. Her research is focused on in
dustrial economics and policy, including comparative analysis of national and region
al industrial policies, innovative and productive processes at firm and territorial lev
els, structural changes and industrial development. She has published numerous arti
cles in Italian and international journals, as well as books such as Industrial Policy af
ter the Crisis: Seizing the Future (Edward Elgar, 2011) and Industrial Policy for the
Manufacturing Revolution: Perspectives on Digital Globalisation (2018, Edward El
gar) with Patrizio Bianchi.
Page 29 of 29
The European Union’s Industrial Policy

Michael A. Landesmann and Roman Stöllinger
This chapter reviews the industrial policy in the European Union in the light of the re
vived interest in the subject and the most pressing challenges ahead. In the current glob
al context these challenges are (i) to keep pace at the technology frontier with the techno
logically most advanced economies; (ii) to meet the challenges of fast catching-up emerg
ing economies; (iii) to contribute to the convergence and cohesion processes within the
European Union; and (iv) to deal with climate change and environmental sustainability is
sues more generally. A quantitative exercise that makes use of the European Union’s bud
get data, including the structural funds, and member states aid expenditures, is used to
identify the European Union’s current industrial policy priorities. The results are the ba
sis for an assessment of the extent to which the key challenges are addressed at the
supranational level and which aspects are primarily dealt with by national governments.
Keywords: industrial policy, EU industrial policy, smart specialization, mission-oriented industrial policy, EU struc
tural funds, EU cohesion policy, EU competitiveness
22.1 Introduction
INDUSTRIAL policy is back on the agenda around the globe. UNCTAD reports that since
2013 no less than eighty-four countries have adopted formal industrial policy strategies
(UNCTAD, 2018). The return of industrial policy (Wade, 2012) followed a period of indus
trial policy fatigue in the 1980s and 1990s. One of the reasons for the renewed interest in
industrial policy is dissatisfaction with the growth and structural outcomes of the non-in
terventionist period of the 1990s and early 2000s. In the European Union (EU), concerns
about de-industrialization in several member states, changing global context, as well as
the experiences of persistent regional disparities and the uneven impact of the financial
and economic crisis after 2008, have played their part in altering the way industrial poli
cies are perceived.
Page 1 of 50
Industrial policy in the EU has traditionally been a mixed approach, incorporating hori
zontal as well as sector-specific measures—with significant variation in focus throughout
the decades. More recently, policy frameworks have multiplied with new approaches
emerging from various fields ranging from Schumpeterian growth theories (Aghion et al.,
2015) to empirically driven patterns of industrial development across product spaces (Hi
dalgo et al., 2007) to new conceptual frameworks that guide policy initiatives at the EU
level, such as ‘mission-oriented’ policies (Mazzucato, 2018) or ‘smart
specialization’ (Foray et al., 2009; Foray, 2013).1 One unifying theme, though, (p. 622)
seems to be the rather general assertion that purely horizontal industrial policy is either
inadequate or simply not possible.2
Identifying the industrial policy stance of the EU is complicated by at least three factors.
First, due to the high importance assigned to competition policy within the single market,
subsidies are prohibited by default. More precisely, the European state aid rules prohibit
subsidies that distort or threaten to distort competition between member states, unless
they are justified by reasons of general economic development. While there are several
exceptions to the general ban (e.g. projects of common European interest such as the Air
bus endeavour), the assessment of the legitimacy of state aid is still governed by a mar
ket-failure approach. So the strict competition rules of the single market, itself part and
parcel of the EU’s industrial policy (and arguably one of its greatest successes), has to be
considered along with continued state aid support by EU member states, all of which
must have been cleared by the European Commission and deemed compatible with the
rules of the Single Market.
Second, and related to the point above, industrial policy action takes place at both mem
ber-state and EU levels. Furthermore, member states, to varying degrees, delegate the
formulation and implementation of industrial policy to the regional level. The same is true
for the EU’s cohesion policy, realized via the numerous European Structural and Invest
ment Funds (ESIF), a cornerstone of EU industrial policy.
Third, at the EU level, making sense of industrial policy action is difficult, given the nu
merous strategies, programmes, actions, and initiatives that are announced, as it seems,
at ever shorter intervals. Recently, the EU has developed the habit of publishing a new or
adopted industrial policy strategy every second year or so (see European Commission,
2010, 2011a, 2012, 2014, 2017a, 2020) with the latest one calling for investment in a
smart, innovative, and sustainable industry. Triggered by the European Commission’s
blocking of the planned Siemens-Alstom merger,3 intense negotiations about yet another
reformed industrial policy (EPSC, 2019; European Commission, 2019) are taking place at
this time. The rise of China with its geopolitically motivated One Belt, One Road (OBOR)
Initiative and its comprehensive and ambitious industrial policy strategy, entitled ‘Made
in China 2025’, is another major trigger for the new European industrial policy drive.
China’s successful catching-up process and ambitious industrial strategy shows that EU
industrial policy (be it at the supranational, the national or the regional level) is not con
ducted in a vacuum but is necessarily context specific. Changing global and (p. 623) socio-
Page 2 of 50
economic circumstances are shaping policy efforts and priorities. The contextual nature
implies that all discussions on European industrial policy are embedded in overarching
economic challenges the EU is facing, namely, the objectives that it has set itself in the
Lisbon Agenda and subsequent strategies. Interestingly, for the EU some of the chal
lenges and associated objectives of industrial policy have remained surprisingly constant
over decades. Others have been added following the EU’s growth in both size and diversi
ty or with new global trends such as climate change and the deterioration of the environ
ment more generally. The most pressing economic challenges may be summarized as fol
lows:
(i) Technological/innovation challenge: Keeping pace with the technologically most

advanced economies in future-oriented technological areas, which includes the digi
tal transformation. This has been one of the long-standing aspirations of EU industri
al policy. In order to defend the European Union’s position at the technology frontier
in many technology-driven industries, it is imperative to build on inherited strengths
of industrial, technological, and human capital structures.
(ii) Emerging markets challenge: Meeting the challenge of fast-catching-up emerg
ing economies through both competition and collaboration. In this, China merits spe
cial attention due to its dynamic growth as well as both its geopolitical ambitions re
alized via OBOR and its technological ambitions as laid down in ‘Made in China
2025’, the country’s ambitious industrial policy strategy.
(iii) Cohesion challenge: To foster progress in the field of convergence and cohesion
policy within the European Union. A major issue in an industrial context here is the
fact that peripheral regions and countries, in order to make progress in their initial
industrialization or structural upgrading process, are confronted with strong agglom
eration tendencies inherent in tradable activities in core areas of Europe’s economy.
(iv) Environmental challenge: Dealing with the challenges of the Paris Agreement on
climate change and sustainability issues more generally. Given the comparatively
strong regulatory framework for environmental protection, European industry is
bound to adjust and incorporate resource efficiency, emission reduction, and sustain
ability in a cost-efficient manner. The implications for the competitiveness of Euro
pean industry of such a shift to the aspired circular economy are as yet unclear and
differ across industries. On the one hand, regulations that force firms to increasingly
rely on sustainable or re-usable goods (including packaging) may develop first-mover
advantages in a series of new materials and technologies. On the other hand, initially
several of these regulations will impose additional costs on firms producing within
the EU.
The structure of the remainder of this chapter is as follows. Section 22.2 briefly reviews
the evolution of industrial policy in Europe and of the supranational (EC and then EU) lay
er in particular. Subsequent section 22.3 puts numbers on the relative importance of
(p. 624) EU industrial policy action. This quantitative exercise identifies the policy priori
ties based on member states’ (compulsory) notifications of state aid4 to enterprises and at
the EU level based on budgetary expenditures on (broadly defined) industrial policy mea
sures. This represents an important supplement to the conventional analysis of European
Page 3 of 50
industrial policy based on the study of policy documents. The quantitative approach
seems appropriate as an effective industrial policy requires not only a sound policy formu
lation but also proper implementation5 including the appropriate funding.6 The penulti
mate section, section 22.4, discusses the extent to which current industrial policies and
the identified priorities can contribute to the key challenges ahead. The conclusion, sec
tion 22.5, presents recommendations for EU industrial policy for the future.
22.2 The Evolution of EU Industrial Policy in a

Nutshell
The history of industrial policy in the EU can be divided into three phases: an early inter
ventionist phase (1950‒80), a liberal phase (1980‒2005), and the current phase which
Bianchi and Labory (2006) have dubbed the ‘pragmatic’ phase.7 The three phases and
some of their main features as well as selected key measures and initiatives are summa
rized in Table 22.1 and briefly reviewed in the following sub-sections (see also Bianchi
and Labory, Chapter 21 this volume).
22.2.1 The Interventionist Phase (1950‒80)
During the interventionist phase, European industrial policy was mainly conducted and
shaped by national governments. As a consequence, industrial policies in most of the six
founding members of the European Economic Community (EEC) were nationally
oriented.8 The public policy grip on industry was much stronger back in the 1950s and
throughout the 1970s than it is nowadays.
Despite notable differences in type and intensity of industrial policy across member
states, a common feature was the identification of, and particular support for, ‘strategic
(p. 625) industries’. There was a considerable overlap with respect to the targeted indus
tries, defined on the basis of specific economic characteristics such as strong linkages
with other parts of the economy and high-technology content (e.g. aircraft, computers,
automotive industry) but also social considerations such as their importance for employ
ment (e.g. coal and steel, shipbuilding) (Owen, 2012). Equally important, network indus
tries with public goods characteristics (i.e. public infrastructure such as electricity gener
ation, telecommunications, railway companies) were administered almost exclusively by
state-owned companies.
Page 4 of 50
Table 22.1 The three phases of industrial policy in the European

Union
Interventionist Liberal phase Pragmatic phase

phase (1950–80) (1980–2005) (since 2005)
Member-state lev
el
Heyday of sector- Full shift of emphasis Return of specific

specific industrial on framework and industrial policies
policies horizontal policies resulting in an in
dustrial ‘policy mix’
Blend of interven Policies influenced by Industrial policies

tions in sunset in the Washington Con constrained by fis
dustries, high-tech sensus; gradual with cal and competition
industries, and pub drawal of national rules
lic utilities subsidy policies
Nurturing of na Privatization of utili Initiatives towards

tional champions ties ecological transfor
mation (some mem
ber states)
Experimentation Active FDI policies

with European
Champions (Airbus,
1970)
EU level
Creation of the Emergence of EU technol Europe 2020

Common Market ogy policy (ESPRIT, 1980) Strategy
(1957) and the
Customs Union
(1960)
Competition rules Tightening of EU competi Further focus

in place (1957) but tion policy: modernization on RTI policies
largely ineffective of state aid rules (2012);
introduction of merger
control (1989)
Page 5 of 50
Sectoral policies European Economic Act ICT-sector pro

for coal, steel, and gramme
nuclear energy
Lisbon Strategy (2000) fo Strategic In

cused on knowledge-based vestment Plans
competitiveness
Smart Speciali
sation Strategy
(2012) institu
tionalized in re
gional policy
Mission-orient
ed industrial in
novation poli
cies (planned)
Global Europe
Strategy (2006)
Source: Authors’ own representation using the periodization in

Bianchi and Labory (2006).
In countries such as France and the United Kingdom,9 the selection of strategic
(p. 626)
industries was inextricably linked with the aim of creating national champions, that is,
dominant domestic companies that should become successful in international markets.10
National champions are the most characteristic element in the European industrial policy
set-up during the 1970s. Taking place in an environment of comparatively protected do
mestic markets (by the standards of the 2010s), the plan for creating a national champion
typically consisted of a complex web of numerous support measures, including direct sub
sidies and preferential loans but also public procurement, the provision of technology by
government-supported agencies, merger promotion, industrial diplomacy, export credits,
and subsidies to domestic upstream suppliers to ensure access to strategic inputs (Owen,
2012). In France, such endeavours were often linked to a ‘grand project’, a prestigious
technological undertaking, which constituted the primary mission of the future superstar
firm.11
Underlying these ambitious industrial policy projects was a belief in the need for techno
logical (and in the case of France also military) independence, especially from the United
States. This shows that even half a century ago, industrial policy in Europe was already
driven by concern about an existing technology gap vis-à-vis the United States. Hence,
one encounters here Europe’s enduring objective of matching (or catching-up with) the
Page 6 of 50
United States in strategic technologies or defending leadership positions already

achieved, such as in car manufacturing. However, more often than not, such as in the
highly subsidised computer industry, international competitiveness and export success
did not materialize.
In addition to the high number of failures and the high cost to taxpayers involved, the di
rigiste policy approach in the industrial sector also lost appeal due to the creation of the
EEC and the common market (the predecessor of the single market), followed by the com
pletion of the customs union in 1968. The European Common Market,12 establishing the
free movement of goods, services, capital, and people, was complemented with far-reach
ing competition rules, including a general prohibition of state aid,13 to ensure its proper
functioning. The enforcement of the competition rules was (p. 627) entrusted to the Euro
pean Commission. This was a remarkable arrangement in a period that was still charac
terized by massive government subsidization and protectionism at the national level, even
if the general prohibition on trade-distorting aid was qualified by a number of exceptions
(e.g. regional aid or aid to projects of common European interest). Nevertheless, control
over state aid by sovereign governments was obviously a delicate issue and the European
Commission had to show a high degree of pragmatism in this respect (Doleys, 2013). In
the historical context of the time this meant that the state aid rules were only laxly en
forced during the early phase of the common market.
Hence, one could argue that it was dissatisfaction with the outcome of interventionist
policies in a more open-economy environment, together with the fiscal strains imposed by
them, rather than the introduction of stringent competition rules, which caused the indus
trial policy pendulum to swing towards the more market-oriented policies that were to
dominate for the following quarter of a century.
22.2.2 The Liberal Phase (1980‒2005)
The 1980s saw a remarkable shift in the industrial policy focus towards less intervention
ism and a more market-oriented policy stance. Pioneered by the United Kingdom, which
joined the EEC in 1973, sectoral planning and support for individual companies was se
verely reduced or abandoned altogether. While also in this case there was considerable
heterogeneity with regard to the stringency and comprehensiveness of the policy shift,
there can be no doubt that competition, ‘horizontal’ measures, and improving the frame
work conditions increasingly replaced the quest for national champions. Especially in the
United Kingdom the emphasis on market principles and competition implied a wave of
privatizations that included most public utilities, including British Telecom in 1984
(Owen, 2012).
A second novel element of the new industrial policies, which were increasingly referred to
as ‘enterprise policy’, was the vanishing resistance against foreign takeovers, including
by US and Japanese companies, though this was truer in the United Kingdom than, for ex
ample, in France, which continued to insist on domestic ownership of the country’s key
industrial assets for much longer.
Page 7 of 50
An event of greater geopolitical significance during this phase was the collapse of the
Soviet bloc and the dramatic transformation of the former communist regimes of Eastern
Europe. Relevant for this discussion is that the radical system change discredited any
form of state planning. This led to a dismantling of existing state capacities that were
deemed to be compromised, and there was a general suspicion of anything that smacked
of ‘industrial policy’. The substitute for industrial policy was the attractiveness of quite a
few of the Central and Eastern European (CEE) economies to foreign direct investment.
This led to a process of ‘re-industrialization’ in some of the CEE economies (those close to
the German–Austrian borders) after an initial phase of (p. 628) collapse of industrial pro
duction. This drive was led by Western European firms that found the existing skilled (and
comparatively low-waged) labour force attractive. Various support measures by govern
ments (such as export-processing zones) encouraged FDI before accession of the CEE
countries to the EU, which then had to be withdrawn to comply with the Acquis Commu
nautaire. Importantly, the geographic closeness to the high-wage industrial zones in Ger
many and Austria encouraged the build-up of vertical cross-border production networks
(see Stehrer and Stöllinger, 2015; Stöllinger, 2016).
These developments (both in Western and Central and Eastern Europe) suggest a certain
degree of convergence in industrial policies across countries between the 1980s and the
middle of the 2000s. This convergence was further nurtured by the changed economic en
vironment characterized by more open economies, the appearance of new players in the
global trade arena and, not least, the emergence of regional and global production net
works which called for an adjustment and, as it turned out, a retreat of active industrial
policies (Landesmann and Stöllinger, 2019).
Less clear was the growing importance of the EU, which emerged as a more relevant
player in the industrial policy sphere during the second half of the liberal phase. The
heightened importance of the supranational level in the field of industrial policy was both
a contributing factor to the convergence in national policies and a consequence of mem
ber states’ growing belief in ‘horizontal policies’ that were seen as compatible with free
and fair competition in open markets.
EU industrial policy took shape in three quite distinct directions. The first and most far-
reaching of these was the Single Market Programme introduced in 1987. Its purpose was
the deepening and strengthening of the already existing common market by further har
monization of regulations and a stronger integration of services markets. As such, the
Single Market Programme is the archetypical framework policy and constitutes the top
layer of industrial policy defining the boundaries and possibilities of other EU industrial
policy activities.
Second, the existing competition rules were broadened and strengthened. Competition
law,14 including state aid rules, merger control, and public procurement, can be seen ei
ther as a part of industrial policy or as its natural antagonist. In any case, it acts as a dis
ciplining layer of industrial policy, monitoring and disciplining state interventions by na
tional governments and uncompetitive behaviour by firms. The third layer, which
Page 8 of 50
emerged during this phase, comprises active horizontally designed industrial policy at the
supranational level, mainly in the domain of technology R&D collaboration and innovation
as well as regional policy (Figure 22.1).
Figure 22.1 The industrial policy structure of the EU
Note: VI = vertical initiatives, including projects of

European interests. Dark grey areas indicate the EU
level, light grey areas indicate the level of member
states.
Source: Authors’ own representation.
Third, active, horizontally designed support policies targeting R&D collaborations were
fostered, as were regional policies, due to the growing heterogeneity of an enlarged EU.
Early attempts to spur R&D collaboration at the EU level in the 1980s were largely
(p. 629) unsuccessful,15 at least, with a view to achieving the main objective of closing the
technology gap with the United States (and increasingly with Japan) in advanced indus
tries such as electronics or aircraft. Despite the limited successes of the early R&D coop
eration programmes, technology and R&D policy remained the cornerstone of the EU’s
support for industrial development. This is reflected in both the wording of the Article on
Industrial Policy16 that was introduced with the Treaty of Maastricht in 1993 and also in
the Lisbon Agenda, the EU’s major growth strategy devised in 2000.17 The new article on
industrial policy, which anchored the competences of the European Commission in this
area in the EU treaties,18 however, has never attained great practical relevance.
With regard to regional policy, which dates back to 1975 with the establishment of the
European Fund for Regional Development (EFRD),19 it was the successive rounds of en
largement20 that advanced ‘cohesion’ to become one of the main EU policy domains. This
implies that the European Structural and Investment Funds (ESIF) are concerned with
distributive impacts (Schwengler, 2013). As will be seen, the scale of spending is quite
large, providing finance not only for investments in infrastructure but (p. 630) for a wide
array of activities including labour-market and qualification measures, technology and en
vironmental policy measures.
The allocation of funds from the ESIF changed fundamentally with the EU enlargement of
2004 because of the much lower per capita income of the CEE countries entering com
pared to the ‘old’ member states. Payments from the ESIF became an essential source for
the financing of infrastructure investments in the CEE countries, which also made exten
sive use of European funds to attract foreign direct investment (FDI).
Page 9 of 50
Despite the propping up of activities in the domain of innovation and technology policy
and the heightened role of regional policy, the EU’s approach to industrial policy through
out the liberal phase is best described as horizontal in design, aiming at creating
favourable framework conditions and correcting market failures, both with the ultimate
objective of supporting the industrial competitiveness of EU firms. While it was to be
come more heterogeneous and multi faceted over time, this approach would also remain
dominant during the subsequent pragmatic phase.
22.2.3 The Pragmatic Phase (Since 2005)
Starting sometime between 2005 and the global financial crisis, the current pragmatic
phase of industrial policy is hard to characterize as it is still evolving. Some authors ar
gue that the Commission’s blocking of the planned Siemens‒Alstom merger in 2019 has
initiated a fourth phase of European industrial policy (see EPSC, 2019). In this view, mem
ber states, dissatisfied with the tough EU competition rules against the background of
growing international competition (often deemed unfair), demand a loosening of the com
petition rules in order to (re-)establish a level playing field—especially vis-à-vis Chinese
companies. However, it seems too early for such a conclusion, so for the purpose of this
chapter the pragmatic phase is considered to be still ongoing.21
It was nevertheless true that a renewed interest in more active industrial policies devel
oped, especially in member countries that experienced a crippling loss of competitiveness
of their industries within the European Monetary Union (vs. Germany) but also those
which had undergone waves of de-industrialization which generated not only macroeco
nomic problems but also deep structural regional imbalances.
The straightjacket of EU competition policy and the interdependencies in an increasingly

integrated world economy tightly linked by global value chains made it (p. 631) impossible
to reactivate the industrial policy toolkit of the past. As a consequence, member states
had to turn to new designs for their industrial strategies that were both compatible with
the single market and in particular with its competition rule, on the one hand, and more
effective (and therefore more specific) policies than the purely horizontal approach fol
lowed until the early 2000s, on the other.
The attempts by governments to square this circle have been diverse, ranging from bare
ly noticeable changes in policy (e.g. in most CEE member states which essentially held
onto their FDI attraction policies) to a quite explicit revitalization of industry, technology,
or skill-specific policies. In all this it is important to note that the search for new industri
al policy tools and concepts did not mean that the cherished preference for horizontal pol
icy objectives was abandoned (Radosevic, 2017).
Hence, the focus on R&D and innovation policy remained, though arguably with a slight
twist as member states increasingly try to link their R&D and innovation efforts to soci
etal challenges such as the digital transformation or mitigating climate change. Likewise,
the EU’s main research support programme, the Horizon 2020 programme, features ‘so
cietal challenges’, one of which is Climate Action, Environment, Resource Efficiency and
Page 10 of 50
Raw Materials.22 This way, ‘Industry 4.0’23 and ‘green’ industrial policies aiming at a
structural transformation towards renewable energies and circular (instead of linear) re
source flows became integral parts of R&D and innovation policies. The Digital Agenda
had already made an appearance in the European Commission’s industrial policy strategy
of 201024 and has since then advanced to become one of the key themes in subsequent
strategies.25
The link between industrial policies and societal challenges further led to the incorpora
tion of ‘mission-oriented innovation policies’ (Mazzucato, 2013) in the industrial policy
frameworks of several member states. Such missions are less encompassing than the
great societal challenges they should help solve and, above all, are defined in a way that
clearly allows assessment of whether the mission has been accomplished.26 At the EU lev
el, missions will become part of the EU’s relabelled Horizon Europe (formerly Horizon
2020) programme in the next multi-annual financial framework period (2021‒7) (see Eu
ropean Commission, 2017b).27
Global value chains (GVCs) and the implied geographically dispersed production
(p. 632)
processes are another phenomenon that new industrial policy concepts try to take into
account. The significant impact of GVCs on changing international trade and investment
patterns in the EU and beyond implies a great potential for achieving industrial policy ob
jectives. In general, however, in the context of industrial policy, GVCs seem so far to have
become a buzzword that has entered the policy discourse but without any substantial ef
fect on policy formulation, not to mention on actual implementation. For example, the Eu
ropean Commission’s industrial policy strategy calling for a European Industrial Renais
sance (European Commission, 2014) urges member states to implement measures that fa
cilitate the integration of EU firms in GVCs to boost their competitiveness. No guidance is
provided though as to what kind of measures would be appropriate, nor are types of val
ue chains or segments of value chains suggested that firms, regions, and countries should
try to get involved in. This raises a question mark over the relevance of the participation
in GVCs in the current EU industrial policy toolkit. This is regrettable as the analysis of
functional specialization (i.e. specialization by value-chain functions such as R&D, pro
duction, or marketing) shows marked differences in functional profiles across member
states (see Stöllinger, 2019). The general pattern is that the more advanced member
states specialize in knowledge-intensive activities such as R&D or managing the produc
tion network (‘headquarters functions’) while especially the Central and Eastern Euro
pean member states act as ‘factory economies’ (Baldwin and Lopez‐Gonzalez, 2015) re
sponsible for the actual production (including final assembly).
Equally related to the global economy, an important change in EU trade policy, with con
sequences for industrial strategies, occurred in 2006 with the adoption of the Global Eu
rope strategy (European Commission, 2006). The strategy implies a marked shift away
from a multilateral (WTO-based) approach to liberalization and towards a bilateral liber
alization strategy. Since then, the EU has been concluding a series of deep and compre
hensive free-trade agreements with several extra-European partners, including Korea,
Japan, and Canada. Given the current disturbances in the global trade architecture ema
Page 11 of 50
nating from the United States, the EU may be forced to change its very liberal stance to
wards international trade, too. Already a marked shift in the EU’s trade policy is notice
able. It seems that the EU is less convinced of the advantages of unilateral liberalization.
And indeed, especially for the poorer member states, asymmetric liberalization efforts
combined with tough EU competition rules and questionable practices by partner coun
tries that follow a state capitalism model poses severe competitive challenges. The aban
donment of the belief in the advantages of unilateral liberalization is also visible in the
2016 reform of the EU’s public procurement rules.28 A key aspect of this reform is the
principle of reciprocity, which means that bidders from third countries that do not open
their public procurement markets to EU companies can in turn be barred from EU ten
ders for public contracts.
In the realm of regional policy, a new variant of place-based innovation-oriented

(p. 633)
industrial policy, known as smart specialization strategies (see Foray et al., 2009)—or 3S
for short—is currently being propagated by the European Commission. It is seen as a tool
to improve the technological capacities especially of lagging European regions, which
could be one way to reduce the currently vast technological asymmetries (as evidenced
by the complementary functional specialization patterns) within the EU. For this reason,
the formulation of a smart specialization strategy has become a (formal) conditionality for
regions to access funds from the EFRD for R&D and innovation purposes.
Finally, there are calls for more projects of European interest, which so far have been
rare, Airbus being one of the few examples. The new French industrial policy calls for the
‘Airbus of tomorrow’ in fields such as energy, multimedia networks, or emissions-free ve
hicles (Cohen, 2007).
Hence, for the time being the EU follows a ‘mixed’ industrial policy approach (which con
tinues to be embedded in the single market and subject to the competition rules). Accord
ing to the European Commission (2014) this policy mix consists of a ‘horizontal basis’
which considers all sectors as important and tailor-made approaches to all these sectors.
To what extent this view is compatible with the growing importance of mission-based
thinking about industrial policy and the concept of smart specialization strategies re
mains to be seen. This issue will be taken up again in section 22.3.
22.3 EU Industrial Policy Priorities: The Num

bers
There is no lack of documents outlining the EU’s industrial policy strategies. Typically,
these documents put forward a large number of objectives and initiatives. Since 2010, the
Commission has launched five industrial policy strategies, each supplementing and fine-
tuning the previous ones. The 2012 strategy (European Commission, 2012) is strongly in
fluenced by the fall out from the global recession of 2008‒9 and the subsequent Euro
pean debt crisis, and therefore focuses on investment in new technologies and innovation
—a recurring topic throughout all strategies. The document defines six investment priori
Page 12 of 50
ties (‘priority action lines’) plus ten further fields for key action (ranging from developing
‘raw materials diplomacy’ to matching skills with jobs, to the introduction of the concept
of smart specialization with regard to the use of regional funds).
The industrial policy strategies are complemented by countless communications, strate

gies, and initiatives which are in one way or another related to the EU’s industrial policy
strategy. The updated industrial policy strategy of 2017 (European Commission, 2017a)
lists no less than forty-eight such communications, strategies, and initiatives, including
the Strategy on Digitising European Industry, the New Circular Economy (p. 634) Pack
age, the extension of the European Fund for Strategic Investment (EFSI), or the free-
trade agreement with Japan.
While the strategy documents are a valuable source for learning about the EU’s general
approach to industrial policy and main objectives, to some extent the vast number of
strategy documents and the varying priorities obscure, even if unintentionally, the true
priorities of the EU’s industrial policy. This chapter therefore supplements the informa
tion on industrial policy priorities available in official documents with data on actual
spending on policy objectives that are related to industrial policy. The two main sources
of data used for this quantitative exercise are the Multi-Annual Financial Framework
(MFF)—the EU’s budget plan—for the period 2014‒17 and the EU State Aid Scoreboard
(SAS) which records the value of the aid elements involved in the subsidies granted by
member states.29 SAS data for the period 2014‒17 are used to align with the MFF data.
The analysis reveals first of all the relative importance of spending on industrial policy at
the EU level on the one hand and at the level of member states or regions within member
states on the other (level dimension). At both levels the data are further dissected to iden
tify the priorities of industrial policy (thematic dimension), and differences in these the
matic priorities across member states or groups of states are discussed (country dimen
sion).30
22.3.1 The Level Dimension: Industrial Policy Spending at EU and

Member-state Levels
Against the background of the historical developments in EU industrial policymaking, it is

not surprising that on the whole, expenditure on industrial policy by member states far
exceeds the amounts spent at the supranational level (Figure 22.2). The financial re
sources from the EU budget flowing to industrial-policy-related measures amounted to
0.35 per cent of the EU’s GDP annually during the period 2014‒17.31 In contrast, indus
trial policy spending by member states was in the order of 0.75 per cent of GDP during
the same period. The latter figure includes, apart from the state aid provided, (p. 635)
member states’ contributions to EU programmes financed by the ESIF, which are known
as co-financing by member states.
Page 13 of 50
Figure 22.2 Spending on industrial policy in the Eu

ropean Union, average 2014–17
Note: EU funding is the sum of expenditure items re

lated to industrial policy in the MFF over the years
2014–17. Co-financing amounts by member states
are estimations which are calculated as the ratio be
tween planned allocation and actual spending by
member states in the MFF 2014–20. All these expen
ditures are expressed as nominal values in per cent
of GDP. National State Aid by member states is the
subsidy element contained in national state aids, also
expressed in per cent of GDP.
Source: European Commission (state aid scoreboard)

available at http://ec.europa.eu/competition/
state_aid/scoreboard/index_en.html; DG budget data
(EU expenditure and revenue 2014–20) available at
http://ec.europa.eu/budget/graphs/
revenue_expediture.html; statistics on European
Structural and Investment Funds (ESIF) available at
https://ec.europa.eu/regional_policy/en/funding/avail
able-budget/), authors’ calculations.
While the subsidy spending by member states obviously forms part of their national in
dustrial policy activities, the co-financing part should, in principle, also reflect to some ex
tent European industrial policy objectives, although specific projects financed by the
ESIF are selected and designed by national or regional governments.
The finding that member states’ funding of industrial policy exceeds that at the EU level
by more than two to one32 also reflects the fact that industrial policy is a shared compe
tence with only subsidiary powers of the supranational level, while member states as
sume the role of main actors in this domain (see also Wyns, 2017).
The combined spending on industrial policy by member states and the EU amounts to
about 1.1 per cent of EU GDP. This is certainly a non-negligible amount of public support
but is low by historical standards, given that during the 1970s subsidy spending in the EU
is estimated to have amounted to 3 per cent of GDP and was still around 2 (p. 636) per
cent during the 1980s (European Commission, 2011b). This can be interpreted in two dif
ferent ways. Either the proclaimed ‘renaissance’ of industrial policy in Europe has not
materialized, or the new industrial policies of the twenty-first century are not just old
Page 14 of 50
wine in new bottles but do indeed have some new features. In particular, these new poli
cies are often described as being more subtle, less interventionist and better aligned with
market forces, with governments or their agencies sometimes only acting as facilitators
(‘honest brokers’).33 In this narrative the comparatively low amounts spent at the aggre
gate level reflect the fact that the new industrial policies involve less of the old-fashioned
‘hard’ support policies and more ‘soft’ policies resulting in less government expenditure.
22.3.2 The Thematic Dimension: Priorities in Industrial Policy
The categorization of industrial policy in this exercise is mandated by the structure of the
MFF and reflects various drivers of competitiveness. More precisely, the following broad
support categories are distinguished: (i) research, development, innovation, and technolo
gy (RDI&T); (ii) support for small and medium-sized enterprises (SMEs); (iii) employment,
education, and training; (iv) ecological transformation (‘green’ industrial policy); and (v)
investment in infrastructure. Two further categories, sectoral industrial policy and re
gional policy, are added, though the latter can be further allocated to the other thematic
fields in the case of industrial policy spending at the EU level.
Page 15 of 50
Table 22.2 EU and member states’ spending on industrial policy by policy field, 2014‒17 (annual averages)
EU industri State aid by EU industri State aid by EU industri State aid by

al policy member al policy member al policy member
states states states
Industrial Absolute amounts (in $ bn) % in GDP % of total IP spending/state

Policy Field aid
Ecological 0.29 51.90 0.00 0.35 0.01 0.52

transforma
tion
Employment, 0.44 3.13 0.00 0.02 0.01 0.03

education,
training
Infrastruc 1.44 0.00 0.01 0.00 0.03 0.00

ture
RDI and 7.47 9.07 0.05 0.06 0.14 0.09

technology
Regional pol 40.96 11.13 0.28 0.08 0.79 0.11

icy
SME support 0.09 4.68 0.00 0.03 0.00 0.05
Page 16 of 50
Sectoral in 1.32 7.02 0.01 0.05 0.03 0.07

dustrial poli
cy
Other 0.00 12.33 0.00 0.08 0.00 0.12
TOTAL 52.01 99.25 0.35 0.67 1.00 1.00
Note: See Figure 22.2. Co-financing amounts by member states are excluded in the table.
Source: European Commission; authors’ calculations. For details see Appendix A.22.6.2.
Page 17 of 50
Table 22.2 shows the absolute amounts spent on these industrial policy fields, along with
their shares in GDP and their share in overall industrial policy spending.
At the supranational level, the data reveal two main policy priorities: regional policy and
support for RDI and technology. Taken by itself, the EU’s regional policy, that is, the activ
ities of the ESIF,34 accounts for the lion’s share of EU industrial policy spending. Forty bil
lion euros or almost 80 per cent of the total industrial policy-related expenditure at the
EU level is disbursed via the ESIF. The fact that industrial policy spending at the EU level
is dominated by regional policy has important implications for policy formulation. This is
because in contrast to industrial policy measures financed via the central EU budget (e.g.
the Trans-European Networks), the projects financed by the ESIF are designed and se
lected by the member states or their regions.35 Hence, with regard to the design of poli
cies it is useful to distinguish between EU industrial policy in a narrow sense, which is fi
nanced and predominantly designed by the EU institutions, and the EU’s regional indus
trial policy.
The second key policy at the EU level, support for RDI and technology policy, ac
(p. 637)
counts for another 14 per cent of the EU’s industrial policy budget. As will be discussed
in detail, the main policy vehicle in this domain is Horizon 2020, a grant-based pro
gramme for research institutions, enterprises, and universities which finances collabora
tive research activities of all kinds.
Moving on to the state aid activities of the member states, Table 22.2 suggests that the in
dustrial policy priorities at the national level are quite different from those at the supra
national level. While member states also spend considerable amounts on RDI and technol
ogy policies (€9 bn) and regional policy (€11 bn), the most prominent state aid category is
green industrial policy,36 that is, state aid granted for ecological transformation. The im
portance assigned to ‘green’ industrial policy measures is attributable above all to
Germany’s massive expenditure on the Energiewende (energy transition), parts of which
constitute state aid.
Also noteworthy is that during the period 2014‒17, only marginal sums were spent on
sector-specific industrial policy. This is true for the EU level where €1.3 bn (3 per cent)
was spent on specific industries, essentially space, aircraft, and electronics, (p. 638) and
€7 bn (7 per cent) by member states. In the latter case, this includes restructuring and
bail-out aid, for example, for airlines (former national carriers).
Returning to the EU level and the ESIF, Figure 22.3 presents the same data but this time
with the ESIF funds split into corresponding categories of industrial policy.37
Page 18 of 50
Figure 22.3 Funding of industrial policy in the Euro

pean Union by themes, average 2014–17
Note: EU Industrial Policy ex ESIF is the sum of ex

penditure items related to industrial policy in the
MFF over the years 2014–17. For the EU Regional
Funds, the total actual financing flows by the ESIF
are assigned to individual funds using the shares of
these funds in the planned budget allocations in the
MFF 2014–20. Co-financing amounts by member
states are calculated using the ratio between
planned allocation and the actual spending by mem
ber states in the MFF 2014–20. National State Aid by
member states is the subsidy element contained in
national state aids, also expressed in per cent of GDP.

At the EU level, this presentation of the expenditure on industrial policy confirms the
point made above that EU industrial policy predominantly coincides with R&D and inno
vation policy. Approximately 70 per cent of ‘central’ EU industrial policy spending is des
tined for RDI and technology. The other two categories of industrial policy that are of
high relevance in financial terms are infrastructure investments and sector-specific poli
cies, which accounted for 13 per cent and 12 per cent of the industrial policy spending
from the ‘central’ EU budget respectively.
The EU’s infrastructure-related industrial policy is embedded in its Trans-European Net

work (TEN) policy. This policy comprises several instruments including the (p. 639) Con
necting Europe Facility (CEF), which provides financial support for strategic investment
in transport, energy, and increasingly digital infrastructure.38 A key objective of the CEF
is to create pan-European energy networks and European transport corridors in order to
overcome Europe’s nationally fragmented energy and transport systems. While this objec
tive is laudable, the effectiveness of the CEF is curtailed by persistent financing gaps in
Page 19 of 50
relation to member states’ declared needs (European Commission, 2017c) and arguably
also by their nationally oriented priorities.
One particularly obvious aspect of the EU’s infrastructure-related industrial policy is that
policy efforts are scattered across too many programmes. While the fact that infrastruc
ture, including the TEN, is apart from the CEF also financed via the ESIF is understand
able given the different policy objectives, the need for an additional infrastructure pro
gramme that was set up with the European Fund for Strategic Investments (EFSI)39 is
less clear. The EFSI, better known as the ‘Juncker Plan’ after its main promoter, former
European Commission president Jean-Claude Juncker, was set up as a response to the fi
nancial crisis. The investment plan was initially intended as a temporary fiscal stimulus
measure40 but has been extended until 2020. In the meantime, it is clear that the EFSI
will become a permanent EU programme in the MFF 2021‒27 under the name EUInvest.
Unfortunately, the design of the EFSI leaves much to be desired, constituting as it does
an old-style industrial and investment approach. To start with, its actual budget is much
too small to have a great impact. Contrary to the €315 billion announced, the actual funds
provided by the EU amounted to only about €3.5 billion for the original three-year period,
with some additional funds coming from the European Investment Bank. The difference
stems from the fact that official EU documents include the private investments that would
supposedly be mobilized by the EFSI. While the figures for the investments actually trig
gered and published by the European Commission suggest that the target of €315 billion
has even been exceeded, auditors have cast doubt on these figures, arguing that the
claims by the European Investment Bank (EIB), which is managing the EFSI, have been
overstated.41
Given the current experiences with the EFSI, it also remains unclear what is strategic
about the fund. The areas eligible for financing by the EFSI are so broad and diverse that
hardly any priorities are discernible. The fields of operation include energy and transport
projects, but also support to SMEs, R&D, and the digital economy.42 (p. 640) Agriculture
and social matters are also covered. Hence, to some extent the EFSI, and the future In
vestEU programme, is likely to duplicate existing programmes. The EFSI is already linked
to the Connecting Europe Facility (CEF), with part of its budget simply redirected from
the CEF, the EU’s main infrastructure investment programme.
The greatest problem of the EFSI, however, is that it does not seem to live up to its pro
claimed principle of additionality. The EFSI was supposed to only support programmes
that otherwise could not be realized (which, by the way, was one of the explanations giv
en for the high leverage effects). A look at the project list, however, casts doubt on the ad
ditionality of projects and the dire need of the beneficiaries for public support from EU
funds. A prime example is the purchase of new rolling stock (CityJets) for regional pas
senger transport by the Austrian Railways at a total cost of €1.7 billion. The project is
supported with a €500 million loan from the EIB backed by an EFSI guarantee. In all like
lihood the Austrian railways would have bought this new rolling stock anyway as the new
material was long overdue. This is just one example demonstrating that, in all likelihood,
Page 20 of 50
the support provided by EFSI is plagued by rent-seeking behaviour of firms who hope to
benefit from improved financing conditions thanks to EFSI.
In summary, the EFSI as well as the InvestEU programme are examples of a poorly tar
geted, old-style industrial policy programme and a rather unfortunate attempt to revive
EU industrial policy. As will be shown, though, the EU is also following more promising in
dustrial policy avenues.
Staying in the domain of infrastructure-related industrial policy but moving to the Euro
pean Structural Funds, an interesting finding regarding the ESIF’s spending profile is
that infrastructure is no longer the primary thematic objective. Though it is still impor
tant, accounting for roughly a fifth of total ESIF spending, the number one priority within
the ESIF according to the categorization employed is funding for ecological transforma
tion, that is, regional green industrial policy.
While it is reassuring to see that the EU is providing funds for the ecological transforma
tion of industry and infrastructure, it should be borne in mind that the projects are select
ed by member states. From this perspective it is surprising that ‘green industrial’ policy
enjoys much greater prominence (as measured by budgetary outlays) in national industri
al policy budgets than at the EU level.43 This is surprising in so far as the mitigation of
climate change, and the protection of the environment more generally, constitute an ar
chetypal global public good (see, e.g. Rodrik, 2014). As such one would expect that the
design and funding of green industrial policies would be ideally placed at the suprana
tional level.
Given the quantitative importance of national governments’ state aid expenditure in EU

industrial policy spending, the thematic priorities set by member states are strongly de
termining the major industrial policy spending patterns. Overall, 40 per cent of industrial
policy spending in the EU is ‘green industrial policy’ of some sort, an issue to which we
shall return. (p. 641)
22.3.3 The Country Dimension: Substantial Differences in Industrial

Policy Expenditure
Going beyond EU averages across member states and zooming into the activities of indi
vidual member states (or groups of member states),44 Figure 22.4 reveals a great deal of
heterogeneity in spending priorities at both governance levels.
A first key insight is that the dominance of industrial policy funding from member states’
own resources, which emerged in the aggregate, is not present across all country groups.
In particular, for the CEE member states and the Southern cohesion countries, the indus
trial policy funding received from the EU budget exceeds national expenditures. The ratio
between industrial policy funding received from the EU and the national sources is 1:3 in
the case of the CEE-5 and even 1:7 for the EU Balkan countries (Croatia, Bulgaria, and
Romania).
Page 21 of 50
Figure 22.4 Spending on industrial policy from the

EU and from member states, by country groups, av
erage 2014–17
Note: EU-Balkans = HR, BG, RO; Other small EU-15

= AT, BE, NL, LU, IE; CEE-5 = HU, CZ, SK, PL, SI;
EU South = CY, EL, ES, MT, PT. For the EU Regional
Funds, the total actual financing flows by the ESIF
are assigned to individual funds using the shares of
these funds in the planned budget allocations in the
MFF 2014–20. Co-financing amounts by member
states are calculated using the ratio between
planned allocation on EU spending and the actual
spending by member states in the MFF 2014–20. Na
tional State Aid by member states is the subsidy ele
ment contained in national state aids equally ex
pressed in per cent of GDP.

At the lower end of the spectrum is Germany which, due to its high expenditure
(p. 642)
on energy transition, has a funding ratio of only 0:1 (or 10:1 in favor of national industrial
policy spending). For this reason, Germany is also the country with the highest aid expen
diture among the EU-15, not only in absolute terms but also in relation to GDP. Including
the funding for industrial policy obtained from the EU budget, German expenditure on in
dustrial policy amounted to 1.4 per cent of GDP, slightly higher than the EU average of
1.1 per cent.
The reason for the patterns of industrial policy spending observed is that the CEE and the
Southern cohesion countries are the main beneficiaries of the ESIF as they have more re
gions eligible for funding from the ERDF and the Cohesion Fund. Therefore, the pro
nounced differences in industrial policy spending across member states are, to a large ex
tent, explained by the large differences in income levels within the bloc.
Page 22 of 50
The combination of comparatively high state aid levels and generous receipts from the
ESIF resulted in expenditures for industrial policy amounting to 3.7 per cent of GDP for
the CEE-5 (with 2 per cent of GDP coming from the EU budget); 3.5 per cent for the
Baltic countries; and 2.9 per cent for the EU Balkan countries. These figures are high
even by historical standards given that the amount of state aid was estimated to have
been around 3 per cent of GDP during the 1970s, the heyday of industrial policy. This has
to be borne in mind when discussing possible industrial policy initiatives, especially be
cause the effectiveness of various ESIF programmes has been shown to decline with the
amount of funding received (e.g. Becker et al., 2012). Hence, in some regions there may
be a risk of overspending. In these cases, new initiatives should not be added to existing
support programmes but should be substituted for some of the less successful support
mechanisms in place.
Section 22.3.4 shows that industrial policy spending across member states not only varies
with regard to the amounts spent but also targets different thematic fields.
22.3.4 The Country Dimension: Different Countries, Different Priori

ties
Differences in industrial policy priorities across member states are discernible in the
funds received from the central budget (Figure 22.5, panel a) and those channelled via
the ESIF (Figure 22.5, panel b).
The most striking feature in the central EU budget is doubtless the dominance of industri
al policy spending on RDI and technology. For the Nordic countries, the United Kingdom,
and the other small EU-15 member states, RDI and technology accounts for more than
three-quarters of the total funding of industrial policy. The shares are similarly high for
Germany and Italy. The most notable exception here is the Central and Eastern European
member states. For the CEE-5, the Baltic countries, and the EU Balkan countries, the cor
responding share is about 40 per cent or less. This comparatively smaller role of RDI and
technology support is compensated by larger (p. 643) (p. 644) spending on infrastructure
financed by the EU budget, mainly through the Trans-European Network (TEN) projects.
Page 23 of 50
Figure 22.5 Spending on industrial policy from the

EU budget, by country groups and themes, average
2014–17

EU South = CY, EL, ES, MT, PT. The numbers are the
share of the respective industrial policy theme in a
country or country group’s total EU policy spending
funded by either the central budget or the ESIF.

While the distribution of funds is more evenly split across the different industrial policy
fields in the case of the ESIF, the pattern that infrastructure-related industrial policy is
more important in the Central and Eastern European member states is also discernible in
the ESIF budget. The same is true for the more limited role of RDI and technology poli
cies in these countries compared to the EU-15.
It also seems that the share of funds destined for ecological transformation account for a
relatively larger share of the overall industrial policy spending financed via the ESIF in
the CEE-5, the Baltic countries, and the EU Balkan countries than in the other member
states. This might reflect the legacy in Central and Eastern European countries still deal
ing with the communist-era priority given to heavy and energy-intensive industries. Given
member states’ state aid spending profile, however, this seems to indicate a compensa
tion for a lack of domestic spending rather than policy priorities.
Page 24 of 50
Finally, it should be mentioned that France’s high share of sector-specific industrial poli
cies in the total funds it received from the EU budget does not reflect the country’s sup
posedly interventionist stance on industrial policy but is linked to France’s leading role in
the European Earth Observation Programme, now called Copernicus, which the EU is op
erating jointly with the European Space Agency (ESA).
Returning to the more general differences across countries in Figure 22.5, it is worth
mentioning that the stronger focus on infrastructure-related industrial policy detectable
for the Central and Eastern European member states and the stronger emphasis on RDI
and technology policies in the EU-15 signal different needs of member states. These dif
ferent needs in turn are to a large extent again explained by the still prominent differ
ences in development levels across member states.
The key policy question in this respect is to what extent not only the EU’s industrial poli
cy spending, but also the possibilities for member states in terms of state aid spending,
should accommodate for these differences. This is a delicate issue as it touches directly
upon the competition rules, that is, the disciplining layer of EU industrial policy and one
of the building blocks of the single market. An alternative (or complementary) approach
would be to ensure that the national innovation systems (NISs) in the Central and East
ern European member states are significantly strengthened in order to allow these coun
tries to benefit to a greater extent from the EU’s R&D and innovation grants and pro
grammes.
In this context the Horizon 2020 programme, the EU’s main R&D flagship programme, is
of key importance, as is the thematic objective R&D and technology within the ERDF.
Both are highly relevant, not only because of the significant amounts of support involved
but also because they incorporate new industrial policy approaches.
The Horizon 2020 programme supports collaborative R&D and innovation projects to the
tune of €7.5 billion in the form of grants allocated via competitive funding. Its main objec
tives are to stimulate excellent science, build industrial leadership, and provide solutions
to the grand societal challenges. The programme is structured into (p. 645) three pillars:
excellent science (targeting ‘frontier research’);45 industrial leadership (targeting closer-
to-the-market research focused on ‘key enabling technologies’);46 and societal challenges
(European Commission, 2017b).47
The most interesting pillar of the Horizon 2020 programme and of its successor in the
coming MFF, the Horizon Europe programme, is the grand societal challenges pillar. Here
the European Commission has discovered the mission-oriented approach to industrial and
innovation policy and seems determined to fully embrace it in the Horizon Europe pro
gramme.48
The mission-oriented approach to industrial and innovation policy is rooted in a systemic

perspective of innovation and is based on a number of assumptions and characteristics.
The first assumption is that successful innovation necessarily involves a large number of
actors, both private and public—hence the notion of an ‘entrepreneurial
Page 25 of 50
state’ (Mazzucato, 2013). Second, the impact of public funds is strongly enhanced if the
money is not spent on narrowly defined and often isolated projects but channelled to re
lated research endeavours that all serve a common purpose. Third, with regard to granu
larity,49 research money should be linked to clearly defined missions. Ultimately, research
projects will serve a grand societal challenge such as health, secure societies, and cli
mate change. Missions are one level down in the granularity (compared to societal chal
lenges) with the key features that they are clearly specified and, above all, measurable.
Unfortunately, for the time being it is not quite clear how these missions will be defined in
an EU context, although some concrete proposals exist (see Mazzucato, 2018). An imple
mentation-related issue of the EU’s main research programmes continues to be the un
even distribution of funds. So naturally, the question of who defines these missions and
who will participate in such missions will arise. Already now, the bias in favour of the
most advanced EU member states is observable in the applications for Horizon 2020
grants (Figure 22.6).
Figure 22.6 Share of Horizon 2020 applications per

EU member state, average 2014–16

EU South = CY, EL, ES, MT, PT.
Source: European Commission (2018). Authors’ own

calculations.
The EU’s regional funds have incorporated elements of new industrial policy thinking. In
particular, since 2016, support from the ERDF in the field of research, technological de
velopment, and innovation (thematic objective 1)50 require that the region has a Research
and Innovation Strategy for Smart Specialisation in place. This (p. 646) ‘ex ante
conditionality’ for accessing funds from the ESIF confirms the view that smart specializa
tion has gained a strategic and central function within the new Cohesion Policy of the EU
(Foray et al., 2009). In quantitative terms, the actual expenditures during the period 2014
Page 26 of 50
to 2017 amounted to approximately €9 billion annually, including co-financing by member

states.
The Strategy for Smart specialization is a regional and innovation-oriented form of indus
trial policy developed within Europe. The central idea underlying S3 is the ‘entrepreneur
ial discovery process’ (Hausmann and Rodrik, 2003). Entrepreneurial discovery is central
because it is the process by which regions can find out about promising niches in which
to specialize. The rationale for the government to get involved in entrepreneurial discov
ery processes (which in principle should be the responsibility of firms) is that first
movers, that is, firms that invest in products that are new to the region or country, can
not, if successful, reap the full benefits from their ‘discovery’ (the well-known ‘externali
ty’ impact of an innovating firm à la Schumpeter). This is the case because, once it has
been revealed that a certain product or service can be successfully produced in the re
gion, other firms from the region will enter production, driving down economic rents ac
cruing to the first mover. Economic rents, however, would be necessary to cover the fixed
costs associated with the investment in the discovery process. The likely scenario is,
therefore, a too-low number of attempts to discover profitable new niches (Hausmann
and Rodrik, 2003). Public support—that is, S3—is therefore needed to encourage entre
preneurial discovery processes. Hence, it is assumed here that countries and regions pos
sess latent comparative advantages which can be turned into effective comparative ad
vantages with the appropriate policy interventions.
While the economic model by Hausmann and Rodrik (2003) would apply to both innova
tion and (physical) investment for the production of products new to the region, (p. 647)
the concept of smart specialization has been developed in the context of R&D and innova
tion policies (see Foray et al., 2009). S3 has been designed as a place-based (i.e. regional)
innovation-oriented industrial policy.
Despite the huge potential of the place-based S3 there is, however, also scepticism about
the appropriateness of this policy for less-developed regions (e.g. Capello and Kroll,
2016). Based on case studies, Trippl et al. (2019), for example, found that while there is
policy learning in less-developed regions, they are struggling with the issue of stakehold
er involvement either due to a lack of capable stakeholders (e.g. universities) or the unfa
miliarity of actors with this type of cooperation. Radosevic (2017) argues that for less-de
veloped regions the switch from an individual entrepreneurial discovery process to a col
lective one is difficult to achieve. Also, in order to avoid conflicts regarding alternative
entrepreneurial opportunities of a region, stakeholders would agree on the lowest com
mon denominator, which often leads to the selection of very broad technological areas. In
such environments the value added of S3 can be expected to be rather limited.
Page 27 of 50
22.4 EU Industrial Policy Efforts in Light of the

Major Challenges Ahead
The wide spectrum of industrial policy and the large set of measures make the overall as
sessment of industrial policy an elusive task. This is all the more true as EU programmes
and initiatives set different, potentially conflicting objectives and have different underly
ing intervention logics (see also Pellegrin et al., 2019). The potential conflict in objectives
becomes evident when considering the four major challenges of industrial policy that
were discussed in the introduction to this chapter. It is against these challenges that the
strengths and achievements as well as the weaknesses and neglects of the current EU in
dustrial policy are discussed in this section. The discussion pays due attention to the fact
that industrial policy, to be effective, needs to be not only well formulated but also appro
priately implemented and continuously evaluated.
22.4.1 The Technological/Innovation Challenge
As regards the long-standing challenge of defending or attaining technological leader

ship, many excellent studies assess the record, scope, and current ambitions of EU inno
vation policy, evaluating the EU’s policy initiatives in the areas of electrical vehicles, digi
tization, the ‘battery alliance’, and others (Veugelers et al., 2018; Veugelers and Bal
tensperger, 2019). In fact, the initiatives and measures taken by the EU in the area of in
novation policy can be considered one of the success stories of EU (p. 648) cooperation.
This is no small achievement given, as was confirmed by the quantitative exercise, that
‘innovation policy’ is at the core of what the EU sees as ‘industrial policy’.
As described above, European R&D and innovation policy could be further sharpened by
fully embracing the mission approach within the upcoming Horizon Europe framework
programme.51 These missions will be defined within the realm of five clusters that have
already been determined.52 Certainly, the effectiveness of the mission approach depends
strongly on the type of missions that are going to be defined. Among the five clusters, cli
mate, renewable energy and the environment figure prominently, which provides an ex
cellent opportunity to link the technological challenge to the environmental challenges.
At this stage it is noteworthy that the Horizon Europe programmes may also tackle the
digital transformation which, for many, is the key technological challenge lying ahead of
EU economies with potentially transformative power for the EU’s entire production sys
tem. While most EU countries seem to be well-positioned to master the digital challenge,
given its innovation and industrial capabilities (see, e.g. World Economic Forum, 2018),
concerns are nurtured by the fact that EU industry does not possess any major global
player in the Internet business.53 This may indeed turn out to be a disadvantage for the
transition towards Industry 4.0, in which cyber-physical systems combining the physical
and virtual worlds are set up. While it is far too early to judge whether the EU will be
falling behind, it is clear that the EU economy will need heavy investment in digital tech
nologies and the required infrastructure. Efforts in this respect vary significantly across
Page 28 of 50
member states,54 which is why a strong impetus originating at the EU level (as envisaged
by the Digital Union) is required.
As in the past with mechanization, and more recently with IT, encompassing technological
changes go along with rapid changes in the composition and the demand for different
types of skills, and changes in occupational structures with some occupations disappear
ing and others emerging. This requires strong adaptation processes from the labour
force, and in education and training programmes. European industrial policy requires a
two-pronged approach: on the one hand, to be at the forefront of technology develop
ments and tapping into the full potential of such new technologies and production
processes for productivity advances and improving ‘the comforts of life’; on the other
hand, to make a major effort to also deal with the distributive and adjustment challenges
of the impacts of rapid technological and structural change.
Against this background, innovation support plays an ever-bigger role in EU in

(p. 649)
dustrial policy. And while European innovation policy has strongly improved since its
shaky beginnings in the early 1980s, there is one important shortcoming which needs to
be dealt with. At the moment European innovation-oriented industrial policy is mostly ori
ented towards the most advanced economies, companies, and regions that are at the fore
front of technological developments. Ideally, the EU’s industrial policy should be con
ceived in a way that gives equal prominence to the developmental needs of those regions,
countries, and (industrial and company) segments that are not on (or close to) the global
technology, in other words, Europe’s middle-ranked as well as peripheral regions, and
countries that have to overcome massive structural challenges.
Some of these countries are seriously lagging or falling behind in economic development.
Again, this assessment is supported by the figures: even though the cohesion countries
benefit disproportionately from industrial policy-related support programmes from the
EU budget—especially when the ESIFs are included—the share of R&D and innovation
support going to the CEE member states is comparatively small. This is related to the ten
dering process through which R&D grants are allocated. While the open tendering
process is doubtless a strong point in the design of EU R&D and innovation support pro
grammes, it is also clear that it favours countries with strong national innovation systems
(NISs).
Essentially, there are two ways to address this issue. One is to further adjust the tender
rules to make sure research consortia that apply for Horizon/Europe 2020 funds become
more balanced geographically with regard to the participating institutions. The alterna
tive approach is to refocus the ESIF towards creating and improving the NISs of CEE
member states. This would entail the creation of tertiary teaching institutions like the Eu
ropean University Institute in Florence in a number of other member states. Another im
portant avenue would be support for technical research agencies comparable to the
Frauenhofer Institute in Germany.
Page 29 of 50
22.4.2 The Emerging Markets Challenge
The dominance of EU industrial policy in support of technological leadership globally to

some extent deflects from the challenge posed by catching-up economies and also the co
hesion process (discussed in sub-section 22.4.3). Both are crucial in the circumstances in
which the EU (and closely connected neighbouring countries) currently finds itself. EU in
dustrial policy needs to face vital structural development issues which the European
economy confronts in today’s global environment. As recognized early on by authors such
as Rosenstein-Rodan (1943), Hirschman (1958), and Gerschenkron (1962), such structur
al development issues were traditionally at the centre of industrial policy strategies (see
also Rodrik, 2009; Wade, 2012; Cimoli et al., 2009) and the EU would be well advised to
reconsider their importance in the formulation of its future industrial policy initiatives.
The structural development issue is closely connected to the emergence of new

(p. 650)
major players in the global trade arena,55 first and above all China, because the greater
overlap in economic structure between the CEE and Southern EU member states and
emerging economies exposes them to fiercer competition. The key concern here is not the
sectoral composition of economies which have strongly converged within the EU. Rather,
it is the more recent functional and vertical dimension of specialization that was opened
up with the emergence of GVCs.56 Recent empirical evidence in this respect (Stöllinger,
2019; Timmer et al., 2019) suggests that the international division of labour with regard
to value-chain functions (e.g. R&D, head office functions, production, back offices) within
the EU continues to be very asymmetric, or in fact complementary,57 with CEE member
states specializing predominantly in production-related activities. In this respect they re
semble countries such as Mexico and the economies in the third and fourth row of the
Southeast Asian flying-geese formation (e.g. Vietnam, Laos, or Thailand). Baldwin and
Gonzalez-Lopez (2015) termed countries which specialize functionally in this manner ‘fac
tory economies’, in contrast to ‘headquarter economies’ whose companies specialize pre
dominantly in R&D and in organizing global value chains.
For these reasons it is not surprising that the ‘China shock’ has affected the CEE and
Southern European member states much more in terms of exports than the rest of the EU
(Ciani and Mau, 2019; Celi et al., 2018).
22.4.3 The Cohesion Challenge
Tightly linked to the stiff global competition presented by emerging economies, but still
in itself a challenge, is the EU internal cohesion process. As has been emphasized
throughout this chapter, a lot of funds are dedicated to cohesion objectives via the EU’s
ESIF. Yet, despite the substantial amount of money spent on cohesion objectives, ‘periph
erization’ has become a critical issue both at regional and country levels. This develop
ment has contributed to the issue of ‘North‒South imbalances’ and severe tensions have
arisen with respect to the formulation of a proper fiscal and structural policy framework
at the EU level to deal with such imbalances (see Celi et al., 2018).
Page 30 of 50
The issue of external imbalances is closely related to the phenomenon of agglomeration

of industrial activity (see Stehrer and Stöllinger, 2015), but also to that of other tradable
activities (business services, financial centres; see also Landesmann, 2015 and 2019 on
this). Apart from the macroeconomic implications, agglomeration phenomena have an in
herent characteristic of giving rise to cumulative virtuous and vicious circles (p. 651) that
can deepen persistent disparities across European regions. Economic geography has
pointed to such dynamic processes where concentration of economic clusters goes along
—especially in the context of the single market with its four freedoms—with factor mobili
ty and thus with disparate demographic developments and the concentration of skills
(people with high levels of training, competence, and education) in some centres and the
outflow of such skills and a deteriorating age profile in other regions (see, e.g. Martin,
2008; Iammarino et al., 2019). This cumulatively saps development potential from periph
eral regions and countries, which has been shown to also have strong political implica
tions (see Rodriguez-Pose, 2018). All these trends raise some questions about the effec
tiveness of the EU’s cohesion policy and the ESIF, an issue on which the empirical evi
dence is still inconclusive.
As mentioned earlier, a very promising avenue for improving the effectiveness of the EU’s
cohesion efforts is the concept of smart specialization, which has been incorporated in
the ERDF, one of the major EU funds. The smart specialization strategies, as a territorial
approach, are argued to favour the relatively poorer countries in the EU, that is, cohesion
and CEE countries. Still, whether this is really the case in the context of the ERDF pro
grammes remains doubtful as several cohesion countries are struggling with numerous
implementation issues that are related to existing deficiencies in their (political) institu
tions. One way to tackle this issue is to change the support mechanism of the ESIF from a
grant-based system to a support system that relies on financial instruments, essentially
EU-backed guarantees administered by the EIB and involving commercial banks that
would provide the actual finance for eligible projects. This would reduce the role of politi
cal institutions in the selection and implementation process, but it might also make the
support less targeted, which must be seen as a major disadvantage in an era where indus
trial policy aspires to be linked to missions and societal challenges.
22.4.4 The Environmental Challenge
The more recent but arguably the most pressing and fundamental challenge which ex
tends far beyond industrial policy and in fact beyond the economic sphere, is climate
change and environmental protection more generally. Doubtlessly, this challenge has re
ceived due attention more recently at EU and national levels through the formulation of
both mission-oriented policies’ and ‘innovation policy’ (i.e. policies directed at the techno
logical leadership challenges), as well as within the context of regional policy (and thus
under the schemes supported by cohesion funds).
As with S3 policies, it will be important to monitor how effective ‘mission-oriented’ poli

cies (which take a systemic view of policy interventions) turn out to be once they have
been implemented for some time and can be evaluated. Given the comparatively strong
Page 31 of 50
regulatory framework in environmental issues, European industry is bound to adjust and

incorporate resource efficiency, emissions reduction, and sustainability. The implications
for the competitiveness of European industry of such a shift towards the aspired circular
economy are as yet unclear and are not the same across different (p. 652) industries. On
the one hand, regulations that force firms to increasingly rely on sustainable or re usable
goods (including packaging) may develop first-mover advantages in a series of new mate
rials and technologies. On the other hand, initially several of these regulations will im
pose additional costs on firms producing within the EU. In a longer-term perspective,
however, the balance can be expected to be positive, especially if the mission approach is
successful and leadership in environmentally sustainable technologies can be obtained.
Acting quickly in order to secure first-mover advantages could be vital in this respect as
Korea’s and China’s latest industrial policy strategies are also clearly targeting ‘green’
technologies and green industrial policies more generally. Unfortunately, acting fast is not
exactly the EU’s greatest strength. Still, the integration of mission orientation and smart
specialization into the EU’s industrial policy formulation is an important step in the right
direction. Ideally, we would see the EU’s three layers of industrial policy interacting with
each other to support the defined missions and the territorial approach demanded by the
smart specialization concept. The coming years will show whether EU member states are
capable of moulding the various interests and objectives into well-defined missions whose
accomplishment will help meet the key challenges ahead.
22.5 Conclusions
Starting with a brief recap of historical developments, this chapter went beyond the usual
investigation of official documents and made a serious attempt to reveal the actual
amounts spent on and priorities targeted by EU industrial policy. A first, rather unexpect
ed, finding is that the supranational level, while on average still second to national gov
ernments’ state aid activities, has become substantial too. Especially for the CEE member
states, the funding of industrial policy by EU programmes is of utmost importance and
much larger than national expenditure. Given the substantial amounts disbursed via the
EU’s regional and structural funds, the overall industrial policy budgets for this country
group are large even by historical standards, and comparable to those seen in the 1970s.
If one adds to the EU industrial policy programmes the European Commission’s regulato
ry power in areas such as competition policy (including inter alia state aid, mergers and
acquisitions, and public procurement) and trade policy, it is obvious that the EU level has
become a central pillar, even if formerly member states remain the primary actors.
The growing importance of the EU institutions in designing and also implementing indus
trial policy is welcome given that none of the key challenges, ranging from technological
leadership to environmental transformation, can realistically be met by any of the mem
ber states individually. The flip side of this is that the EU’s industrial strategies as a
whole, as well as individual programmes, need to be designed and implemented in a co
herent and effective manner. With the incorporation of mission- (p. 653) oriented thinking
Page 32 of 50
about industrial and innovation policies and at the regional level the concept of smart
specialization, the EU has made formidable progress in the formulation of industrial poli
cy. The effective implementation of the sophisticated strategies, however, in many in
stances remains curtailed by policy incoherence (with rival or even contradictory objec
tives) and a host of programmes and initiatives, many of which lack critical mass. If the
official announcements are to be believed, the upcoming budget period (2021‒7) will see
a noticeable reduction in the number of programmes and the elimination of duplications.
With regard to policy priorities, the existing focus of industrial policy on R&D and innova
tion support is likely to remain and these areas will be strengthened, according to the Eu
ropean Commission’s current budget proposal for the period 2021‒7. What seems impor
tant here is that efforts in the realm of R&D and innovation policy are not only aiming at
the technological frontier and the leading edge vis-à-vis the United States (and increas
ingly China). Rather, it is imperative that, in line with the initial smart specialization con
cepts, R&D and innovation support is also a policy tool for the EU cohesion countries.
While this is fully acknowledged by official rhetoric, the actions on the ground seem to be
different, as regions from CEE economies (and laggards such as Bulgaria and Romania in
particular) are underrepresented in the various pilot projects and R&D cooperation initia
tives within the smart specialization strategy (e.g. when it comes to European Innovation
Hubs).
An additional aspect of the EU R&D and innovation strategies is that they too suffer from
an overload of rival programmes. While this is understandable, given that such pro
grammes are necessarily a compromise, reflecting the interests and priorities of all mem
ber states, it is necessary to have better-defined technological priorities which are de
rived from a vision of the technological trajectory European societies want to embark on.
This is where a mission-oriented policy comes into play again. Guidance on how this tra
jectory may look can be taken from national governments’ industrial policy priorities,
where the state aid figures clearly identify the ecological transformation as the number
one target.58 In other countries, concentration on avoiding a drift into a vicious circle of
‘peripherization’ and aiming towards a more balanced pattern of economic development
across the European Union, will be the priority.
Whatever member states decide the missions are going to be (which will be enshrined in
the next R&D framework programme, the Horizon Europe programme), it should be clear
that they need to be limited in numbers, ideally not more than three to four at any time.
Otherwise, there is a risk that projects for the achievement of the (p. 654) missions remain
underfunded so that the fragmented policy efforts will not develop their full potential.
Provided member states identify the missions and the underlying societal challenges, it is
also likely the projects at the regional level funded by the structural funds are aligned
with the defined missions. If the smart specialization strategies work out, each region
should find its niches in the key enabling technological domains needed for completing
the missions. Hence, what is needed first of all is a consensus on appropriate missions
Page 33 of 50
which, if achieved, would substantially support the EU’s competitive position in

tomorrow’s industries.
Page 34 of 50
22.6 Appendix
22.6.1 Member States and Country Groupings
Table A22.1 List of member states, country codes, and country groups
Country Iso-3-code EU code Country group
Austria AUT AT Other small EU-15
Belgium BEL BE Other small EU-15
Bulgaria BGR BG EU Balkan
Cyprus CYP CY EU South
Czechia CZE CZ Central Europe (CEE-5)
Germany DEU DE Germany
Denmark DNK DK Nordic
Estonia EST EE Baltics
Greece GRC EL EU South
Spain ESP ES EU South
Finland FIN FI Nordic
France FRA FR France
Croatia HRV HR EU Balkan
Hungary HUN HU Central Europe (CEE-5)
Ireland IRL IE Other small EU-15
Italy ITA IT Italy
Lithuania LTU LT Baltics
Luxembourg LUX LU Other small EU-15
Page 35 of 50
Latvia LVA LV Baltics
Malta MLT MT EU South
Netherlands NLD NL Other small EU-15
Poland POL PL Central Europe (CEE-5)
Portugal PRT PT EU South
Romania ROU RO EU Balkan
Sweden SWE SE Nordic
Slovenia SVN SI Central Europe (CEE-5)
Slovakia SVK SK Central Europe (CEE-5)
United King GBR UK United Kingdom

dom
Source: Authors’ own country groupings.
22.6.2 Selection of EU Budget Items and Assignment to Indus

(p. 655)
trial Policy Themes
22.6.2.1 Mapping of budget items in the MFF 2014‒20 to the industrial poli
cy themes
The discussion of the EU’s budgetary expenditure on industrial policy-related issues dur
ing the period 2014‒17 is based on a selection of items within the Multi-Annual Financial
Framework (MFF) 2014‒20, which are subsumed in the realm of industrial policy. Table
A22.2 lists all budget items that were selected to qualify as industrial policy related. The
numbers used refer to actual disbursements.
Page 36 of 50
Table A22.2 Industrial policy item within the MFF 2014–20
Section Description Assigned indus

in MFF trial policy cat
egory
1.1.10 European Fund for Strategic In Infrastructure

vestments (EFSI)
1.1.11 Implementation and exploitation of Sectoral policy

European satellite navigation sys
tems (EGNOS and GALILEO)
1.1.13 European Earth Observation Pro Sectoral policy

gramme (Copernicus)
1.1.31 The Framework Programme for RDI

Research and Innovation (Horizon
2020)
1.1.4 Programme for the Competitive SME

ness of Enterprises and Small and
Medium-sized Enterprises
(COSME)
1.1.6 EU Programme for Employment Employment, ed

and Social Innovation (EaSI) ucation, training
1.1.81 Energy Infrastructure
1.1.82 Transport Infrastructure
1.1.83 Information and Communications Sectoral policy

Technology (ICT)
1.1.9 Energy projects to aid economic Infrastructure

recovery (EERP)
1.2.5 Youth Employment Initiative (spe Employment, ed

cific top-up allocation) ucation, training
1.2.6 Contribution to the Connecting Eu Infrastructure

rope Facility (CEF)
Page 37 of 50
2.0.4 Programme for the Environment Ecological

and Climate Action (Life)
1.2.11 Less-developed regions (Regional Regional—for the

convergence) thematic break
down see Table
A22.3
1.2.12 Transition regions Regional—for the

thematic break
down see Table
A22.3
1.2.13 More-developed regions (Competi Regional—for the

tiveness) thematic break
down see Table
A22.3
1.2.14 Outermost and sparsely populated Regional—for the

regions thematic break
down see Table
A22.3
1.2.15 Cohesion fund (incl. contribution Regional—for the

to the CEF) thematic break
down see Table
A22.3
Source: European Commission at http://ec.europa.eu/budget/graphs/

revenue_expediture.html.
22.6.2.2 Mapping of thematic objectives in the ESIF to the industrial

(p. 656)
policy themes
For the budget items representing expenditure of the European Structural and Invest
ment Funds (ESIF) additional information on the ‘thematic objectives’ of the funds were
used. However, for these targets, only the figures from the projected expenditure as allo
cated in the MFF 2014‒20 were available. This information was exploited to assign the
regional policy expenditures to individual targets/objectives. The implicit assumption here
was that the actual disbursements (both from the EU budget and national co-financing)
have the same distribution across objectives as the projected amounts in the MFF 2014‒
20. Since the classification of the MFF 2014‒20 does not allow distinguishing between
expenditure under the European Social Funds (ESF) and the European Funds for Region
al Development (EFRD), these two funds are treated together, distinguished, however, by
Page 38 of 50
objectives. The data on regional fund allocations is found at: https://ec.europa.eu/

regional_policy/EN/funding/available-budget/.
Spending under the Cohesion Fund, which was not found in the budget allocations, is as
signed to infrastructure as it must be assumed that this spending is made in respect of
the Connecting Europe Facility (CEF) which is also included in this budget item.
Page 39 of 50
Table A22.3 Categorization of the thematic objectives in the ESIF into

industrial policies themes
Code Target objective of struc Assigned industrial

tural funds policy category
1 Research and innovation RDI
2 Information and communica Sectoral policy

tion technologies
3 Competitiveness of SMEs SME
4 Low-carbon economy Ecological
5 Climate change adaptation Ecological

and risk prevention
6 Environment protection and Ecological

resource efficiency
7 Network infrastructures in Infrastructure

transport and energy
8 Sustainable and quality em Employment, education,

ployment training
9 Social inclusion Social
10 Educational and vocational Employment, education,

training training
11 Efficient public administration Institutions
12 Outermost and sparsely popu Social

lated
TA Technical assistance Institutions
DM Discontinued measures Not assigned
Source: European Commission at https://ec.europa.eu/regional_policy/

EN/funding/available-budget/.
Page 40 of 50
22.6.2.3 Assumption made for the co-financing by member states of ESIF

projects
For national co-financing no actual numbers are available. Therefore the ratio of budget
allocation between EU financing and national co-financing was assumed to prevail also
for the actual financing.
References
Aghion, Philippe, Jing Cai, Mathias Dewatripont, Luosha Du, Ann Harrison, and Patrick
Legros (2015) ‘Industrial Policy and Competition’, American Economic Journal: Macroeco
nomics 7(4): 1–32.
Baldwin, Richard and Lopez‐Gonzalez, Javier (2015) ‘Supply‐chain Trade: A Portrait of

Global Patterns and Several Testable Hypotheses’, The World Economy 38(11): 1682‒721.
Becker, Sasha, Peter Egger, and Maximilian von Ehrlich (2012) ‘Too Much of a Good
Thing? On the Growth Effects of the EU’s Regional Policy’, European Economic Review
56(4): 648‒68.
Bianchi, Patrizio and Sandrine Labory (2006) ‘Empirical Evidence on Industrial Policy us
ing State Aid Data’, International Review of Applied Economics 20(5): 603–21.
Capello, Roberta and Henning Kroll (2016) ‘From Theory to Practice in Smart Specialisa
tion Strategy: Emerging Limits and Possible Future Trajectories’, European Planning
Studies 24(8): 1393‒406.
Celi, Giuseppe, Andrea Ginzburg, Dario Guarascio, and Annamaria Simonazzi (2018) Cri
sis in the European Monetary Union: A Core-Periphery Perspective. New York: Routledge.
Ciani, Andrea and Karsten Mau (2019) ‘European Integration, Trade, and Globalization:
Eastern Europe’s Response to Chinese Competition’. Paper presented at the International
Conference Economics of Global Interactions: New Perspectives on Trade, Factor Mobili
ty and Development, Bari, 10–11 September.
velopment: The Political Economy of Capabilities Accumulation. Oxford: Oxford Universi
ty Press.
Cohen, Elie (2007) ‘Industrial Policies in France: The Old and the New’, Journal of Indus
try, Competition and Trade 7(3): 213–27.
Doleys, Thomas (2013) ‘Managing the Dilemma of Discretion: The European Commission
and the Development of EU State Aid Policy’, Journal of Industry, Competition and Trade,
13(1): 23–38.
European Commission (2006) ‘Global Europe—Competing in the World: A Contribution to

the EU’s Growth and Jobs Strategy’. COM(2006)0567 final. Brussels: European Commis
sion.
Page 41 of 50
European Commission (2009) ‘Preparing for our Future: Developing a Common Strategy
for Key Enabling Technologies in the EU’. COM(2009)512 final. Brussels: European Com
mission.
European Commission (2010) ‘An Integrated Industrial Policy for the Globalisation Era.
Putting Competitiveness and Sustainability at Centre Stage’. COM(2010)614. Brussels:
European Commission.
European Commission (2011a) ‘Industrial Policy: Reinforcing Competitiveness’.

COM(2011) 642 final. Brussels: European Commission.
European Commission (2011b) ‘State Aid Scoreboard: Report on State Aid Granted by the
EU Member States, Autumn 2011 Update’. COM(2011)848 final. Brussels: European
Commission.
European Commission (2012) ‘A Stronger European Industry for Growth and Economic
Recovery. Industrial Policy Communication Update’. COM(2012)582/3. Brussels: Euro
pean Commission.
European Commission (2014) ‘For a European Industrial Renaissance’. COM(2014)014.

Brussels: European Commission.
European Commission (2017a) ‘Investing in a Smart, Innovative and Sustainable

(p. 658)
Industry: A Renewed EU Industrial Policy Strategy’. COM(2017)479 final. Brussels: Euro

pean Commission.
European Commission (2017b) ‘Towards a Mission-oriented Research and Innovation Poli

cy in the European Union, ESIR Memorandum: Executive Summary’. Brussels: European
Commission. Available at https://ec.europa.eu/info/sites/info/files/
an_esir_memorandum-towards_a_mission-oriented_research-and-
innovation_policy_in_the_european_union-executive_summary.pdf.
European Commission (2017c) ‘Delivering TEN-T: Facts and Figures’, September 2017.
Available at http://www.connectingeu.eu/documents/Delivering_TEN_T.pdf.
European Commission (2019) ‘EU Industrial Policy’. Brochure, March 2019. Available at
https://ec.europa.eu/commission/sites/beta-political/files/
brochure_industrial_policy_euco13032019_0.pdf.
European Commission (2020) ‘A New Industrial Strategy for Europe’. Communication by

the Commission, COM(2020)102 final. Brussels: European Commission.
European Political Strategy Centre (EPSC) (2019) ‘EU Industrial Policy after Siemens-Al
stom: Finding a New Balance between Openness and Protection’. Brussels. Available at
https://ec.europa.eu/epsc/sites/epsc/files/epsc_industrial-policy.pdf.
Foray, Dominique (2013) ‘The Economic Fundamentals of Smart Specialisation’, Ekonomi

az 83(2), 54‒78.
Page 42 of 50
Foray, Dominique, Paul David, and Bronwyn Hall (2009) ‘Smart Specialisation: The Con
cept’. Knowledge Economists Policy Brief No. 9, June.


Journal of Development Economics 72(2): 603‒33.
Hidalgo, Cesar, Baily Klinger, Albert-Laszlo Barabási, and Ricardo Hausmann (2007) ‘The
Product Space Conditions the Development of Nations’, Science 317(5837): 482‒7.
University Press.
Iammarino, Simona, Andrés Rodriguez-Pose, and Michael Storper (2019) ‘Regional

Inequality in Europe: Evidence, Theory and Policy Implications’, Journal of Economic Ge
ography 19: 273‒98.
Landesmann, Michael A. (2015) ‘The New North–South Divide in Europe: Can the Euro
pean Convergence Model Be Resuscitated?’, in Jan Fagerberg, Staffan Laestadius, and
Ben R. Martin (eds) The Triple Challenge for Europe: The Economy, Climate Change and
Governance. Oxford: Oxford University Press, pp. 60–88.
Landesmann, Michael A. (2019) ‘External Imbalances and European Integration’, in

Christina Marcuzzo, Antonella Palumbo, and Paola Villa (eds) Economic Policy, Crisis and
Innovation: Beyond Austerity in Europe. Milton: Routledge, pp. 84–112.
Landesmann, Michael A. and Roman Stöllinger (2019) ‘Structural Change, Trade and
Global Production Networks: An “Appropriate Industrial Policy” for Peripheral and Catch
ing-up Economies’, Structural Change and Economic Dynamics 48: 7‒23.
Martin, Ron (2008) ‘National Growth versus Spatial Equality? A Cautionary Note on the
New “Trade-off” Thinking in Regional Policy Discourse’, Regional Science, Policy and
Practice 1: 3‒13.
tor Myths. London: Anthem Press. (p. 659)
Mazzucato, Mariana (2018) Mission-Oriented Research and Innovation in the European

Union: A Problem-solving Approach to Fuel Innovation-led Growth. Luxembourg: Publica
tions Office of the European Union. Available at https://ec.europa.eu/info/sites/info/
files/mazzucato_report_2018.pdf.
Owen, Geoffrey (2012) ‘Industrial Policy in Europe since the Second World War: What
Has Been Learnt?’, ECIPE Occasional Paper No. 1, The European Centre for International
Page 43 of 50
Political Economy, Brussels. Available at http://eprints.lse.ac.uk/41902/1/

Industrial_policy_in_Europe_since_the__Second_World_War_what_has_been_learnt%28lsero%
Pellegrin, Julie, Louis Colnot, Francesco Prota, Francesca Ardizzon, Laura Delponte,
Michele Capriati, Agnieszka Olechnicka, and Chris Smith (2019) ‘How to Tackle Chal
lenges in a Future-oriented EU Industrial Strategy? Volume 1’. Study commissioned by
ITRI, June. Available at http://www.europarl.europa.eu/RegData/etudes/STUD/
2019/638409/IPOL_STU(2019)638409_EN.pdf.
Pellegrin, Julie, Maria Giorgetti, Camilla Jensen, and Alberto Bolognini (2015) ‘EU Indus
trial Policy: Assessment of Recent Developments and Recommendations for Future Poli
cies’. Study commissioned by the Directorate-General for Internal Policies, Policy Depart
ment A. Economic and Scientific Policy, February. Available at http://
www.securepart.eu/download/ipol_stu-2015-536320_en150416142150.pdf.
Radosevic, Slavo (2017) ‘Assessing EU Smart Specialisation Policy in a Comparative Per

spective’, in Slavo Radosevic, Adrian Curaj, Radu Gheorghiu, Liviu Andreescu, and Imo
gen Wade (eds) Advances in the Theory and Practice of Smart Specialisation. London:
Academic Press, pp. 4–40.
Rodriguez-Pose, Andres (2018) ‘The Revenge of the Places That Don’t Matter (and What
to Do about It)’, Cambridge Journal of Regions, Economy and Society 11: 189‒209.
Rodrik, Dani (2009) ‘Industrial Policy: Don’t Ask Why, Ask How’, Middle East Develop
ment Journal 1(1): 1–29.
469–91.
Rosenstein-Rodan, Paul (1943) ‘Problems of Industrialisation of Eastern and South-East

ern Europe’, The Economic Journal 53: 202‒11.
Schwengler, Barbara (2013) ‘Einfluss der europäischen Regionalpolitik auf die deutsche
Regionalförderung’. IAB Discussion Paper No. 18/2013. Available at http://doku.iab.de/
discussionpapers/2013/dp1813.pdf.
Stehrer, Robert and Roman Stöllinger (2015) ‘The Central European Manufacturing Core:
What is Driving Regional Production Sharing?’. FIW-Research Report No. 2014/15(2).
Stöllinger, Roman (2016) ‘Structural Change and Global Value Chains in the EU’, Empiri
ca 43: 801‒29.
Stöllinger, Roman (2019) ‘Testing the Smile Curve: Functional Specialisation in GVCs and
Value Creation’. wiiw Working Paper No. 163, February.
Timmer, Marcel, Sebastien Miroudot, and Gaaitzen J. de Vries (2019) ‘Functional Speciali
sation in Trade’, Journal of Economic Geography 19(1): 1‒30.
Page 44 of 50
Trippl, Michaela, Elena Zukauskaite, and Adrian Healy (2019) ‘Shaping Smart Specialisa
tion: The Role of Place-specific Factors in Advanced, Intermediate and Less-Developed
European Regions’. Online. Regional Studies 53(9).
UNCTAD (2018) World Investment Report 2018. Investment and New Industrial Policies.
Geneva: UNCTAD. (p. 660)
Veugelers, Reinhilde and Michael Baltensperger (2019) ‘Europe: The Global Centre for
Excellent Research’. Study requested by the ITRE Committee of the European Parlia
ment, Brussels.
Veugelers, Reinhilde, Gustav Fredriksson, Alexander Roth, Simone Tagliapietra (2018) ‘Is
the European Automotive Industry Ready for the Global Electric Vehicle Revolution?’. Pol
icy Contribution No. 26. Brussels: Breugel.
Wade, Robert H. (2012) ‘Return of Industrial Policy?’, International Review of Applied

Economics 26(2): 223‒39.
World Economic Forum (2018) Readiness for the Future of Production Report 2018. Gene
va: WEF.
Wyns, Tomas (2017) ‘A Mapping of the EU Industrial and Innovation Policy’. VUB IES
Working Paper, November. Available at http://i2-4c.eu/wp-content/uploads/2017/10/
Industrial-policy-mapping_i24c_T.Wyns_2017.pdf.
Notes:
(1) For an overview see, for example, Radosevic (2017).
(2) This is because any form of industrial policy, even if designed as a horizontal measure
such as R&D support or even a slightly undervalued exchange rate, will affect sectors dif
ferently or, as Rodrik puts it: ‘In practice most interventions, even those that are meant to
be horizontal, necessarily favor some activities over others’ (Rodrik, 2009: 6).
(3) The motivation behind the deal was to create a European player in the rail sector
which has sufficient scale to compete with the rapidly expanding (state-owned) Chinese
rivals. The European Commission’s decision has intensified the perception that the tough
European competition rules in the areas of merger control, state aid, or also public pro
curement put European firms at a disadvantage vis-à-vis foreign competitors who do not
always play according to the same rules.
(4) ‘State aid’ is EU jargon for subsidies provided by national governments.
(5) To policy formulation and implementation, one should also add evaluation.
(6) This is true at least for active industrial policy measures. In contrast, framework poli
cies might be budget-neutral.
Page 45 of 50
(7) Bianchi and Labory (2006) use a slightly different temporal delimitation of the liberal
and the pragmatic phase.
(8) The founding members of the EEC (as well as the European Community for ECSC)
were Belgium, France, Germany, Italy, Luxembourg, and the Netherlands.
(9) The United Kingdom joined the EEC in 1973 (together with Ireland and Denmark).
(10) Approaches to industrial policy, while highly selective and comprehensively employed
during this era, were by no means uniform—neither throughout the three decades
stretching from the 1950s to the 1980s, nor across member states. The German economic
model differed in so far as it embraced ordoliberalism as its guiding economic principle.
The specificity of the ordoliberal concept, often considered as the key to the German ‘eco
nomic miracle’ of the post-war period, was its strong emphasis on competition and mar
ket forces. Still, the state had a much greater role to play than in the Anglo-Saxon liberal
conception in so far as it had to provide vital infrastructure, an inclusive training and edu
cational system, and to ensure regional cohesion.
(11) A prime example dating from the 1970s was the construction of a super-fast train, the
Train à Grande Vitesse (TGV), developed by Alstom.
(12) The European Common Market came into being with the creation of the EEC in 1958.
(13) According to these rules, governments were not allowed to grant aid ‘which distorts
or threatens to distort competition…in so far as it affects trade by member states’. See
Article 107 of the Treaty on the Functioning of the European Union (TFEU).
(14) The EU competition rules form part of the single market programme but given its
great importance and its distinct features and consequences, they are considered as a
separate component.
(15) The first collaboration programmes included the European Strategic Programme on
Research in Information Technology (ESPRIT) and the European Research Coordination
Programme (EUREKA).
(16) Article 173(1) of the TFEU dealing with industrial policy stipulates that the EU ‘shall
ensure that the conditions necessary for the competitiveness of the Community’s industry
exist’.
(17) The Lisbon Agenda aimed at making the EU ‘the most competitive and dynamic
knowledge-based economy’.
(18) Article 173(1) makes industrial policy a ‘shared competence’, which means that mem
ber states remain the main actors but with complementary activities and regulatory pow
er at the EU level.
Page 46 of 50
(19) The ERDF was the first of the European Structural and Investment Funds (ESIF).
Currently, the ESIF family comprises five distinct funds: the ERDF, the European Social
Fund (ESF), the Cohesion Fund (CF), the European Agricultural Fund for Rural Develop
ment (EAFRD), and the European Maritime & Fisheries Fund (EMFF).
(20) Greece entered the EC in 1981, followed by Portugal and Spain in 1986.
(21) Apparently, France and Germany pressed the European Commission to come up with
a new Industrial Policy Communication which should facilitate the creation of European
champions, ideally by amending EU competition law (EPSC, 2019). The new Industrial
Policy Communication (European Commission, 2020) was finally issued in March 2020
and only marginally reflects the Franco-German demands and focuses instead on techno
logical leaders in digital markets and a transformation to a ‘green’ economy.
(22) See: https://ec.europa.eu/programmes/horizon2020/en/h2020-sections.
(23) The term ‘Industry 4.0’ originates from the German High-Tech Strategy and has since
become a general term for either preparing for, and/or successfully embracing, the new
production technologies associated with the Fourth Industrial Revolution (notably cyber-
physical production systems). Industry 4.0 constituted the German government’s Project
for the Future, initiated in 2011 and developed into a platform in 2013. See: https://
www.bmbf.de/de/zukunftsprojekt-industrie-4-0-848.html.
(24) See An Integrated Industrial Policy for the Globalisation Era (European Commission,
2010).
(25) See Investing in a Smart, Innovative and Sustainable Industry (European Commis
sion, 2017a).
(26) The prime example of a mission from the past is sending a man to (and returning him
from) the moon.
(27) See: https://www.ucl.ac.uk/bartlett/public-purpose/partnerships/ucl-commission-mis

sion-oriented-innovation-and-industrial-strategy-moiis#UK%20IS’s%20GC. For this pur
pose the European Commission also engaged Mariana Mazzucato as an adviser and com
missioned a report on mission-oriented innovation and industrial strategy (Mazzucato,
2018).
(28) A summary of the reform is provided on the website of the European Commission at:
https://ec.europa.eu/growth/single-market/public-procurement/rules-implementation_en.
(29) For details see Appendix A22.6.2.
(30) Note that this approach captures only policy actions that have budgetary implications
and, therefore, other potentially important measures such as regulations or the conclu
sion of a free-trade agreement are left out.
Page 47 of 50
(31) Note that this share of industrial-policy-related spending is lower than that in Pelle
grin et al. (2015) who subsume 87 per cent of the EU MFF 2014‒20 to ‘Relevant pro
grammes from an industrial policy perspective’ (see table 2, pp. 31‒2). This would result
in EU industrial-policy spending of about 0.8 per cent of EU GDP. The main difference be
tween the numbers in this chapter and those in Pellegrin et al. (2015) is that these au
thors include all spending on the Common Agricultural Policy (CAP) and the Maritime and
Fisheries Policies (amounting together to about 40 per cent of the EU budget) in their
analysis.
(32) While not surprising, this is still a remarkable result given the rather generous alloca
tion of funds from the EU budget to the realm of industrial policy in the calculations,
which includes public infrastructure expenditure and parts of active labour-market poli
cies.
(33) See also Wade (2012) describing US industrial policy.
(34) Note that this quantitative exercise does not include all spending under the ESIF but
only that of the European Social Funds (ESF) and the European Funds for Regional De
velopment (EFRD) as well as the Cohesion Fund (CF).
(35) EU institutions are monitoring the projects and the disbursement of funds though,
which is also an important function.
(36) This result is influenced by the fact that we defined industrial-policy spending by
member states rather narrowly, limiting it to state aid which therefore excludes most of
the investment expenditure on public infrastructure.
(37) The disaggregation of ESIF funds into categories of industrial policies is possible be
cause ESIF funds are structured into ‘thematic objectives’.
(38) See: https://ec.europa.eu/transport/themes/infrastructure_en.
(39) The European Fund for Strategic Investments (EFSI) is the central pillar of the In
vestment Plan for Europe. See EU regulation 2015/1017 on the European Fund for
Strategic Investments, the European Investment Advisory Hub and the European Invest
ment Project Portal.
(40) The initiative was initially scheduled for three years (2015‒17).
(41) See: ‘EIB and Commission Defend Juncker Plan Following Auditors’ Criticism’, Eu
roActiv, 30 January 2019. https://www.euractiv.com/section/economy-jobs/news/eib-and-
commission-defend-juncker-plan-following-auditors-criticism/.
(42) For a distribution of the projects financed by the EFSI by country and topic, see:
https://ec.europa.eu/commission/priorities/jobs-growth-and-investment/investment-plan-
europe-juncker-plan/investment-plan-results_en.
Page 48 of 50
(43) Certainly, the priorities of the central EU budget are also ultimately decided by mem
ber states.
(44) For the groupings of member states see Appendix 22.6.1.
(45) See ERC website at: https://erc.europa.eu/about-erc/mission.
(46) The key enabling technologies (KETs) include, but are not limited to: advanced mate
rials, nanotechnology, micro- and nano-electronics, biotechnology, and photonics (Euro
pean Commission, 2009). The KETs are regarded as the basis for innovation in a range of
products across all industrial sectors. In particular, they should support the moderniza
tion of Europe’s industrial base and the shift to a greener economy.
(47) These numbers refer to the evaluation period 2014‒16.
(48) See https://ec.europa.eu/info/horizon-europe-next-research-and-innovation-frame

work-programme_en.
(49) See also the discussion in section 22.4.1.
(50) The same is true for thematic objective 2, information and communication technolo
gies (ICT). In the quantitative exercise, these two objectives are subsumed under the
themes ‘RDI and technology’ and ‘sector-specific industrial policies’, respectively. Note
that the S3 conditionality also applies to TO1 of the EAFRD (Fostering knowledge trans
fer and innovation in agriculture, forestry, and rural areas). However, we have excluded
the EAFRD from the analysis in this chapter.
(51) Horizon Europe is the R&D and innovation framework programme of the MFF 2021‒
28, which is to replace the current Horizon 2020 programme.
(52) These clusters are: Health; Culture, Creativity, and Inclusive Society; Civil Security
for Society; Digital, Industry, and Space; Climate, Energy, and Mobility; and Food, Bioe
conomy, Natural Resources, Agriculture, and Environment.
(53) See EPSC (2019). The German software company SAP may be an exception here.
(54) This is, for example, revealed by the Digital Economy and Society Index (DESI) avail
able at https://ec.europa.eu/digital-single-market/en/desi.
(55) Arguably, the CEE member states are also new actors in the global economy but
when referring to emerging economies we think of extra-EU partner countries.
(56) Functional specialization refers to the specializing in specific segments of the value
chain within an industry-specific value chain.
(57) The same is true for other tightly integrated regional blocs such as (former) NAFTA
and Southeast Asia.
Page 49 of 50
(58) Translating ‘green’ industrial policies into missions therefore seems the logical next
step. If industrial policies aim at altering the economic structure, then the shift towards a
sustainable, less resource-intensive and circular economy based on renewable energies is
the obvious policy target given the pressing challenge of climate change but also the poli
cy priorities revealed by member states. To this one should add that the prevalence of
multiple market failures in the context of a greening of the economy, which include mas
sive externalities, path dependencies favouring pollution-intensive technologies, and the
public good characteristic of a clean environment, makes striving for an ecological trans
formation the number one objective for a credible industrial policy at the European level.
Michael A. Landesmann
Michael A. Landesmann was scientific director of the Vienna Institute for Internation
al Economic Studies (wiiw) from 1996 to 2016 and is professor of economics at Jo
hannes Kepler University, where he is head of the Department of Economic Theory
and Quantitative Research. His research focuses on international economic relations,
European economic integration, globalization and labour markets, and migration. He
has a DPhil from the University of Oxford, was a lecturer and fellow (Jesus College)
at the University of Cambridge, and has held visiting professorships at Harvard Uni
versity (Pierre Keller, Schumpeter) and a range of other universities (Brandeis,
Jerusalem, Central European University, Padova, Basel, Osaka, Mumbai).
Roman Stöllinger
Roman Stöllinger is a staff economist at the Vienna Institute for International Eco
nomic Studies (wiiw). In his research he focuses on issues related to international
trade and global value chains, industrial policy and structural change. Recent publi
cations include a paper on the impact of global value chains on structural change and
the need for appropriate industrial policies in Europe and its neighbourhood. He is a
regular contributor to research reports by international institutions such as the Euro
pean Commission and UNIDO. He is strongly involved in the Research Centre Inter
national Economics (FIW), an Austrian think-tank and infrastructure platform in the
field of international economics. Since 2012 he has also been a lecturer at the Vienna
University of Economics and Business, teaching international macroeconomics and
industrial policy. Before joining wiiw in 2008, he worked at OeKB, the Austrian Ex
port Credit Agency, where he gained real-world experience in implementing state
support policies. He holds a master’s degree in international economics from the Uni
versity of Innsbruck and a PhD in economics from the University of Vienna.
Page 50 of 50
Diverse Tools of Industrial Policy in Korea: A Schumpeterian and Capabili
ty-based View
Diverse Tools of Industrial Policy in Korea: A Schum

peterian and Capability-based View
Keun Lee
The chapter presents a Schumpeterian and capability-based view of industrial policy, re
flecting upon its practices in Korea over the last several decades. Given that it is typical
for many developing countries to suffer from capability failure, industrial policy should go
beyond correcting market failure and aim at overcoming capability failure. It is not about
picking winners but about picking good students and allowing them time to learn and
build capabilities until they are able to compete with incumbent firms from developed
countries. This chapter discusses specific industrial policy tools practised in Korea at dif
ferent stages of its development: tariffs to protect infant industry; technology import li
censing to promote building of absorptive capacity; entry control guaranteeing rents for
fixed and R&D investment; and public‒private joint R&D to break into higher-end prod
ucts and sectors. While these tools look different in their concrete contents, they all allow
some rents for the targeted sectors, which can be used to pay for building production ca
pabilities in the case of tariffs or technology licensing in the 1970s, investment capabili
ties in entry control in the 1980s, and technological (R&D) capabilities in the case of pub
lic‒private joint R&D in the 1990s.
Keywords: industrial policy, tariffs, licensing, public–private joint R&D, entry control, infant industry, leapfrogging
23.1 Introduction
A notable result of the 2008 global financial crisis has been the revival of industrial poli
cy, initiated by the work of Stiglitz and Lin (2013). While classical work, such as Johnson
(1982), defined industrial policy as policies that improve the structure of a domestic in
dustry in order to enhance a country’s international competitiveness, the meaning has
evolved to meet the context of the twenty-first century (Radosevic et al., 2017). This chap
ter presents a Schumpeterian and capability-based view of industrial policy (Lee and
Malerba, 2018; Lee, 2013b), reflecting upon its practices in Korea over the last several
decades.
Page 1 of 19
ty-based View
It is critical for a developing country to enhance its capability to produce and sell prod
ucts in the international market so that the country may earn foreign currency that it can
then use to pay for imports of investment goods. However, the challenging part of this
process is how to increase that capability. Thus, it is typical for many developing coun
tries to suffer from capability failure (Lee, 2013c; Lee and Malerba, 2018). Further, indus
trial policy should go beyond correcting market failure to aim at overcoming capability
failure (Lee, 2013c, 2019). The market-failure perspective focuses on providing optimal
incentives to correct externalities associated with public goods like R&D, with a hidden
assumption that firms are already equipped with capabilities. However, in the absence of
capabilities, incentives alone are not sufficient to start innovation. It is not about picking
winners, but more about picking good students and (p. 662) allowing them time to learn
and build capabilities until they are able to compete with incumbent firms from devel
oped countries (Lee and Malerba, 2018).
Capability building was the focus of a World Bank study compiled by Chandra (2006). A
World Bank (2005) assessment of the reform decade of the 1990s concluded that growth
entails more than the efficient use of resources, and that growth-oriented action may be
needed, for example, for technological catch-up or the encouragement of risk-taking for
faster accumulation. Lee and Mathews (2010) synthesize the capability-based view as the
Beijing‒Seoul‒Tokyo (BeST) consensus, which is commensurate with their firm-level
study (Lee and Mathews, 2012).
Capability building, not the state‒market dichotomy, is the essence of the Korean model
of catch-up. This chapter argues that because Korea built and enhanced the capabilities
of private firms, it was able to sustain growth for several decades until it joined the club
of high-income economies (Lee, 2013b). If we consider industrial development as a long-
term process that takes over ten or twenty years, it is natural for the policy tools to
change over the course of economic growth. Such a dynamic view of industrial policy is
warranted, because the capability level of the beneficiaries of such intervention would
change over time as well.
This chapter discusses specific industrial policy tools practised in Korea at different
stages of the country’s development: tariffs to protect infant industry (Shin and Lee,
2012); licensing of technology imports to promote building of absorptive capacity (Chung
and Lee, 2015); entry control to guarantee rents to be paid for fixed and R&D investment
(Jung and Lee, 2010); and public‒private joint R&D to break into higher-end products and
sectors (Lee et al., 2005; Lee, 2013a). While these tools differ in their concrete contents,
a common aspect is that they allow some rents (extra profits) for targeted sectors which
can be used to pay for building variants of capabilities, such as production capabilities in
the case of tariffs or technology licensing in the 1970s, investment capabilities in entry
control in the 1980s, and technological (R&D) capabilities in the case of public‒private
joint R&D in the 1990s.
Sections 23.2, 23.3, and 23.4 discuss technology licensing, tariffs and entry control, and
public‒private joint R&D, and are followed by some conclusions.
Page 2 of 19
ty-based View
23.2 Technology Licensing for Absorptive Ca

pacity
Absorptive capacity (AC) was first introduced by Cohen and Levinthal (1989, 1990) as the
ability of a firm to identify, value, assimilate, and exploit knowledge from the environ
ment. AC is also recognized as an important binding constraint in the development of
latecomer economies. Borensztein, Gregorio and Lee (1998) performed a country panel
regression and found that foreign direct investment will produce a growth effect only if a
country has a certain level of AC. Although several empirical (p. 663) studies emphasize
the importance of absorptive capacity using in-house R&D or human capital as proxy
(Keller, 1996; Mowery and Oxley, 1995), recent findings indicate that AC cannot be appro
priately proxied by R&D or staff quality alone (Flatten et al., 2011; Lane et al., 2006). The
earlier studies fail to discover other origins of AC aside from formal R&D or education.
Relying heavily on Chung and Lee (2015), this section examines the origin of AC in Korea.
How can we tell whether this capacity is established in a firm? This question is particular
ly relevant in the context of latecomer countries in which firms are usually uncertain
about conducting their own R&D and continue to rely on imported technology by special
izing in assembly-type production. Scholars studying Korea as an example of a successful
latecomer economy have emphasized the importance of AC in enabling Korean firms to
learn and assimilate external knowledge (Evenson and Westphal, 1995; Keller, 1996; Pack,
1992).
As a latecomer economy, Korea has successfully transformed itself from a technology-im

porting to a technology-generating country. Korean firms only began to conduct in-house
R&D in the mid-1980s after undergoing a period of learning, assimilating, and adapting
foreign technologies (OECD, 1996; Lee, 2013b). Chung and Lee (2015) observed that the
number of foreign technologies acquired by Korean firms increased from as early as the
late 1960s, followed by an increasing trend towards private R&D two decades later. In
other words, a significant increase in foreign technology inflow preceded local R&D ef
forts and innovation outcomes in Korea. Many researchers assert that access to external
knowledge is especially important in the development of latecomer firms (Bell and Pavitt,
1993; Kim, 1997; Laursen and Meliciani, 2002; Jaymin Lee, 1996; Park and Lee, 2006).
Leading firms in Korea generally acquired various forms of know-how, such as opera
tional skills and elementary process technology, before starting their own capital invest
ment (Enos and Park, 1988; Kim, 1997, 1998). These firms built their basic technology
proficiency during the building and testing of their production facilities, enabling Korean
engineers to assume responsibility for daily operations as early as possible.
The Foreign Exchange Control Act of 1960 addressed two objectives relating to foreign
technology acquisition. The first was to ensure that foreign exchange, which became
scarce after the Korean War, would only be used for critical technologies. Second, the
government wanted to use acquired technology as a stepping stone on which Korean
firms could build their own technological capabilities (Korea Development Bank, 1991).
Page 3 of 19
ty-based View
The scarcity of foreign exchange during the 1960s compelled Korean firms to seek gov
ernment approval prior to signing a contract with a foreign counterpart if they wanted to
receive technical assistance for a year or more and if they paid their counterparts in for
eign currencies. All applications were scrutinized by the Ministry of Commerce and In
dustry (Korea Development Bank, 1991).
Korean industries attempted to build production and export capabilities in labour-inten

sive or light industries, such as textiles, wigs, rubber footwear, and stuffed toys, from the
1960s and into the 1970s. However, during the 1970s firms realized that these (p. 664) in
dustries had low profit margins and insufficient cash inflow to produce the necessary for
eign debt services. Therefore, both the government and the private sector wanted to inte
grate backwards into intermediate goods which, if secured within the country, would re
duce the need for foreign exchange in the long run. Under the Economic Development
Plan, a raft of legislation was enacted to promote the general machinery, electronics, oil
refinery and petrochemicals, transport equipment, steel and shipbuilding industries
(Byun and Park, 1989). The approval procedure for the acquisition of foreign technologies
in these target sectors was also simplified.
By the late 1970s, many of the initial entrants into the ‘heavy’ industries were acquiring
both physical capital and relevant technology from foreign sources. Westphal, Kim, and
Dahlman (1985: 190–1) reported that more than a quarter of gross domestic investment
in Korea from 1977 to 1979 was spent on foreign capital goods. In 1978, an automatic ap
proval system for the acquisition of foreign technology was introduced in general and
electrical machinery, shipbuilding, chemicals, textiles, and finance under the following
conditions: (1) the duration of the contract must be three years or shorter; (2) the down
payment must be US$30,000 or less; (3) the running royalty rate must be 3 per cent or
lower; and (4) the fixed fee must be US$100,000 or less. From 1979 onwards, most sec
tors, except weapons, explosives, and nuclear power, were granted automatic approval
for their projects subject to satisfying the conditions.
Deregulation continued in the 1980s and 1990s, with the filing requirement abolished in
1994 and the approval process simplified to a filing-and-confirmation process from 1984.
From 1988, foreign exchange banks were assigned to handle a certain level of foreign
technology acquisition applications (Korea Development Bank, 1991; Korea Industrial
Technology Association, 1995: 6).
The Korea Industrial Technology Association, a semi-government agency, published a data

book, KOITA (1995), covering foreign acquisition in the period 1970–93. The value of this
unique data set lies in the fact that all contracts are reported and classified into three cat
egories: know-how-only acquisition, know-how-and-patent-rights acquisition and patent-
rights-only acquisition. Know-how-only acquisition typically consists of technical services
and training that are bundled with relevant documents. Know-how-and-patent-rights
transfer consists of technical services, training, and documentation protected by the
patent system. Patent rights only consists of patent-right licensing.1 Chung and Lee
(2015) made a number of findings from analysis of this database.
Page 4 of 19
ty-based View
First, they found that contracts for know-how licensing dominated in the early years,
whereas contracts that involve patents followed later. The shares of know-how-only, know-
how-and-patent-rights, and patent-only contracts during the period 1970–5 were 69 per
cent, 27 per cent and 4 per cent, respectively. However, these shares amounted to 50 per
cent, 45 per cent and 5 per cent between 1976 and 1981, reflecting (p. 665) the subse
quent increase in know-how-and-patent-rights licensing. This pattern may imply that
those firms that successfully assimilated basic operational skills and elementary process
technology through know-how acquisition advanced to the acquisition of technologies
that involve patent rights at later stages. These contracts include not only printed infor
mation and blueprints, but also technical services and training information. Expatriate
engineers usually come to Korea to ensure that the initial operation of new facilities goes
according to plan. Selected Korean engineers are sometimes sent abroad for training,
which emphasizes the importance of human capital investment in building AC, as demon
strated by leading firms in Korea, such as Hyundai Motors (Kim, 1998) and POSCO (Po
hang Steel Co.) (Song, 2002).
Various types of training, particularly overseas and on-site training, were arranged or
provided by the firms’ foreign suppliers of facilities and equipment. For instance, after
Hyundai entered into an agreement with Ford to assemble compact cars on a semi-
knocked-down basis, Ford transferred ‘packaged’ technologies to Hyundai with an accom
panying set of explicit knowledge, such as blueprints, technical specifications, and pro
duction manuals. Ford also provided tacit knowledge to Hyundai, sending ten Ford engi
neers to Hyundai and training Hyundai engineers at Ford sites in procurement planning,
procurement coordination, production engineering, process engineering, production man
agement, welding, painting, after-service, and marketing (Kim, 1998). In the case of POS
CO, the company sent thirty-nine engineers to Japan in 1968–9 and 1,861 staff members
abroad from 1968 to 1983 during the construction of its first mill (Song, 2002: 128).2
Second, they found that the acquisition of foreign technologies is dominated by four capi
tal-intensive sectors—electricals and electronic equipment, chemicals, transport equip
ment, and general machinery—rather than by labour-intensive sectors. These findings re
flect the industrial policy of the Korean government and the efforts of firms to enter these
sectors from the early 1970s. The contracts in these four industries made up more than
70 per cent of the total contracts filed in the sampling period and throughout each sub-
period.3 This finding indicates that the state’s control over foreign technology acquisition
was critical for the structural transformation of Korean industries from labour intensive
to capital intensive, which has ultimately helped them achieve industrial upgrading.
Korea’s ‘pillar’ companies, such as Samsung Electronics, Hyundai Motors, and POSCO,
were all established around 1970.
The entry of Korean firms into these industries typically involved the manufacture of
products that were new to Korea yet common in the developed world. According to a sur
vey by the Korea Development Bank (1991) on foreign technology acquisition in the
1980s, 55 per cent of these acquisitions by Korean firms related to mature technologies
(p. 666) in developed countries, whereas 70 per cent related to the expansion of product
Page 5 of 19
ty-based View
mix. With Korean firms finding the knowledge embedded in manufacturing facilities insuf
ficient for their operations, they sought additional services and training (or know-how-
and-patent licensing) from firms in developed countries at an appropriate price.
From the suppliers’ point of view, concealing mature technologies was pointless because
providing know-how was a way of exporting large manufacturing facilities. The Japanese
government’s decision to move away from ‘pollution-prone’, ‘natural-resource-consuming’
heavy and chemical industries in 1971 provided a favourable environment for Korean
firms (Enos and Park, 1988). Selling unnecessary technologies proved profitable for Japan
(Enos and Park, 1988: 34), and was consistent with the ‘flying-geese’ pattern of economic
growth in East Asia, where Japanese companies serve as leaders for their follower firms
from Korea or Taiwan by transferring their technologies or relocating their factories
abroad (Akamatsu, 1962; Kojima, 1973). In this regard, as noted by several scholars, in
cluding Kiyota and Okazaki (2005), Korean industries followed a similar path to those of
Japan, with foreign technology acquisition in the form of licensing rather than FDI nurtur
ing their domestic absorptive capacity and improving their performance (Lee and Kim,
2010).
According to the Korea Development Bank (1991), technologies that were bundled with
patent rights were more expensive or had a higher value than those that were only bun
dled with know-how. This arrangement suggests that Korean firms may have demanded
something more than the mere operation of manufacturing facilities after stabilizing their
daily production. Patented technologies may have been adopted as a means of completing
the assimilation and improvement processes that were initiated by investment and know-
how acquisition.
Imported capital goods have been regarded as among the most important forms of tech
nology transfer in Korea (Lee and Kim, 2010; OECD, 1996). However, these goods become
ineffective without the transfer of technology, and especially of tacit knowledge, through
know-how-only contracts. Thanks to frequent on-site training by foreign expatriate engi
neers, Korean engineers quickly learnt to manage their daily operations efficiently. If the
knowledge were deemed insufficient, the turnkey contractor and/or other sources, includ
ing R&D specialty companies or equipment providers, were contacted for additional infor
mation. Enos and Park (1988) argue that even in the most successful cases, such as POS
CO and Hanyang Chemicals, time and effort were necessary for Korean firms to become
able to use foreign technologies effectively.
Finally, the econometric analysis in Chung and Lee (2015) shows that know-how licensing
associated with imported capital led Korean firms to build AC and then to start in-house
R&D, whereas patent-only licensing was not significantly related to being able to conduct
R&D. Therefore, a substitution effect may be observed between the introduction of for
eign patents and the initiation of own R&D activities at the early stages of development. A
similar econometric exercise for the second step shows that conducting in-house R&D
leads firms to generate innovations, in terms of either patent applications or increased
productivity, during the later stages of their development. (p. 667) Moreover, there is a
Page 6 of 19
ty-based View
significant correlation between know-how licensing experience and firms’ first-time
patent applications, whereas patent-only licensing helps stimulate the subsequent gener
ation of patents. The study also finds that firms generally spend at least three to four
years building their AC, from the first year of know-how licensing to the first year of
patent applications.
Chung and Lee’s study (2015) was the first to verify the dynamic link between the learn
ing of tacit knowledge and the formation of AC, as well as the first to measure the actual
length of time firms take to build this capacity. This finding verifies the two-step-based
differentiation of ‘potential’ and ‘realized’ AC that is proposed by Zahra and George
(2002), as well as the decomposition of AC into learning capacity and problem-solving ca
pacity by Kim (1998). In contrast to Jaymin Lee (1996), our view is that the substituting
or inducing effect of foreign technology acquisition on indigenous R&D depends on the
specific licensing method. Patent-only licensing exhibits a substituting relationship,
whereas know-how licensing tends to have an inducing effect on in-house R&D. More
over, a learning process that involves foreign technology, especially tacit knowledge in
the form of know-how, occurs before firms can conduct in-house R&D and innovation.
However, this specific learning process may not be the only way to build AC, given the im
portance of worker education, on-the-job training, and overseas training.
Lee (2013b), arguing that it may be difficult to derive generalizable lessons from the
state-versus-market view of Korean economic development, proposed instead a ‘capabili
ty-based’ view and elaborated various modes of capability building that may be useful in
other latecomer countries. Chung and Lee’s (2015) econometric study provides a micro
economic foundation for a macro-level view of economic development, from which several
generalizable implications may be derived.
First, the building of AC is a dynamic process that may become more effective when com
bined with access to foreign knowledge, particularly tacit knowledge (know-how). This
suggests, second, that know-how transfer should be an essential element in technology-li
censing contracts for a latecomer firm. Moreover, third, the potential substituting effect
of foreign patent licensing may interfere with formation of in-house AC, especially if it is
not linked to the start of in-house R&D activities or contracting for know-how transfer.
23.3 Infant Industry Protection by Tariffs and

Entry Control
One of the most traditional industrial policy tools is infant industry protection through
tariffs. However, empirical studies on the effectiveness of tariffs report conflicting results.
According to Beason and Weinstein (1996), tariff protection, preferential tax rates, and
subsidies did not affect the rate of capital accumulation or total factor productivity
(p. 668) (TFP) in Japan from 1955 to 1980. Lee (1996) found no impacts of tariffs on TFP
in Korean industries from 1963 to 1983, such that nominal tariff was negative and signifi
Page 7 of 19
ty-based View
cant to the growth rate of labour productivity and TFP at the sectoral level. Nevertheless,
several studies verify the positive contribution of industrial policy, in particular tariffs.
Shin and Lee (2012), using the same period and sectoral data as Lee (2006), find that tar
iff protection, especially when combined with export-market discipline, leads to the
growth of export share and RCA (revealed comparative advantages). They also argue that
in the early stages the goal of this type of industrial policy was not productivity—as in the
1970s—but output or growth in market share. Aghion et al. (2011) also find that subsidies
widely distributed among Chinese firms have had a positive impact on both TFP and the
innovation of new products in sectors with a high level of competition. Both of these re
cent studies identify competition or discipline as a common precondition for effective in
dustrial policy.
An example of success with tariffs is the case of Hyundai Motors, established in 1970.
Hyundai’s first own-brand car was the Pony, which had a 44 per cent market share in Ko
rea in 1976. However, it was protected by tariffs on imported cars, for example from
Japan, as high as 82 per cent. While its domestic market price was about US$4,500 dol
lars, it was exported to the US market for US$1,850. In other words, it was only this
‘dumping’ that enabled Hyundai to compete with other cars, and it was only possible ow
ing to extra profits associated with the oligopoly market situation based on tariffs; at that
time, Japanese or German cars in a similar segment were selling at US$2,300 in US mar
kets. In other words, domestic profit compensated for the loss of foreign markets, and it
was these guaranteed profits that helped Hyundai to survive and pay for fixed and R&D
investment for expansion.
Thus, it can be argued that if Korea had been opened up from the beginning without tar
iffs, the Korean economy would not have been so successful in promoting indigenous
firms and sustaining their catch-up in market share. A hidden assumption of trade liberal
ization is that local firms are sufficiently competitive to potentially compete against for
eign companies or imported goods. This assumption is not true in many cases. In such cir
cumstances, naïve trade liberalization may lead to monopoly by foreign goods or the de
struction of local industrial bases.
A smart or better opening strategy, as discussed in Shin and Lee (2012), is ‘asymmetric
opening’, in which latecomer economies liberalize the import of capital goods for the pro
duction of final or consumer goods, while protecting their own consumer goods industries
by charging high tariffs on imported goods. Korea implemented an asymmetric tariff poli
cy for consumer and capital goods, with extremely high tariffs for consumer goods (e.g.
around 70 per cent for household electrical appliances in the 1970s) which were promot
ed as export industries, but considerably lower tariffs for capital goods, such as machin
ery which it needed to import for its assembly industries, mostly in the consumer goods
sector.
Another form of industrial policy in Korea was entry control. Put simply, the idea is that
five profit-making firms is better than ten firms in the same sector with no profits.
(p. 669) This type of entry control has typically been practised in the past by Japan, and
Page 8 of 19
ty-based View
copied in Korea, where it has been regarded as an industrial policy copied from the
Japanese (Johnson, 1982). The practice has two meanings: to sort the ‘good’ from the
‘bad’ producers; and to allow stable profits for the selected producers to encourage them
to invest further in fixed capitals for business expansion.
This practice also has the effect of return rates that are higher than interest rates. This is
good for boosting private investment in manufacturing, which corresponds to low rates of
return with longer time horizons. Manufacturing sectors can earn ‘rents’ associated with
entry control, and the government’s industrial policy consisted in working out the opti
mum number of each sector, taking market size into account, so that the firms admitted
are guaranteed a minimum level of profits (rents) which can provide the following
period’s investment funds.
Of course, the protection of local firms through tariffs and entry control leads to an oli
gopolistic domestic market. However, a study by Jung and Lee (2010) demonstrates that
monopoly rents can be used to fund R&D investment because firms are exposed to the
discipline of world export markets, while their privileged protected status is not free but
rather is linked to their export performance. In other words, the combination of rent-gen
erating protection in the domestic market and discipline from world markets was one of
the most important aspects of industrial policy in Korea during its catching-up stage
(mid-1980s and 1990s). Jung and Lee’s (2010) study confirms that R&D financed in this
way led to Korean firms developing enhanced innovation capabilities, which enabled their
productivity catch-up with Japanese firms from 1985 to 2005.
In other words, in-house R&D eventually became more important than foreign technology
acquisition because (1) foreign firms became increasingly reluctant to provide core tech
nology to their potential competitors in Korea, (2) labour cost-based competitiveness
gradually disappeared, and (3) government support for private R&D increased (OECD,
1996: 91–2).
23.4 Public‒Private Joint R&D

The final stage of industrial policy consists of public‒private R&D consortia that can
serve as important vehicles for breaking into higher-end segments or sectors which de
mand both more capital and more risk.
An early example was the government-led R&D consortia in the telecommunications

equipment industry, specifically the accompanying local development of telephone switch
es. This led to the successful localization of telephone switches in the 1980s and 1990s in
several latecomer countries, including China, Korea, India, and Brazil (Lee, Mani, and
Mu, 2012). In the 1970s and 1980s most developing countries experienced serious tele
phone service bottlenecks. With neither domestic telecommunications equipment manu
facturing industry nor their own R&D programme, they imported expensive equipment
and related technologies, and local technicians merely (p. 670) installed foreign switching
systems into the country’s domestic telephone networks. With industrial and commercial
Page 9 of 19
ty-based View
bases developing rapidly—along with population growth—a number of countries decided
to build their own manufacturing capabilities.
First in Brazil in the 1970s, followed by Korea and India in the mid-1980s, and finally by
China towards the late 1980s, a state-led system of innovation in the telecommunications
equipment industry was developed with a government research institute at the core. The
research institute developed more or less ‘indigenous’ digital telephone switches that
were then licensed to public and private domestic enterprises. In all four countries, a
common pattern in the indigenous development of digital switches was the tripartite R&D
consortium between government research institutes in charge of R&D functions, state-
owned enterprises or the ministry in charge of financing and coordination, and private
companies in charge of manufacturing at the initial or later stages. However, subsequent
waves of industry privatization and market liberalization in Brazil and India contrasted
with consistent infant industry protection in Korea and China (Lee, Mani, and Mu, 2012).
At one extreme, the indigenous manufacturers of China and Korea took over from the im
porters and MNCs. The enhanced capabilities in wired telecommunication they had been
accumulating over the preceding decades led to the growth of indigenous capabilities in
wireless telecommunications. At the other extreme, Brazil and India have increasingly be
come net importers of telecoms equipment, and their industries are now dominated by af
filiates of the MNCs.
As noted by Mathews (2002), examples from Taiwan include calculator and laptop com
puter production. The calculator is an example of the acquisition of more fundamental de
sign capability (basic design platform) enabled with the help of a government entity such
as the Industrial Technology Research Institute (ITRI). Another example is the 1990‒1
public‒private R&D consortium to develop laptop computers (Mathews, 2002). This con
sortium developed a common mechanical architecture for a prototype that could easily
translate into a series of mass-produced standardized components. The consortium repre
sented an industry watershed, and even after several failed attempts, it succeeded in es
tablishing new ‘fast-follower’ industries in Taiwan.
While the telephone switch case localized somewhat mature technologies, the same pub
lic‒private joint R&D model can be used to leapfrog into emerging technologies or prod
ucts. A Korean example is digital TV development, the decisive final watershed that en
abled Korea to begin overtaking Japan in the TV business. Examples from China include
recent moves towards solar power and electric cars. In these areas, there are no products
for latecomers to imitate: advanced and latecomer countries enter the market at the same
time. If the former latecomers succeed first, there is a strong momentum for them sur
passing the middle-income group and joining the rich-country club. The public‒private
R&D consortium plays a more vital role in this leapfrogging endeavour, given that the risk
involved is huge and different. Furthermore, coordinated initiatives on exclusive stan
dards and incentives for early adopters would be important in reducing the risk faced by
the weak initial market.
Page 10 of 19
ty-based View
In these public‒private R&D consortia in mature stage, private firms take the lead over
the public labs in conducting R&D jointly whereas the opposite would work (p. 671) dur
ing the early stage of public-private joint R&D. Thus, in the final stage of the R&D consor
tium, the public research arm would be in charge of monitoring the technology trends as
well as providing information about technology standards and identifying suitable foreign
partners (e.g. Qualcomm for mobile phone development and Zenith for digital TV develop
ment) for collaborative development. The foreign company’s role will also change. In the
second stage, the foreign company is the direct teacher in the co-development contract,
but in the final stage, it becomes the supplier of source technology to be commercialized
by the latecomer firm or their consortium. The final stage features horizontal collabora
tion or an alliance based on complementary assets. Some Korean firms (e.g. Samsung)
have reached this stage, and are now engaged with Intel, Sony, Toshiba, and Microsoft in
diverse modes of alliances.
The probability of leapfrogging success may therefore be higher when a new techno-eco
nomic paradigm or a new generation of technologies begins to emerge. Pérez and Soete
(1988) and Freeman and Soete (1997) observe that some latecomers may be able to
leapfrog older versions of technology, bypass heavy investment in previous technology
systems, and grasp new technologies to take over the market from the incumbent firms or
countries. This leapfrogging strategy makes more sense at the time of a paradigm shift,
because every country or firm is a beginner in using the new techno-economic paradigm,
and the entry barriers tend to be low. Furthermore, the ‘winner’s trap’ may operate in the
sense that the incumbent tends to ignore new technologies and continue to use the exist
ing dominant technologies until it exhausts its investment in the existing facility. The con
cept of leapfrogging is consistent with the idea of technological discontinuity proposed by
Anderson and Tushman (1990) and Tushman and Anderson (1986), that competence-de
stroying discontinuity may lead to the emergence of new entrants.
Korea’s catch-up with Japan in the development of high-definition TVs (HDTVs) would not
have been successful if Korean electronics companies, such as Samsung and LG, had not
targeted the emerging digital technology-based products more aggressively than Japan
ese companies that opted to continue manufacturing the dominant analogue products.4
The Japanese firms developed analogue-based HDTV in the late 1980s and suggested that
Korean companies follow new technologies and products by learning from them—which
was the strategy they used to pursue in the 1970s and 1980s. Instead, Korean companies
tried a leapfrogging strategy of developing an alternative and emerging technology—pro
ducing digital technology-based HDTVs. They successfully formed the public‒private
R&D consortium that marked the beginning of Korean hegemony in the global display
market previously dominated by Japan. If it had not been for this risk-taking leapfrogging
strategy, Korean catch-up with Japan would have taken much longer or might have never
happened.
Leapfrogging is more likely to happen where there are sectors with more frequent
changes in technologies or generation changes in products. This is closely linked with
(p. 672) the length of the technology cycle, as these features indicate the speed with
Page 11 of 19
ty-based View
which technologies change or become obsolete over time, paving the way for the contin
ued emergence of new technologies. We can reason that it is advantageous for qualified
latecomers to target and specialize in short-cycle technology-based sectors. Although this
is risky, it is also logical because latecomers do not have to rely substantially on the exist
ing technologies dominated by the incumbents; moreover, there are always more growth
opportunities associated with constantly emerging technologies.
Finally, we should note the importance of carefully handling the risks involved in choos
ing a leapfrogging strategy. As Lee et al. (2005) explain, one of the biggest risks is choos
ing the right technology or standard in the ex post sense. In the competition for standard
setting and market creation, the role of the government is to facilitate the adoption of
specific standards, thereby influencing the formation of markets at the right time, espe
cially when the target involved is in the area of information or other emerging technology.
Isolated development without considering standards might cause the entire project to
fail. Collaboration and partnership with rivals or suppliers of complementary products is
essential for standard setting. Another key factor is determining who creates and reaches
the market first, given the fact that market size determines the success or failure of one
standard in relation to another.
The above case for successful public and private joint R&D for leapfrogging in Korea can
be contrasted with a mixed case in South Africa, which was also discussed in Lee (2017).
Swart (2015) explains that the South African government provided the initial funding to
establish Optimal Energy in 2005, which by December 2010 had developed four roadwor
thy electric car prototypes. The Joule Electric Vehicle was a ‘born electric’ five‐seater pas
senger car incorporating a locally developed battery, motor, and software technologies.
However, the government decided to stop the funding required to start large-scale pro
duction because of uncertainties over marketing success, and the company closed in June
2012. The failure of Joule cars was caused by the lack of private companies to take on vol
ume production and sales. Existing foreign multinational companies and local auto com
panies did not want this new ‘disruptive innovator’, a state-owned company, to grow into
another rival.
A better outcome would have been for the government to form a public‒private consor
tium so that volume production would be carried out by private actors after the consor
tium developed the prototype. Private firms need to be involved for two reasons: they
know where market demand is, and they eventually run the show. However, public-sector
agents are better placed to deal with technological and financial uncertainty. This South
African case can therefore be considered one of ‘design failure’ rather than a ‘targeting
failure’. The sources are often mixed together. While one might expect more cases of tar
geting failure, this is not always the case. Uncertainty diminishes if targeting is seen in
terms of identifying potential or existing markets as long as the private sector, with its
knowledge of markets, is involved. If not on the frontier, the targets may be obvious be
cause there is often a clear benchmark case which enables niches between existing firms
Page 12 of 19
ty-based View
and projects to be identified. Numerous public initiatives fail because of design or capa
bility (low execution capability) failure. (p. 673)
23.5 Concluding Remarks

Reports on the effectiveness of industrial policy in the empirical literature provide con
flicting or ambiguous results. This may be so because it is difficult to verify the average
positive impact of industrial policy as the effects tend to appear only in certain condi
tions, depending upon specific contexts (countries or sectors). Moreover, the results
change depending upon the criteria used to assess the effectiveness of industrial policy.
As discussed in this chapter, while expansion of production volume and sales to increase
market share had been the primary goal in the earlier stages of catch-up, productivity has
become an important criterion only since the late 1980s, when the Korean government
shifted its policy from tariffs to R&D subsidies as well as joint R&D (Jung and Lee, 2010).
Given that structural change in an economy is a long-term process, the idea of adopting
different policy tools over time is consistent with the reasoning that industrial policy
should deal with the various dimensions of capabilities of firms and industries in the late
comer countries. Different tools are necessary, depending on whether the target involves
simple operational or production capabilities, investment capabilities, or advanced tech
nological capabilities.
In Korea, tariffs and other forms of protection led to export and output expansion through
fixed investment during the early period (1970s and 1980s), according to the study by
Shin and Lee (2012), whereas Jung and Lee (2010) find that for a later period (the
mid-1980s to 2005), R&D investment stimulated by tax exemptions led to productivity
growth. The two studies find that for both periods, the disciplinary impact of export orien
tation is significant, because it promoted the rents associated with tariffs (earlier period)
and with an oligopolistic market structure (late period) used for fixed (earlier period) and
R&D investments (later period), respectively. In sum, while the specific tools and contents
of the industrial policies were different, they both allowed some rents for the targeted
sectors or firms so that the extra profits could be used to build variants of capabilities,
such as production capabilities in the case of tariffs or technology licensing in the 1970s,
investment capabilities in entry control in the 1980s, and technological (R&D) capabili
ties in the case of public‒private joint R&D in the 1990s.
Such a dynamic shift in policy tools is not simply imposed by the government but also re
flects the available and/or desired level of firm capability, which changes over time. Al
though Korea had been growing fast with exports of labour-intensive low-end goods, this
growth strategy reached a peak by the mid-1980s. Around that time, Korea saw an in
crease in its own wage rate, which coincided with the emergence of lower-wage countries
able to compete in world markets. As Korean firms realized the need to upgrade to high
er-end or value-added goods, they began, for the first time, to establish in-house R&D
Page 13 of 19
ty-based View
centres, after which the industrial policy tools switched towards tax exemption on R&D
(Lee, 2013b; Lee and Kim, 2010).
The Korean experience indicates a dynamic shift in the form of government ac
(p. 674)
tivism from traditional industrial policy (tariffs and undervaluation) in the early stages of
development, to technology policy (R&D subsidies and public‒private R&D consortia) in
the later stages. This dynamic shift is required for a developing country to evolve from
low-income to middle-income status, and eventually move on up to higher-income status.
It can be argued that without such a shift, any country may be stuck in what is called the
middle-income trap, struggling to remain competitive as a site for low-cost, high-volume
production (World Bank, 2010; Lee, 2013b).
Acknowledgements
The author thanks the editors of this volume and the participants of the Review Work
shop, held in Addis Ababa in September 2019, in particular Arkebe Oqubay and John
Weiss, for their useful comments and discussion.
References
Aghion, Philippe, Mathias Dewatripont, Liqun Du, Ann Harrison, and Patrick Legros
(2011) ‘Industrial Policy and Competition’. CEPR Discussion Paper No. 8619.
Akamatsu, Kaname (1962) ‘The Historical Patterns of Economic Growth in Developing

Countries’, Developing Economies 1(s1): 3‒25.
Anderson, Philip and Michael Tushman (1990) ‘Technological Discontinuities and Domi
nant Designs: A Cyclical Model of Technological Change’, Administrative Science Quarter
ly 35: 604‒33.
Beason, Richard and David E. Weinstein (1996) ‘Growth, Economies of Scale, and Target
ing in Japan (1955–1990)’, The Review of Economics and Statistics 78(2): 286‒95.
Bell, Martin and Keith Pavitt (1993) ‘Technological Accumulation and Industrial Growth:
Contrasts between Developed and Developing Countries’, Industrial and Corporate
Change 2(1): 157‒210.
Borensztein, Eduardo, Jose R. De Gregorio, and Jongwha Lee (1998) ‘How Does Foreign
Investment Affect Growth?’, Journal of International Economics 45(1): 115‒35.
Byun, Hyung-Yoon and Dong Chul Park (1989) ‘Development of Capitalism in Korea: For
mation and Development of Monopolistic Capital’, in Hyung-Yoon Byun (ed.) The Korean
Economy. Seoul: Yupoong Publishing Company (in Korean).
Chandra, Vandana (ed.) (2006) Technology, Adaptation, and Exports: How Some Coun
tries Got It Right. Washington, DC: World Bank.
Page 14 of 19
ty-based View
Chung, Moon Young and Keun Lee (2015) ‘How Absorptive Capacity is Formed in a Late
comer Economy: Different Roles of Foreign Patent and Know-how Licensing in Korea’,
World Development 66: 678–94.
Cohen, Wesley M. and Daniel A. Levinthal (1989) ‘Innovation and Learning: The Two
Faces of R&D’, The Economic Journal 99(397): 569‒96.
Cohen, Wesley M., and Daniel A. Levinthal (1990) ‘Absorptive Capacity: A New Perspec
tive on Learning and Innovation’, Administrative Science Quarterly 35(1): 128‒52. (p. 675)
Enos, John L. and Woo Hee Park (1988) The Adoption and Diffusion of Imported Technolo
gy: The Case of Korea. New York: Croom Helm.
Evenson, Robert E., and Larry Westphal (1995) ‘Technological Change and Technology
Strategy’, in Jere Behrman and T. N. Srinivasan (eds) Handbook of Development Econom
ics, Vol. 3. Amsterdam: North Holland, pp. 2209–99.
Flatten, Tessa, Andreas Engelen, Shaker Zahra, and Malte Brettel (2011) ‘A Measure of
Absorptive Capacity: Scale Development and Validation’, European Management Journal
29(2): 98–116.
Freeman, Chris and Luc Soete (1997) ‘Development and the Diffusion of Technology’, in
Chris Freeman and Luc Soete (eds) The Economics of Industrial Innovation. London: Pin
ter Publishers, pp. 351–71.
Johnson, Chalmers (1982) MITI and the Japanese Miracle: The Growth of Industrial Policy
1925–1975. Palo Alto, CA: Stanford University Press.
Joo, Si H. and Keun Lee (2010) ‘Samsung’s Catch-up with Sony: An Analysis Using US
Patent Data’, Journal of the Asia Pacific Economy 15(3): 271–87.
Jung, Moosup and Keun Lee (2010) ‘Sectoral Systems of Innovation and Productivity
Catch-up: Determinants of the Productivity Gap between Korean and Japanese Firms’, In
dustrial and Corporate Change 19(4): 1037–69.
Keller, Wolfgang (1996) ‘Absorptive Capacity: On the Creation and Acquisition of Technol
ogy in Development’, Journal of Development Economics 49(1): 199–227.
Kim, Linsu (1997) Imitation to Innovation: The Dynamics of Korea’s Technological Learn
ing. Boston, MA: Harvard Business School Press.
Kim, Linsu (1998) ‘Crisis Construction and Organisational Learning: Capability Building
in Catching-up at Hyundai Motors’, Organization Science 9(4): 506‒21.
Kiyota, Kozo and Tetsuji Okazaki (2005) ‘Foreign Technology Acquisition Policy and Firm
Performance in Japan, 1957–1970: Micro-aspects of Industrial Policy’, International Jour
nal of Industrial Organization 23(7–8): 563‒86.
Page 15 of 19
ty-based View
Kojima, Kiyoshi (1973) ‘A Macroeconomic Approach to Foreign Direct Investment’, Hitot
subashi Journal of Economics 14: 1‒21.
Korea Development Bank (1991) Analysis of Effects of Technology Acquisition. Seoul: Ko
rea Development Bank (in Korean).
Korea Industrial Technology Association (1995) Records of Technology Acquisition Con

tracts: 1962–1995. Seoul: Korea Industrial Technology Association (in Korean).
Lane, Peter J., Balaji R. Koka, and Seemantini Pathak (2006) ‘The Reification of Absorp
tive Capacity: A Critical Review and Rejuvenation of the Construct’, Academy of Manage
ment Review 31: 833‒63.
Laursen, Keld and Valentia Meliciani (2002) ‘The Relative Importance of International vis‐
à‐vis National Technological Spillovers for Market Share Dynamics’, Industrial and Corpo
rate Change 11(4): 875‒94.
Lee, Jaymin (1996) ‘Technology Imports and R&D Efforts of Korean Manufacturing
Firms’, Journal of Development Economics 50(1): 197‒210.
Lee, Jong-wha (1996) ‘Government Interventions and Productivity Growth’, Journal of

Economic Growth 1: 391‒414.
Lee, Keun (2006) ‘The Washington Consensus and East Asian Sequencing: Understanding
Reform in East and South Asia’, in José María Fanelli and Gary McMahon (eds) Under
standing Market Reforms, Vol. 2. Basingstoke: Palgrave, pp. 99–140.
Lee, Keun (2013a) Schumpeterian Analysis of Economic Catch-up: Knowledge, Path-cre

ation and the Middle-income Trap. Cambridge: Cambridge University Press. (p. 676)
Lee, Keun (2013b) ‘How Can Korea Be a Role Model for Catch-up Development? A “Capa
bility-based View”’, in Augustin K. Fosu (ed.) Achieving Development Success: Strategies
and Lessons from the Developing World. Oxford: Oxford University Press, pp. 24–49.
Lee, Keun (2013c) ‘Capability Failure and Industrial Policy to Move beyond the Middle-in
come Trap: From Trade-based to Technology-based Specialization’, in Joseph E. Stiglitz
and Justin Yifu Lin (eds) The Industrial Policy Revolution, Vol. I. Basingstoke: Palgrave
Macmillan, pp. 244–72.
Lee, Keun (2017) ‘Smart Specialization with Short-cycle Technologies and Implementa
tion Strategies to Avoid Both Target and Design Failures’, in Slavo Radosevic, Adrian Cu
raj, Radu Gheorghiu, Liviu Andreescu, and Imogen Wade (eds) Theory and Practice of
Smart Specialization. Amsterdam: Elsevier, pp. 201–24.
Lee, Keun (2019) The Art of Economic Catch-up. Cambridge: Cambridge University Press.
Lee, Keun and Yee Kyoung Kim (2010) ‘IPR and Technological Catch-up in Korea’, in Hi
royuki Odagiri, Akira Goto, Atsushi Sunami, and Richard R. Nelson (eds) Intellectual
Page 16 of 19
ty-based View
Property Rights, Development, and Catch Up: An International Comparative Study. Ox
ford: Oxford University Press, pp. 133–67.
Lee, Keun and Franco Malerba (2018) ‘Economic Catch-up as Evolutionary Process’, in
Richard R. Nelson (ed.) Modern Evolutionary Economics: An Overview. Cambridge: Cam
bridge University Press, pp. 172–207.
Lee, Keun and John A. Mathews (2010) ‘From the Washington Consensus to the BeST
Consensus for World Development’, Asian-Pacific Economic Literature 24(1): 86‒103.
Lee, Keun and John A. Mathews (2012) ‘Firms in Korea and Taiwan: Upgrading in the
Same Industry and Entries into New Industries’, in Edmund Amann and John Cantwell
(eds) The Innovative Firms in the Emerging Market Economies. Oxford: Oxford University
Press, pp. 223–48.
Lee, Keun, Chaisung Lim, and Wichin Song (2005) ‘Emerging Digital Technology as a
Window of Opportunity and Technological Leapfrogging: Catch-up in Digital TV by the Ko
rean Firms’, International Journal of Technology Management 29(1–2): 40‒63.
Lee, Keun, Sunil Mani, and Qing Mu (2012) ‘Divergent Stories of Catch-up in Telecom:
China, India, Brazil, and Korea’, in Franco Malerba and Richard R. Nelson (eds) Economic
Development as a Learning Process. Cheltenham: Edward Elgar, pp. 21–71.
Mathews, John A. (2002) ‘The Origins and Dynamics of Taiwan’s R&D Consortia’, Re
search Policy 31: 633‒51.
Mowery, David C. and Joanne E. Oxley (1995) ‘Inward Technology Transfer and Competi
tiveness: The Role of National Innovation Systems’, Cambridge Journal of Economics 19:
67‒93.
OECD (1996) Reviews of National Science and Technology Policy: Republic of Korea.
Paris: OECD.
Pack, Howard (1992) ‘Learning and Productivity Change in Developing Countries’, in Ger
ald K. Helleiner (ed.) Trade Policy, Industrialization, and Development: New Perspectives.
Oxford: Clarendon Press, pp. 21–44.
Park, Kyoo-Ho and Keun Lee (2006) ‘Linking the Technological Regime to the Technologi
cal Catch-up: Analyzing Korea and Taiwan Using the US Patent Data’, Industrial and Cor
porate Change 15(4): 715‒53.
Pérez, Carlota and Luc Soete (1988) ‘Catching-up in Technology: Entry Barriers and Win
dows of Opportunity’, in Giovanni Dosi et al. (eds) Technical Change and Economic Theo
ry. London: Pinter Publishers, pp 459–79. (p. 677)
Radosevic, Slavo, Adrian Curaj, Radu Gheorghiu, Liviu Andreescu, and Imogen Wade
(eds) (2017) Theory and Practice of Smart Specialization. Amsterdam: Elsevier.
Page 17 of 19
ty-based View
Shin, Hochul and Keun Lee (2012) ‘Asymmetric Trade Protection Leading Not to Produc
tivity But to Export Share Change: The Korean Case from 1967 to 1993’, Economics of
Transition 20(4): 745–85.
Song, Sungsoo (2002) ‘Historical Development of Technological Capabilities in Korean

Steel Industry: Posco from the 1960s to the 1990s’, PhD thesis, Seoul National University
(in Korean).
Stiglitz, Joseph E. and Justin Yifu Lin (eds) (2013) The Industrial Policy Revolution, Vol. I.
Basingstoke: Palgrave Macmillan.
Swart, Gerhard (2015) ‘Innovation Lessons Learned from the JOULE EV Development’,
The IAMOT 2015 Conference Proceedings.
Tushman, Michael and Philip Anderson (1986) ‘Technological Discontinuities and Organi
zational Environments’, Administrative Science Quarterly 31: 439‒65.
Westphal, Larry E., Linsu Kim, and Carl J. Dahlman (1985) ‘Reflections on the Republic of
Korea’s Acquisition of Technological Capability’, in Nathan Rosenberg and Claudio
Frischtak (eds) International Technology Transfer: Concepts, Measures, and Compar
isons. New York: Praeger, pp. 167‒221.
World Bank (2005) Economic Growth in the 1990s: Learning from a Decade of Reform.
Washington, DC: World Bank.
World Bank (2010) Exploring the Middle-income Trap. World Bank East Asia Pacific Eco
nomic Update: Robust Recovery, Rising Risks, Vol. 2. Washington, DC: World Bank.
Zahra, Shaker and Gerard George (2002) ‘Absorptive Capacity: A Review, Reconceptual
ization, and Extension’, Academy of Management Review 27(2): 185–203.
Notes:
(1) We follow Kiyota and Okazaki (2005) in using the term ‘foreign technology acquisi
tion’.
(2) These foreign-trained engineers played very substantial roles in the early days of POS
CO, with their share in the workforce in charge of facility operation and maintenance
reaching 62 per cent and 24 per cent, respectively (Song, 2002: 128).
(3) However, this was not evident during the period from 1976 to 1981, when heavy in
vestment in social overhead capital increased the demand for technology in cement and
utility firms, such as electricity.
(4) The case of digital TV production is further explained by Lee et al. (2005). A direct
comparison of Samsung and Sony can be found in the work of Joo and Lee (2010).
Page 18 of 19
ty-based View
Keun Lee
Keun Lee is a professor of economics at the Seoul National University, fellow of the
CIFAR programme on Innovation, Equity and Prosperity, and founding director of the
Center for Economic Catch-up. He is an editor of Research Policy and an associate
editor of Industrial and Corporate Change. He served as president of the Internation
al Schumpeter Society (2016‒18), a member of the Committee for Development Poli
cy of UN (2014‒18), and a council member of the World Economic Forum (2016‒19).
He is the winner of the 2014 Schumpeter Prize for his monograph Schumpeterian
Analysis of Economic Catch-up (Cambridge University Press, 2013), as well as the
2019 Kapp Prize from the EAEPE for his article on national innovation systems. He
obtained his PhD in economics from the University of California, Berkeley. One of his
most cited articles is the paper ‘Korea’s Technological Catch-up’ published in Re
search Policy, with 1,100 citations (Google Scholar). His H-index is 40, with about
100 papers with more than ten citations. His latest book is The Art of Economic
Catch-up: Barriers, Detours, and Leapfrogging (Cambridge University Press, 2019).
Page 19 of 19
Industrial Policy and Industrialization in South East Asia
Industrial Policy and Industrialization in South East

Asia
Rajah Rasiah
The latecomer industrialization thesis has acted as a powerful instrument in promoting

industrialization. However, there is little explanation in the literature as to why some
economies that attempted to industrialize are facing premature de-industrialization. This
chapter provides an assessment of industrialization in the South East Asian market
economies of Indonesia, Malaysia, the Philippines, Singapore, and Thailand, examining
policies that spurred either strong or little technological upgrading. Driven by an agile
state, Singapore managed to become a developed economy through strong industrial up
grading. Malaysia launched upgrading policies, while Indonesia, the Philippines, and
Thailand introduced ad hoc strategies to support technological upgrading. Foreign
transnational corporations have played a major role in stimulating manufacturing expan
sion in these economies. However, while all five countries have begun to experience de-
industrialization, Singapore’s de-industrialization has been accompanied by strong tech
nological upgrading, while the remaining four countries remain stuck in the middle-in
come trap owing to a lack of technological upgrading.
Keywords: industrialization, industrial policy, de-industrialization, technological upgrading, developmental states,

South East Asia
24.1 Introduction
THE latecomer industrialization thesis has acted as a powerful policy instrument in the
promotion of the industrialization, as well as the economic growth, of the United States,
Germany, Japan, and the East Asian economies. However, there has been little explana
tion in the literature as to why several other economies that attempted to industrialize
are facing premature deindustrialization. Thailand was never colonized; Indonesia,
Malaysia, the Philippines, and Singapore gained independence in 1945, 1957, 1898, and
1959 respectively. Despite introducing industrial policies as a major instrument to stimu
late economic development, only Singapore has managed to sustain sufficiently rapid
growth to enjoy a per capita income of US$64,581 in 2018 (current US$), 5.7 times high
Page 1 of 35
er than that of Malaysia (US$11,239), and 8.9 times higher than that of Thailand.
Malaysia and Thailand enjoyed the second- and third-highest per capita income in South
East Asia in 2018. On average in US$ 2010 prices, the GDP per capita incomes of Singa
pore, Thailand, Malaysia, Indonesia, and the Philippines grew by 5.0, 4.2, 3.8, 3.2, and
1.8 per cent per annum respectively between 1960 and 2018.1 Thailand’s growth rate
over the period 1960‒2018 was higher than that of Malaysia, even though the per capita
income in US$ 2010 prices of Malaysia (US$12,109.5) was much higher than of Thailand
(US$6,361.6) in 2018. This was because Thailand’s starting per capita income (US$570.9)
base was much smaller than Malaysia’s (US$1,354.0).
This chapter seeks to explain how industrial policy has driven sustained rapid eco
(p. 682)
nomic growth and structural change in Indonesia, Malaysia, the Philippines, Singapore,
and Thailand.2 However, whereas Singapore has carefully managed upgrading to become
a developed country, the rest are stuck in the middle-income trap. Going beyond the typi
cal broad-brush framework used by latecomer exponents, the chapter examines the short
comings associated with policies in these countries that failed to spur differentiation and
division of labour, and technological upgrading—both across and within industries.
Malaysia launched explicit industrial policies and developed the science, technology, and
innovation infrastructure to stimulate upgrading, while Indonesia, the Philippines, and
Thailand had ad hoc strategies in place to support technological upgrading. Unlike the
experiences of Japan, South Korea, and Taiwan, where industrialization was propelled by
national firms, foreign transnational corporations (TNCs) have played a major role in
stimulating manufacturing expansion in the South East Asian market economies. Howev
er, all five countries have begun to experience de-industrialization. As a share of value
added in GDP, Thailand and Malaysia enjoyed the highest expansion of manufacturing
among these countries, but they have neither managed to stimulate upgrading to high
value-added activities, nor succeeded in developing national firms to reach the technolo
gy frontier. Singapore has managed to specialize in the high-value-added segments in
shipbuilding and petrochemicals, in which it enjoys relative comparative advantage, and
in TNC-driven export-oriented manufacturing, by quickly shifting its industrial specializa
tion to sustain industrial upgrading.
This chapter seeks to explain the industrial policies introduced in Indonesia, Malaysia,
the Philippines, Singapore, and Thailand, and their impact on industrialization. We begin
with key theoretical arguments to locate the analysis, before discussing first, the import-
substitution industrialization (ISI) policies, and second, the export-oriented industrializa
tion (EOI) policies implemented in the South East Asian market economies. Section 24.3
assesses their industrialization and de-industrialization experiences, followed by an exam
ination of the impact of technological upgrading initiatives in these countries. Section
24.7 presents the conclusions.
Page 2 of 35
24.2 Theoretical Considerations

South East Asia’s industrialization did not follow the paths begun by latecomers, such as
the United States, Germany, Japan, South Korea, and Taiwan. The early introduction of
(p. 683) export-oriented industrialization in the South East Asian market economies of In
donesia, Malaysia, the Philippines, Singapore, and Thailand largely took the form of at
tempts to diversify exports to reduce dependence on a narrow range of commodities and
avoid the occurrence of ‘fallacy of composition’ problems (see Prebisch, 1950; Singer,
1950).
The initial focus was on the promotion of light labour-intensive manufactured goods to
generate employment through the stimulation of foreign direct investment (FDI). Indone
sia, Malaysia, the Philippines, and Thailand did introduce ISI but that strategy failed to
contribute significantly to rapid economic growth as it was not accompanied by a rapid
switch to export orientation. In addition, little attention was devoted to technological up
grading. In contrast, the massive growth and structural change enjoyed by Singapore was
achieved by switching completely to export orientation in 1965 and focusing on techno
logical upgrading since 1979.
Smith (1776) and Young (1928) argued over the differentiation and division-of-labour po
tential of manufacturing to produce dynamic increasing returns. Industrial expansion
through technological upgrading can be traced to latecomer strategies. The latecomer
thesis on industrialization originally focused on broad-brush approaches that simply docu
mented how Britain industrialized; the subsequent catch-up experiences of the United
States, Germany, and other countries were documented by Gerschenkron (1952) and
Abramovitz (1956). This argument was later extended by Kaldor (1967), Chang (1994),
and Reinert (2007).3
Amsden (1989, 1993) and Kim (1997) offered empirical dynamics to the latecomer catch-
up thesis by focusing on the diffusion of innovation from abroad using specific host-site
firms in South Korea. They extended the developmental state arguments of Johnson
(1982) and Wade (1990) on how industrial policy was used successfully to stimulate eco
nomic development. It is in this tradition that Mazzucato (2013) was able to frame their
mission-oriented and smart intervention industrial policy arguments. Ocampo (Chapter 3,
this volume) connects policy with technological learning and upgrading through incre
mental innovations. Rasiah (2018) went further by incorporating Schumpeter’s (1934,
1943) concepts of incremental and radical innovations using the experience of firms in
South Korea (e.g. Samsung Semiconductors) and Taiwan (e.g. TSMC), showing how, de
spite their contrasting industrial policies, they successfully shaped catch-up and leapfrog
ging in the integrated circuits industry (Yap and Rasiah, 2017).
South Korea and Taiwan did not integrate into the world economy in the same way as the
South East Asian economies. Both countries focused on the latecomer catch-up model
from the early 1970s by prioritizing national ownership in strategic industries (Amsden,
1989; Wade, 1990). They also introduced monetary policies to ensure that volatile fluctua
Page 3 of 35
tions in capital and trade flows did not destabilize their industrialization (p. 684) initia
tives. While prudent monetary policies ensured that the New Taiwan dollar’s exchange
rate against the US$ did not fluctuate sharply, South Korea imposed foreign exchange
controls from 1961 till the 1970s, including on borrowings from abroad by private enti
ties, maintaining ownership of banks by the state until the 1980s. South Korea also fixed
the won against the US$ between 1974 and 1979 to stem the negative impact of growing
balance-of-payment imbalances and debt service problems following the first oil crisis,
when oil prices rose by a factor of 4 from 1973 to 1975. With targeted firms, especially
chaebols, enjoying subsidized interest rates, the deliberate imposition of high arbitrage
interest rate differentials between borrowing and lending assisted the government to
gradually lower its external debt, as well as support targeted firms (Chang, 1994).4
The generation and adaptation of technologies, incentive systems, and appropriation of

knowledge flows were critical for stimulating individuals, firms, and organizations to con
nect and coordinate with knowledge nodes that provide the synergy for catching up.
Since the knowledge necessary to stimulate innovation can be regarded as a public good,
its governance should prioritize technological upgrading. Embodied knowledge—both for
eign and domestic—is continuously adapted to raise industrial productivity. The organiza
tional set-up can vary between countries, as initial conditions and economic structures af
fect how they are shaped. Combining the key arguments on the manufacturing experi
ences of particularly South Korea and Taiwan (the most recent successful industrializers),
we present a stylized model of industrial policy in Figure 24.1, showing a framework of
how incremental and radical innovations (including interactive learning) support techno
logical upgrading in successful economies.
Institutions must be structured to pursue macroeconomic policies that provide the financ
ing (incentives and grants) essential to support innovative economic activity, as well as in
sulation from external shocks. Institutions need strengthening to meet stringent appraisal
standards to check unproductive rent seeking. Amsden (1993) and Kim (1997) provide a
lucid account of innovation appropriation and economic catch-up from foreign sources in
South Korea and Taiwan. The public-good characteristics of knowledge creation and ap
propriation (innovation) were harnessed effectively in Taiwan and South Korea.
The critical organizations relevant to institutionalizing knowledge creation and appropria

tion on a national scale, such as universities, public labs, standards organizations, and in
cubators at science parks, must play an active role in stimulating incremental and radical
innovations with a focus on evolving experiential and tacit knowledge (see Penrose, 1959;
Polanyi, 1966; Dosi, 1988; Rasiah, 1995; Best, 2001).
While first movers initiate cycles of innovation, latecomers engage in incremental innova
tion (Schumpeter, 1943). As economies move from the ‘least-developed countries’ catego
ry to middle-income countries, governments should ensure that gross (p. 685) expenditure
on R&D (GERD) gradually rises, with a focus on R&D targeted at generating new stocks
of knowledge. Instead of simply seeking to gradually replace government expenditure
with business expenditure in GERD, non-business public expenditure should be included
Page 4 of 35
to address the public goods characteristics of science, technology, and innovation (STI)
infrastructure.
Figure 24.1 Innovation-driven industrial policy
Source: Adapted from Rasiah (2018).
24.3 Import-substitution Policies

Influenced by advice from the World Bank and the United Nations Industrial Development
Organization (UNIDO) to limit manufacturing to consumption goods, in the late 1950s In
donesia, Malaysia, the Philippines, Singapore, and Thailand all began to promote ISI to
replace imports (Chenery et al., 1986). Only in some extended backward-linkage indus
tries, such as iron and steel, and cement, was there a focus on intermediate products.
Corruption through smuggling was rampant, with rising tariffs encouraging clientelist re
lationships. Capital goods were not emphasized until the 1970s in Indonesia and Philip
pines, and the 1980s in Malaysia and Thailand. Singapore (p. 686) introduced EOI immedi
ately after its separation from Malaysia in 1965,5 so the ISI focus in this chapter will be
confined to the other four countries.
24.3.1 Indonesia
Suharto became president of Indonesia in 1966, following a period of international isola

tion arising from Sukarno’s declaration of independence and nationalist policies. The New
Order government had the support of the United States as it sought to promote import-re
placement policies following the launching of the First Plan (1969‒74). Foreign owner
ship was discouraged but joint ventures were allowed (Anwar, 1979). Incentives and im
port credits cheapened the processing and assembly of consumption goods in the country.
Incentives for existing projects included carry-over losses over the first six years, acceler
ated depreciation allowance up to a maximum of 25 per cent, exemption from dividend
tax, an investment allowance of 20 per cent over four years, and exemption from property
tax (Poot, 1981). The Investment Coordinating Board offered new projects with additional
Page 5 of 35
fiscal incentives, which included exemptions from import duties and restrictions on im
ported machinery and raw materials, and exemption from corporate income tax, dividend
tax, and company tax on reinvestments over five years.
Booming oil revenues following the first oil crisis (1973‒75) supported the launching of
the aircraft industry in 1976 with special support from the state. Krakatau Steel was also
launched by the state to support heavy industrialization (Prawiro, 1998). While diversifi
cation into manufacturing was viewed by policymakers as important for averting the ‘fal
lacy of composition’ syndrome, three major instruments undermined the growth of ISI in
dustries in Indonesia. First, protection and incentives were given without clear perfor
mance standards, while there was no condition requiring import substitution industries to
switch to EOI as in South Korea and Taiwan. Second, there was no institutional support
for technological upgrading by firms. The combination of these two drawbacks saw a rise
in production costs for downstream firms, and a lack of pressure to bring costs down.
Third, the oil boom that began in 1973 (which raised the share of oil to 50 per cent of In
donesian exports), resulted in the appreciation of the rupiah, which undermined manufac
turing as cheap imports replaced domestic production over the period 1973‒80 (Tee,
2006). Although manufacturing grew by an annual average of 11.3 per cent over the peri
od 1970‒79, (almost double the growth over 1960‒69), its ‘Dutch Disease’ effect stunted
inward-oriented industries, such as food and beverages.
The steep fall in oil prices after 1982 forced Indonesia to consider a shift to EOI, and par
ticularly to manufactured exports. However, until the late 1980s serious balance-of-pay
ments problems forced the government to focus on monetary stabilization, (p. 687) finan
cial deregulation to allow state banks to set deposit and lending rates, and a massive de
valuation of the rupiah in March 1983 (Hill, 1996: 155; Prawiro, 1998). Steel and aircraft
manufacturing continued, as did state-led ISI, with the launching of large state-owned re
source-processing heavy industries, which included the Timor‒KIA joint venture deal to
assemble cars (Rasiah, 2010).
24.3.2 Malaysia
Malaysia raised tariffs on final consumption goods following the Pioneer Industry Ordi
nance of 1958. A combination of a small domestic market and a lack of technological up
grading policies confined manufacturing to ‘screwdriver operations’ (Edwards, 1975). Not
only had manufacturing growth become sluggish by the mid-1960s (Malaysia, 1965), the
sector had also become a major source of imports as capital goods and a variety of con
sumption goods continued to be imported on a large scale (Alavi, 1996).
The extension of ISI policies that took place from the 1970s without due attention to their
impact on downstream industries and technological upgrading merely sapped the domes
tic economy of resources. Protection levels remained high in the country (Edwards, 1975;
Alavi, 1996), though estimations of effective rates of protection (ERP) did not take ac
count of tax and tariff holidays of firms located in export-processing zones (EPZs), which
were opened in Malaysia from 1972. Free-trade zones (FTZs) and licensed manufacturing
Page 6 of 35
warehouses followed (Rasiah, 1988). The lack of export promotion as a condition for ISI
undermined the competitiveness of firms. The lack of integration between the IS and the
export-oriented sectors also discouraged linkages between the two (Jomo and Edwards,
1993).
A second round of ISI began in 1981 with the Fourth Malaysia Plan and the launch of the
Look East Policy (Malaysia, 1981). Perwaja (steel), Proton (cars), and Kedah Cement were
among the heavy industries promoted by the Heavy Industry Corporation of Malaysia (HI
COM) in an attempt to imitate South Korea through subsidized credit and heavy protec
tion (Malaysia, 1986). The construction of the North‒South Highway in the 1980s helped
to stimulate these industries. While the focus of ISI was on national firms for the first
time, the lack of effective selection and appraisal, technological upgrading, and exposure
to export markets rendered the first two industries inefficient, though Proton did export
using a questionable export subsidization strategy in the late 1980s (Rasiah, 1995). Nev
ertheless, the massive expansion of construction and technology transfer made cement
and construction firms competitive (Chang et al., 2016).
Arguably the most catastrophic reason for the failure of IS strategies in Malaysia was the
lack of policies to stimulate technological upgrading. With little emphasis on developing a
science and technology infrastructure in the country, production costs began to rise,
cheaper locations emerged from abroad (e.g. China and Vietnam (p. 688) following eco
nomic reforms in 1978 and 1986 respectively), and Malaysian manufacturing could not
upgrade to higher value-added activities. Meanwhile, while export diversification helped
avert the ‘fallacy of composition’ syndrome, the appreciation in the ringgit in the face of
chronic current account deficits between 1989 and 1997 caused by strong FDI and port
folio equity inflows undermined inward-oriented industries, including food and beverages
(Rasiah, 2018).
24.3.3 The Philippines
The Philippines gained independence from Spain in 1898; largely due to balance-of-pay
ment (BoP) problems, ISI policies were introduced in 1949 (Bautista, 1983). Controls on
both imports and foreign exchange became the prime instruments to curb imports, and
the manufacturing sector lacked competitiveness due to lack of policies to stimulate tech
nological upgrading. By the 1960s it became obvious that ISI had become a drain on the
Philippines’ economy, which drove the government to stimulate FDI inflows. Reconstruc
tion assistance from the United States offset the negative balance of trade in the Philip
pines. However, while import restrictions stimulated manufacturing, import controls re
sulted in the substitution of domestic production by imports of textiles, paper and paper
products, and non-metallic mineral products. Nevertheless, manufacturing grew at 12 per
cent per annum on average between 1960 and 1970 (Bautista, 1983).
However, the Philippines faced economic and political turmoil in the late 1960s as the
peoples’ car launched under ISI policies by the government failed to take off (Ofreneo,
2015) which, along with rising imports, caused a return of the BoP crisis in 1969. Eco
Page 7 of 35
nomic growth remained sluggish, particularly in the manufacturing sector which was
heavily dependent on processing of primary commodities such as coconut products, sug
arcane, forest products, copper, and gold. To combat its BoP difficulties, the government
implemented the Republic Act 5186 and the Export Incentives Act of 1970 to encourage
FDI through 100 per cent ownership of pioneering industries (textiles and garments, and
electrical and electronics). Non-pioneering industries were allowed to be 40 per cent for
eign owned and were offered accelerated depreciation allowance, tax exemptions on im
ported capital equipment, and tax holidays.
However, in the 1970s the bulk of the FDI was heavily import dependent (Ofreneo, 2015).
Additionally, non-tariff barriers were implemented in the country to protect industries like
vehicles and equipment, textiles, basic metals, and fabrics in the principal customs area.
Instead of discouraging imports, the devaluation of the peso negatively affected the ISI
industries that were dependent on imports of capital equipment, intermediate goods, and
raw materials. While the proliferation of EPZs raised manufacturing growth to 8.6 per
cent in the period 1970‒79, driven largely by TNCs assembling and processing imported
items from their own value chains, the Philippines’ economy suffered from rising external
debt caused by the oil crises of 1973‒75 (p. 689) and 1979‒80. The strong oligarchy sup
porting the government applied clientelist pressures to retain unproductive protection
rents. Growing poverty and unemployment drove hundreds of thousands of Filipinos
abroad, which eventually made overseas remittances a major source of revenue for the
Philippines.
24.3.4 Thailand
Following the First Development Plan (1961‒66), but especially after the enactment of
the New Investment Promotion Act of 1962, the Thai government raised tariffs and of
fered indirect subsidies to promote ISI. Three groups of industries were promoted by the
Board of Investment (BoI). Category I industries were given full exemption from import
duties on raw materials and intermediate inputs and business taxes, category II indus
tries, one-half exemption, and category III industries, one-third exemption. Tariffs target
ed at final goods for the domestic market continued to be promoted under the Second
Five-Year Plan (1967‒71), but duties on imported inputs were reduced to a third. Manu
facturing value added grew at 11 per cent on average per annum over the 1960s with ISI
policies the prime propellant (Bautista, 1983: 23).
Food processing, beverages, and tobacco dominated manufacturing in the early 1960s,
but began to contract from the late 1960s with the growth of import-replacing industries,
such as petroleum products, transport equipment, non-durable consumer goods, and EOI
industries, such as textiles, clothing, and electronic products (Bautista, 1983: 24). Al
though EOI began in the 1970s, ISI remained potent with the government’s stimulation of
domestic investment under the 1977 Investment Promotion Law, through which the BoI
reconstituted ISI strategies in Thailand (Akrasanee, 1979). The BoI’s promotion of mini
mum plant size also led to the establishment of large-scale manufacturing firms, which in
general were found to be more capital intensive and import dependent than small firms
Page 8 of 35
(Tambunlertchai and Loohawenchit, 1981). While manufacturing value added still grew by
an average of 11 per cent per annum during 1970‒79, chronic BoP deficits and mounting
external debt service (followed by a 2.5 times rise in oil prices in 1979‒80) drove the gov
ernment to seek assistance from the World Bank and the International Monetary Fund
(IMF). Between 1989 and 1997, with massive inflows of FDI and portfolio equity invest
ment and no trade surplus, the baht began to appreciate, leading to the ‘Dutch Disease’
effect of damage to inward-oriented industries (Akrasanee, 1979).
24.3.5 Summary
In short, high domestic production costs and the absence of initiatives to gradually lower
protection levels undermined export competitiveness, but the inability of ISI to support
long-term growth in manufacturing was principally due to a lack of effective (p. 690) tech
nological upgrading policies to transform the economies of Indonesia, Malaysia, the
Philippines, and Thailand from low to high value-added activities. To make matters worse,
unproductive rentier operations through clientelism behind high tariffs and quotas aggra
vated the state’s capacity to sustain them over the long run.
24.4 Export-oriented Policies

A combination of small domestic markets, the disastrous costs of poorly conceived IS poli
cies, and TNCs relocating their labour-intensive production stages drove the South East
Asian market economies to switch to EOI policies. Singapore and Thailand attracted for
eign TNCs with tariff-free operations and tax holidays. Malaysia launched manufacturing
as the engine of growth through its New Economic Policy (Malaysia, 1971). Indonesia al
lowed export-oriented TNCs to relocate in Batam from 1992 with total ownership and tar
iff-free and tax holiday privileges (Prawiro, 1998).
24.4.1 Indonesia
With the Plaza Accord of 1985 and further falls in the oil price, Indonesia effectively be
gan EOI in 1986. The withdrawal of the generalized system of preferences (GSP) from
South Korea, Taiwan, and Singapore in 1988 drove massive FDI inflows into South East
Asia. The surge in Indonesia’s manufactured exports since 1987 is largely a consequence
of the introduction of the duty exemption and drawback scheme. The Badan Koordinasi
dan Penanaman Modal (Investment Coordination Board) promotes industrial investment
in Indonesia (Prawiro, 1998).
Indonesia also offered total equity ownership of export-oriented firms in selected loca
tions, such as Bantam, leased to Temasik Holdings of Singapore to manage from the late
1980s. As the EOI strategy began to proliferate in Indonesia, continued stagnation in the
oil price and cumbersome integration of inward-oriented firms in heavy industries caused
by dualistic policies restricted the capacity of EOI to synergize the whole economy. The
rupiah appreciating against major currencies from the 1990s undermined the IS indus
tries owing to cheap imports, which aggravated the balance of payments and debt ser
Page 9 of 35
vice. The unsustainable exchange rates affecting most East Asian economies caused In
donesia to succumb to the 1997‒98 Asian financial crisis, which brought down the rupiah
and resulted in a massive capital flight from Indonesia (UNCTAD, 1996).
The IMF-imposed structural adjustment package, as well as the low exchange rate, began
to attract FDI inflows into Indonesia in 2000, reviving manufacturing growth. The re
structuring included the closure of the Timor‒KIA aircraft manufacturing joint venture
(Rasiah, 2010). Although Indonesian manufacturing was not seriously affected (p. 691) by
the 2007‒08 global financial crisis, since its trade was decoupled from exports to the
United States, a combination of cumbersome administrative procedures and a lack of
technological upgrading undermined further growth.
24.4.2 Malaysia
Although the Investment Incentives Act of 1967 initiated the shift to EOI, it was only fol
lowing the opening of FTZs (and their equivalent in licensed manufacturing warehouses)
in 1972 that export-oriented manufacturing expanded strongly (Malaysia, 1965, 1976).
Through aggressive promotion by the Malaysian Industrial Development Agency (an or
gan of the Ministry of International Trade and Industry (MITI)),6 EOI in the 1970s focused
on lowering unemployment with light industries and generating exports. Giant TNCs from
the United States, Europe, and Japan relocated labour-intensive operations to Malaysia.
The electrical and electronics, and textile and clothing industries dominated EOI in the
1970s (Rasiah, 1995).
Without significant backward linkages, the fortunes of EOI fluctuated with the vicissi
tudes of volatile external demand swings. Hence, not only did manufacturing wages grow
little in the 1970s, but workers were also often exposed to retrenchments. Although a sec
ond round of ISI policies beginning in 1981 changed the emphasis to heavy industrializa
tion, rising external debt service and falling export commodity prices pressured the gov
ernment to revive EOI policies, and new financial incentives were offered to TNCs to in
vest in Malaysia from the mid-1980s (Rasiah, 1988).
As the EOI phase gained momentum with massive inflows of FDI from Japan, South Ko
rea, Taiwan, Hong Kong, and Singapore in the late 1980s, the government shifted its fo
cus from simply creating jobs to promoting technological upgrading in the 1990s. Plans
were announced limiting financial incentives to capital- and technology-intensive TNCs,
while R&D incentives were offered to stimulate technological upgrading. Parastatals were
launched to create and strengthen the STI infrastructure in the country. The Human Re
source Development Council, Malaysia Industry Government High Technology, the Multi
media Development Corporation, and the Multimedia Super Corridor were launched in
the 1990s for this purpose. The Malaysian Institute of Microelectronics Systems was cor
poratized, and science and technology parks were built over the Western Corridor to
house incubators.
Page 10 of 35
While carrots were given to selected firms to undertake R&D, there was little monitoring
and no sticks were applied to non-performers. The selection procedure for leadership of
STI parastatals and designated technology firms excluded Malaysians with experiential
and tacit knowledge in the respective industries. The preference was for extending the
legacy of creating Bumiputera chief executive officers, which had become the
government’s important goal since the promulgation of the Permodalan (p. 692) Nasional
Berhad in 1978, and had been actively pursued since 1981. The ethnic coloration of in
dustrial policy resulted in standards taking a back seat, rendering the STI parastatals in
effective, while the firms created for support failed to compete in export markets. Perwa
ja, Proton, and Silterra are three examples of national firms that have continued to sap
the national economy. The control of Proton was finally taken over by China’s Geely in
2017 (Zhang et al., 2017), while Silterra is languishing at the bottom of the world semi
conductor ladder (Rasiah and Yap, 2019).
Although EOI was launched in the Philippines in the 1970s when the Board of Investment
(BoI) started promoting export-oriented FDI, ISI continued to lead, as industries located
outside EPZs remained protected. Indeed, there were virtually no links between the tariff-
free operations in the EPZs that attracted primarily electrical and electronics, and textile
and garment firms from abroad. However, chronic BoP deficits and debt service attracted
structural adjustment packages from the IMF that, inter alia, required the liberalization
of tariffs from the 1980s as the Philippines’ integration into ASEAN intensified following
the ASEAN Free Trade Area (AFTA) agreement in 1992.
The Foreign Investment Act of 1991 increased foreign equity ownership to 100 per cent
in non-EPZ industries. Tariffs were reduced on raw materials and finished products,
which helped reduce external debt (Menardo, 2004). Although largely ineffective, these
generous incentives attracted FDI into particularly electrical, electronics, and automotive
components. However, the lack of a strong STI infrastructure, together with smuggling
and the absence of proactive promotion of technological upgrading, restricted manufac
turing expansion so that its share in overall exports only reached 21.0 per cent in 2011
before falling again.
24.4.4 Singapore
As one of the first to promote EOI in Asia, and located strategically as an entrepôt, Singa
pore was in a position to promote industries relevant to the country’s economic develop
ment. Given the high unemployment rates of the 1960s, Singapore’s Economic Develop
ment Board (EDB) promoted labour-intensive export-oriented FDI from 1965 until the sat
uration of the labour market and rising wages shifted the emphasis from 1979 to attract
ing capital- and technology-intensive industries. Exports are promoted principally
through financial incentives, withdrawal of which discourages incompatible industries.
Strategic industries, such as shipbuilding, petrochemicals, information communication,
Page 11 of 35
and bio-technologies, have also been offered R&D grants and facilities (Singapore, 2019a,
2019b).
Singapore fits the mission and smart industrial policy argument advocated by
(p. 693)
Mazzucato (2013). The EDB’s agile leveraging role takes account both of Singapore’s en
dowments and of changes in TNC strategies, to attract the adaptation required in a broad
Schumpeterian sense to sustain technological upgrading and phase out incompatible in
dustries (see also Wong, 2001; Rasiah and Yap, 2016). Three clear strategies can be ob
served in the conduct of the EDB. First, Singapore took advantage of its intermediate role
in trade to launch oil refineries that expanded into petrochemicals and to develop ship
building. Second, incompatible industries were replaced with compatible industries. For
example, Singapore phased out disk-drive firms in 1989‒90 once it became clear that the
industry was labour intensive, and the semiconductor industry from 2015 because of the
lack of related research capability to support upgrading (Rasiah and Yap, 2019). Indeed,
the notable semiconductor firms in Singapore, such as TSMC, Seagate, and Avago, were
only fabricating low-end chips.7 Third, a science park was established to stimulate com
mercialization.
Singapore’s leveraging strategy to transform the economy from low to high value-added
activities by offering R&D grants and facilities attracted regional operations from TNCs.
Similarly, the government’s promotion of shipbuilding and downstream petrochemical in
dustries allowed the country to become industry leaders. Both industries offered Singa
pore strategic positioning as the country gradually rose, through refining imported oil,
and repairing and building ships, rigs, and platforms, to the status of the busiest in the
world. The EDB’s quick approach to world-class firms, handling of registration, customs
coordination, and developing institutional support by linking R&D labs and universities
has gone a long way to wards evolving a world-class shipbuilding and petrochemical in
dustry (Singapore, 2019a, 2019b; ASMI, 2019).
In addition, Singapore’s entrepôt trade has continued to thrive, despite ASEAN

economies gradually establishing their own trade routes. Around half of merchandise ex
ports over the period 1980‒2016 were re-exports, which in 2014 benefited from a net
work of twenty-one regional and bilateral free-trade agreements with thirty-two trading
countries. In fact, merchandise trade through preferential trading agreements accounted
for about 80 per cent of Singapore’s imports and 74 per cent of exports in 2014 (WTO,
2016).8 Through lucrative incentives and grants, and careful selection of foreign experts,
Singapore has evolved a strong biotechnology industry.
24.4.5 Thailand
Although the Export Promotion Act of 1972 initiated Thailand’s export orientation, EOI
only became the dominant industrial strategy of Thailand from the early 1980s (p. 694)
when mounting debt and BoP problems forced the government to accept an IMF structur
al adjustment package amounting to US$150 million. The promotion of exports became a
key instrument to clear the BoP deficit, which on average amounted to around US$1 bil
Page 12 of 35
lion since the second half of the 1970s (Akrasanee, 1979). Through increased production
and productivity, EOI was expected to lead manufacturing value-added growth to an aver
age annual rate of 7.6 per cent during the plan period 1982‒86. The government began
to reduce tariffs, promote regional dispersal of industry, and offer tax incentives to ex
porting firms. Five industrial estates with excellent infrastructure were added to the four
that had existed earlier.
Thailand faced a contraction in manufacturing value added owing to falling commodity

prices and a cyclical downturn in the electronics industry in the mid-1980s. However, se
rious unemployment problems were largely avoided as factory workers returned to rural
farms to support themselves (Phongpaichit and Baker, 1996). The country also experi
enced a massive inflow of export-oriented manufacturing investment from the late 1980s,
which helped fuel rapid growth again. The appreciation of the yen, won, NT dollar, and
Singapore dollar from 1985 following the Plaza Accord, and the withdrawal of the GSP
from South Korea, Taiwan, and Singapore in February 1988 resulted in substantial capital
flight from these countries to South East Asia. Thailand also depreciated the baht to at
tract further FDI inflows. Further deregulation in the 1990s resulted in giant foreign au
tomobile firms relocating their operations to Thailand.
While manufacturing value added expanded over the period 1988‒2007, the lack of tech
nology policies undermined the capacity of manufacturing to sustain wage increases and
technological upgrading from low to high value-added activities. Minor ad hoc incentives
and high-tech facilities were offered to large firms in the automobile (including an auto
mobile university) and electronics industries (Intarakumnerd and Chaoroenporn, 2013),
but the lack of development of a comprehensive STI infrastructure to support incremental
and radical innovations resulted in manufacturing value added’s contribution to GDP
plateauing in 2010 and subsequently declining.
24.4.6 Summary
Overall, EOI offered the appropriate scale to stimulate expansion in exports and employ
ment creation. With foreign frontier TNCs spearheading EOI, the difficulties associated
with competition from abroad were reduced. While Singapore enjoyed first-mover advan
tages in Asia, the creation of EPZs allowed the other four market economies to offer the
excellent basic infrastructure essential for attracting labour-intensive operations. Howev
er, the lack of a focused industrial policy to stimulate upgrading and the lack of stringent
appraisal mechanisms for firms enjoying incentives and grants stifled their upgrading in
global value chains. Lack of connectivity between (p. 695) ISI and EOI firms also discour
aged the development of linkages. Singapore did not face these problems; its EDB man
aged to stimulate industrial upgrading by promoting compatible industries while dis-in
centivizing incompatible industries using a strong appraisal mechanism.
Page 13 of 35
24.5 Industrialization and De-industrialization

While industrial expansion began with the launch of ISI in the 1950s and 1960s, the
growth was fastest during EOI owing to a combination of bigger export markets enjoyed
by TNCs relocating operations from abroad compared to small domestic markets and the
use of labour-intensive technologies. Indonesia, Malaysia, the Philippines, and Thailand’s
experience with ISI drained the economies of foreign exchange from imports of interme
diate inputs and capital machinery equipment, while undermining performance owing to
a lack of strong technological upgrading policies. While the powerful liberalizing initia
tives through Asia-Pacific Economic Cooperation (APEC) and the World Trade Organiza
tion (WTO) reduced the significance of IS policies, (particularly after the formation of the
AFTA in 1992), it is also the lack of effort to pressure ISI industries to shift to EOI that re
sulted in manufacturing either stagnating or declining in importance in Indonesia,
Malaysia, the Philippines, and Thailand (Figure 24.2).9 Although manufacturing has also
started to fall in Singapore, (p. 696) its almost 20 per cent share in GDP for a city-state
and being characterized by high-value-added activities is positive.
Although EOI became the dominant industrial strategy, all five South East Asian market
economies examined in this chapter have begun undergoing de-industrialization. At its
peak, manufacturing contributed over 30 per cent of GDP in Indonesia, Malaysia (1999‒
2000), and Thailand (2008‒09). The steady expansion of manufacturing in these countries
began to slow down and contract once FDI-led light manufacturing firms could not up
grade to compete with cheaper sites in China, Bangladesh, and Indo-China. Heavy indus
tries, such as iron and steel, cement, and automobiles have remained owing to the prefer
ential tariffs and incentives enjoyed under the ASEAN Economic Community (AEC), which
stipulates a rule-of-origin condition of 40 per cent value to be added cumulatively for
firms to enjoy liberal access to domestic markets.
Figure 24.2 Manufacturing value added share in

GDP, selected economies, 1960–2017
Source: World Bank (2019).
As liberalization and incentives propelled exports, the trade structure of the five South
East Asian market economies quickly shifted towards manufactured goods (Figure 24.3),
particularly electricals and electronics, automotive components, and textiles and gar
Page 14 of 35
ments. Except for declines caused by external crises, manufactured export shares in the
total exports of Indonesia, Malaysia, the Philippines, and Singapore reached their peaks
in 1998‒2000. Thailand enjoyed its highest export share in 2015. The rapid expansion of
manufacturing, including diversification within the sector, helped the South East Asian
economies avert problems caused by the ‘fallacy of composition’.
Figure 24.3 Manufactured exports share in total

merchandise exports, selected economies, 1962–
2017
Source: Computed from data collected from World

Bank (2019).
Owing to problems of data availability and reliability, in evaluating the perfor

(p. 697)
mance of manufacturing among the five economies we focus on the most common indus
tries: machinery; electricals and electronics; transport vehicles and equipment; food, bev
erages and tobacco; and chemical industries. We have used the trade balance with South
Korea and Taiwan of the key export-oriented industry of electricals/electronics (HS85)
and the key import-substituting industry of motor vehicles (HS87) over the period 2000 to
2018 as the benchmark to examine the competitiveness of these industries.
Using trade balance (TB) as the indicator of competitiveness is not without its
problems.10 Although export volume and TB are used to measure the competitiveness of
critical high-technology electric/electronics, and motor vehicles and components indus
tries in the four countries,11 as value added in gross output, the figures are often distort
ed by both tariff and non-tariff barriers. Non-tariff barriers are still important in the mo
tor vehicles industry, although Thailand has largely phased them out since the late 1990s
in its efforts to turn the country into a regional platform for motor vehicle manufacturing.
Singapore led initially on electrical and electronics (EE) exports, a massive proportion of
which were re-exports. In fact, Singapore exported more EE goods than South Korea until
2012 (Figure 24.4). South Korea and Taiwan overtook Singapore in 2012 and 2015 re
spectively. The three countries enjoyed their highest exports of US$185 billion, US$144
billion, and US$129 billion, respectively, in 2018. Exports from Malaysia followed, peak
ing at US$83 billion in 2018. Exports from Thailand and the Philippines amounted to
Page 15 of 35
US$35 billion and 33 billion respectively in 2018, while Indonesia exported US$9 billion
in 2018. Most South Korean and Taiwanese exports are generated by their domestic
firms, while those of the South East Asian market economies are generated by foreign
firms, which often has negative BoP implications once profits are repatriated away from
the host countries. Consequently, firms from these countries have better control over
global value chains.
Figure 24.4 Electric and electronics exports, select

ed economies, 2001–18
Source: Computed from data gathered from WTO

(2019).
South Korea enjoyed a clear lead in the export of motor vehicles and parts, reaching a
peak of US$73 billion in 2014 before falling to US$61 billion in 2018 (Figure 24.5). Ex
cept for 2007‒08 when external demand fell owing to the global financial crisis, motor ve
hicle and parts exports have risen steadily from the countries illustrated. Thailand over
took Taiwan in 2004 to become the second-largest exporter of motor vehicles, with ex
ports reaching US$30 billion in 2018. Exports from Taiwan, Indonesia, Singapore,
Malaysia, and the Philippines followed, reaching US$10 billion, US$8 billion, US$4 bil
lion, US$2 billion, and US$1 billion respectively in 2018. Again, most South Korean and
Taiwanese exports are generated by domestic firms, and (p. 698) (p. 699) those of the
South East Asian market economies are generated by foreign firms, with negative BoP
implications. A significant proportion of motor vehicle and parts exports from South Ko
rea originate from lead national firms, including from their subsidiaries abroad, while
there are only two national car assemblers in Malaysia (both heavily reliant on foreign
technology) and none in the remaining South East Asian market economies. It therefore
appears that firms from South Korea have better control over their global value chains.
Page 16 of 35
Figure 24.5 Motor vehicle exports, selected

economies, 2001–18
Source: Computed from data gathered from WTO

(2019).
The least export-oriented economy among the five countries until the turn of the millenni
um, Indonesia, enjoyed the highest TB in the EE industry during 2001‒06. However, fol
lowing liberalization, Indonesia’s TB fell to –0.42 in 2018 (Figure 24.6). In the same year
Thailand had the second-lowest TB, and Taiwan and South Korea the highest (at 0.37 and
0.31). Malaysia, Singapore, and the Philippines also enjoyed a positive TB, but owing to
the dominance of foreign capital in the ownership of the EE industry much of the profits
are likely to have been repatriated abroad.
Figure 24.6 Trade balance, electric and electronics,

selected economies, 2001–18
Note: X – exports; M – imports.
Source: Computed from data collected from WTO

(2019).
Page 17 of 35
Thailand was the only country to enjoy a positive TB in the motor vehicles and parts in
dustry in 2001‒18 (Figure 24.7). Indonesia and South Korea had a positive TB in this sec
tor in some years, but Taiwan, Malaysia, and the Philippines consistently faced negative
TB over the period. Although the automotive industry cluster generated 850,000 jobs and
10 per cent of GDP in 2017, the foreign-dominated industry still specialized in low-margin
vehicles that paid low salaries and generated little savings for the national economy ow
ing to repatriation of profits abroad (Intarakumnerd and Chaoroenporn, 2013; Maikaew,
2019).
Figure 24.7 Trade balance, motor vehicles, selected

economies, 2001–18
Notes: X – exports; M – imports.
Source: Computed from data collected from WTO

(2019).
We next discuss the key industries by country to assess how they have fared, in
(p. 700)
cluding additionally the outward-oriented industry of textiles and garments, and the in
ward-oriented industries of food, beverages and tobacco, and chemicals. All real growth
rates were computed using local currencies and national base years.
24.5.1 Indonesia
In 1969, as Indonesia was emerging from a political crisis, manufacturing accounted for
only 4.5 per cent of GDP. By 1979 it had risen to 10.6 per cent (Figure 24.2). The 1973‒75
and 1979‒80 oil crises resulted in booming exports and with that an appreciation of the
rupiah as oil prices soared. A flood of cheap imports saw the food, beverages, and tobac
co industry contracting, with growth of –7.0 per cent. Textiles and garments, and the
chemical industry grew on average by 12.4 and 6.6 per cent, respectively, per annum.
Nevertheless, government-protected heavy industries grew rapidly to record the highest
annual average growth rate of 19.3 per cent during 1970‒80. Thus, the low base and the
expansion in heavy industries caused manufacturing value added to grow by 12.5 per
cent on average per annum over this period.
Page 18 of 35
Despite falling oil prices in the 1980s, protected IS industries (including downstream
wood-based industries) drove manufacturing value added to grow by 9.9 per cent per an
num on average between 1980 and 1990. The annual average growth rate of machinery,
electrical and electronics, transport vehicles and equipment (24.9 per (p. 701) cent), food,
beverage, and tobacco (21.8 per cent), and chemicals (20.6 per cent) was the highest in
1980‒90. The share of manufacturing in GDP continued to expand in trend terms to reach
a peak of 32.0 per cent of GDP in 2002, despite a fall in 1997‒2000 following the 1997‒
98 Asian financial crisis. Manufacturing value added grew by 6.5 per cent on average per
annum in 1990‒2000. Machinery, electrical and electronics, and transport vehicles and
equipment recorded the highest growth rate of 15.7 per cent on average per annum in
the same period. Electronics exports soared as the liberal environment in Batam and Bin
tan attracted massive inflows of low-end manufacturing.
Indonesia began to de-industrialize in 2002, with the share of manufacturing in GDP

falling in trend terms from 32.0 per cent in 2002 to 20.5 per cent in 2017 (Figure 24.2).
Manufacturing value added grew on average by 4.5 per cent per annum in 2000‒10. The
restructuring of incentives and deregulation of tariffs from 2000 slowed down the growth
of machinery, electricals and electronics, and transport vehicles and equipment to an av
erage of 5.6 per cent per annum, while food, beverages and tobacco, and the chemical in
dustry grew on average by 7.9 and 7.6 per cent per annum in 2000‒10. The end of the
Multi-Fibre Arrangement (MFA) in 2004 negatively affected the textile and garment in
dustry, which recorded an average annual growth rate of –0.7 per cent. The advent of lib
eral policies shifted the focus to the food, beverages, and tobacco industry, which grew on
average by 9.1 per cent per annum in 2010‒16. The rapid scaling down of protection has
resulted in machinery, electrical and electronics, and transport vehicles and equipment
industries growing by –0.1, and the chemical industry by 3.9 per cent on average per an
num in 2010–16. The textile and garment industry recorded an annual average growth
rate of 6.7 per cent in 2010‒16 as new locations outside Java Island became attractive for
low-wage operations.
24.5.2 Malaysia
The manufacturing sector increased its share in GDP in trend terms from 1963 till 2000
(Figure 24.2). The early increase, from 1963 till 1970, was largely due to a decline in oth
er sectors, as the Malaysian economy was facing a deflationary environment caused by a
fall in tin and rubber prices and ethnic riots in 1969 (Rasiah and Salih, 2019). The low
starting base, as well as massive relocation of labour-intensive export-oriented TNCs, ex
panded the manufacturing sector, which grew by 11.6 per cent on average per annum in
1970‒80. The highest average annual growth of 19.8 per cent was recorded in the textile
and garment industry, followed at 15.2 per cent by machinery and transport vehicles and
equipment (which also comprises electrical and electronics equipment). It was in the
1970s that the giant polyglot company Toray relocated its massive textile manufacturing
operations to Penang. Intel, Advanced Micro Devices, National Semiconductor, Texas In
Page 19 of 35
struments, Hewlett Packard, and Hitachi also relocated their assembly operations in
Malaysia in the 1970s (Rasiah, 1993).
The manufacturing sector grew little in the period 1980‒86 owing to a shift in in
(p. 702)
centives to ISI heavy industries and a cyclical downturn in the electronics industry. How
ever, it grew again from the late 1980s with the renewal of incentives for EOI firms and
massive inflows of FDI into the sector (Rasiah, 1995). Consequently, manufacturing grew
by 9.8 per cent on average per annum in 1980‒90. The highest growth rate of 17.6 per
cent was recorded by the chemical industry, followed by the machinery, and transport ve
hicles and equipment industry (12.2 per cent).
The manufacturing sector grew by 9.9 per cent on average per annum in 1990‒2000.
While food, beverages and tobacco (4.6 per cent), chemicals (6.6 per cent), and textiles
and garments (4.9 per cent) began to slow down, machinery, and transport vehicles and
equipment recorded its highest annual average growth rate of 26.5 per cent in 1990‒
2000. The export-oriented electrical and electronics industry was the biggest contributor
to this growth, which resulted in machinery, EE, and transport vehicles and equipment
accounting for a 37.8 per cent share of manufacturing in 2000, before falling to 31.0 per
cent in 2016.
Deindustrialization began to creep into Malaysia from 2000 as the share of manufactur
ing value added in GDP began to contract from 30.9 per cent in 2000 to 22.3 per cent in
2016 (Figure 24.2). The manufacturing sector recorded an annual average growth rate of
3.6 per cent in 2000‒10. Labour shortages, rising wages, and the emergence of cheaper
sites abroad saw textiles and garments contract with a growth rate of –3.7 per cent. Food,
beverages and tobacco (7.9 per cent), and chemicals (7.6 per cent) recorded the highest
average annual growth rates. Machinery, electrical and electronics, and transport vehi
cles and equipment (0.5 per cent) experienced their lowest growth.
Deindustrialization continued in 2010‒16 as manufacturing recorded its slowest growth

of 2.8 per cent on average per annum. Machinery, electrical, electronics, and transport
vehicles and equipment grew by 3.9 per cent per annum on average, food, beverages, and
tobacco by 3.3 per cent, textiles and garments by 2.8 per cent, and chemicals by 1.3 per
cent.
The Philippines was the earliest to industrialize rapidly in South East Asia. Its manufac
turing share in GDP rose from 24.6 per cent in 1960 to its peak of 26.6 per cent in 1973
(Figure 24.2). However, manufacturing’s share of GDP has largely stagnated through the
period 1960‒2016. Between 1970 and 1980, manufacturing value added grew on average
by 6.1 per cent per annum as textiles and garments, machinery, electricals, electronics,
transport vehicles and equipment, and the chemical industries grew by 10.7, 11.0, and
7.2 per cent respectively. De-industrialization began to grip the Philippines from the
1980s, with average annual manufacturing value added growing by only 0.9 per cent per
annum in 1980‒90. Machinery, electricals and electronics, and (p. 703) transport vehicles
Page 20 of 35
and equipment (–4.7 per cent), the chemical industry (–0.9 per cent), and textiles and gar
ments (–0.9 per cent) contracted.
Deregulation of tariffs and the provision of export-oriented financial incentives stimulated

rapid FDI-driven growth in the machinery, electricals and electronics, and transport vehi
cle and equipment industry, which grew by 22.5 per cent on average per annum from
1990 to 2000. However, a contraction in textiles and garments (–1.9 per cent), and food,
beverages, and tobacco (–0.7 per cent), and a slowdown (0.8 per cent) in the chemical in
dustry slowed manufacturing value-added growth to an annual average of 2.6 per cent.
Manufacturing growth began to pick up in 2000‒10 and 2010‒16, growing on average by
3.7 per cent and 6.9 per cent respectively. The deregulation process began in 1992 with
the introduction of AFTA, and the subsequent formation of the ASEAN Economic Commu
nity (AEC) in 2015 saw the rationalization of tariff measures in South East Asia, resulting
in food, beverages, and tobacco, and textiles and garments growing on average by 14.9
per cent and 10.6 per cent per annum in 2010‒16. However, the trend towards de-indus
trialization continued, with manufacturing’s share in GDP falling to 19.6 per cent in 2016
(Figure 24.2).
24.5.4 Singapore
After launching EOI in 1965, Singapore enjoyed first-mover advantages in Asia, with
manufacturing’s share of GDP reaching 27.0 per cent in 1980 (Figure 24.2). The peak of
28.2 per cent was reached in 2004. While labour-market saturation offered the govern
ment the opportunity to promote skills development and higher value-added activities,
rising wages and labour shortages discouraged further FDI inflows. Manufacturing value
added grew by 13.7 per cent per annum in 1970‒80. Machinery, electricals and electron
ics, and transport vehicles and equipment, textiles and garments, and chemicals grew by
16.1 per cent; textiles and garments by 14.2, and chemicals by 14.7 per cent on average
per annum.
The saturation in the labour market, the floating of the Singapore dollar in 1985, and the
withdrawal of the GSP in 1988 did not seriously affect the manufacturing sector, which
grew by 6.7 per cent on average per annum in 1980‒90 following restructuring of incen
tives and a focus on the STI infrastructure. Further upgrading initiatives and careful
macroeconomic management sustained manufacturing value-added growth of 8.4 per
cent on average per annum in 1990‒2000. The machinery, electricals, electronics, and
transport vehicle and equipment (especially electronics and ships), and chemical (espe
cially petrochemicals), industries grew on average by 15.4 and 12.3 per cent respectively.
Although de-industrialization began to take root after manufacturing’s share in GDP

peaked at 28.2 per cent in 2004, with the 2018 share at 19.6 per cent, Singapore’s manu
facturing sector has remained strong despite the country’s small size. The country’s com
petitive advantage regionally within ASEAN has ensured that the (p. 704) chemical, food,
beverages, and tobacco industries, machinery, electricals, and electronics industries, and
transport vehicles and equipment industries enjoyed annual average growth rates of 11.3,
Page 21 of 35
11.1, and 6.7 per cent respectively in 2000‒10. The textile and garment industry con
tracted (–6.6 per cent) in this period. Consequently, manufacturing value added grew on
average by 6.6 per cent per annum. The food, beverage, and tobacco, and chemical indus
tries grew further by 10.8 and 8.1 per cent respectively while the textiles and garments
industry, and the machinery, electricals, electronics, and transport vehicles and equip
ment sector slowed down to grow by only 3.4 and 2.2 per cent respectively in 2010‒16.
Annual average manufacturing value added grew by 4.6 per cent in 2010‒16.
24.5.5 Thailand
Burgeoning BoP deficits following the 1973‒75 and 1979‒80 oil crises drove the govern
ment to promote exports. The share of manufacturing in GDP rose from 12.5 per cent in
1960 to 30.7 per cent in 2007‒08 before falling in trend terms to 27.4 per cent in 2016
(Figure 24.2). Manufacturing value added grew on average by 7.9 per cent per annum in
1973‒81. The shift to EOI raised manufacturing value-added growth as the machinery,
electrical and electronics, transport vehicles and equipment, textile and garment indus
tries grew on average by 36.0 per cent and 14.0 per cent respectively per annum follow
ing the relocation of TNCs from abroad in 1981‒90. Consequently, manufacturing value
added grew by 11.3 per cent per annum despite the food, beverages and tobacco industry
contracting (–8.7 per cent) in this period.
The food, beverages, and tobacco industry grew by 12.9 per cent, chemicals by 14.6 per
cent, and machinery, electrical and electronics, and transport vehicles and equipment by
10.7 per cent on average per annum in 2000‒10. However, manufacturing value added
only grew by 5.8 per cent per annum on average in 2000‒10 because of a decline in in
ward-oriented industries, such as iron and steel and metals in the first half of the 1990s
as a strong baht attracted massive imports. The machinery, electricals and electronics,
and transport vehicles and equipment industry grew by 6.2 per cent on average per an
num in 2000‒12 following Thailand’s emergence as South East Asia’s regional production
base for cars. Food, beverages, and tobacco, and chemicals grew on average by 5.8 and
4.0 per cent respectively. However, manufacturing value-added growth slowed to 4.7 per
cent on average in 2000‒12 owing to a contraction (–0.4 per cent) in the textile and gar
ment industries.
24.5.6 Summary
In short, after a long period of expansion since particularly the 1970s, the South East
Asian market economies has all faced de-industrialization. Export-orientation through
TNC operations have been the prime propellant of industrialization in these countries.
(p. 705) However, only Singapore has so far avoided premature de-industrialization
through strong industrial upgrading. Malaysia, Indonesia, the Philippines, and Thailand
are experiencing premature deindustrialization as these countries have not managed to
achieve significant industrial upgrading.
Page 22 of 35
24.6 Technological Upgrading Initiatives

With the exception of Singapore, the South East Asian market economies have under-per
formed compared to South Korea and Taiwan, especially in the two critical high-technolo
gy industries of electrical/electronics, and motor vehicles and parts. Although all the
South East Asian market economies eventually switched to EOI, not only has institutional
support to stimulate technological upgrading varied, but Indonesia, Malaysia, the Philip
pines, and Thailand have not managed to transform their manufacturing industries to be
come global leaders. While South Korea and Taiwan expertly used export quotas to quick
ly expose IS industries to external competition (a glaring omission in Indonesia, Malaysia,
the Philippines, and Thailand), the underfunded STI infrastructure and its weak gover
nance in the latter four countries undermined the capacity of their governments to fi
nance industrial upgrading. Whereas Singapore and Malaysia began to stimulate industri
al upgrading on a national scale in 1979 and 1991 respectively, Indonesia, the Philip
pines, and Thailand only launched broad-based STI development policies in 2005. Also,
whereas Singapore’s mission-oriented and strategic industrial policy drove rapid techno
logical upgrading, a lack of appraisal mechanisms and connectivity and coordination be
tween STI infrastructure organizations, and institutions and firms restricted technologi
cal upgrading in the other countries. Ad hoc incremental innovation from knowledge in
flows from abroad has been critical to supporting infrastructure and maintenance, min
ing, and agriculture in all these countries. This section explains the link between STI in
frastructure and industrial upgrading in the five economies.
24.6.1 Indonesia
Strategic policies targeted heavy industries in Indonesia, for example, for the aircraft in
dustry from 1976 until it was discontinued following the 1997–98 Asian financial crisis.
The first formal STI policy in Indonesia can be traced to the 2005 launch of the Vision and
Mission of Indonesian Science and Technology Statement which incorporated four-year
plans to be rolled out until 2025 by the National Research Council of Indonesia (Aminul
lah, 2015; UNESCO, 2015; OECD, 2013). The first two plans (2005‒09 and 2010‒14) ad
dressed the need to support business R&D in strategic (p. 706) sectors. However, R&D ex
penditure did not increase significantly, and hence much of Indonesia’s STI capacity has
remained in public organizations, with only 1 per cent of the budget allocated to the Insti
tute of Sciences by the National Research Council of Indonesia to spearhead R&D activi
ties (Oey-Gardiner, 2011; see also Figure 24.8).
The focus has been on resource-based industries with steel, shipbuilding, palm oil, and
coal identified for Sumatra; food and beverages, textiles, transport equipment, shipping,
information communication technology (ICT), and defence identified for Java; steel, baux
ite, palm oil, coal, oil, gas, and timber strategized for Kalimatan; nickel, food and agricul
ture (including cocoa), oil, gas, and fisheries specified for Sulawesi; tourism, animal hus
bandry, and fisheries classified for Bali and Nussa Tengarra (Lesser Sunda Islands); and
nickel, copper, agriculture, oil and gas, and fisheries targeted for Papua and Maluku Is
lands (UNESCO, 2015). By 2015, the government had only committed 10 per cent of the
Page 23 of 35
US$300 million for infrastructure development (Rasiah, 2018: 286). In addition, the role
played by the private sector is modest, and the ratio of GERD to GDP (0.08 per cent) was
negligible in 2009 (Aminullah, 2015), reaching about 0.2 per cent in 2016.12
24.6.2 Malaysia
In the 1980s, with increased demand for proximate sourcing, technology diffusion
through TNCs appeared promising (Rasiah, 1989). Indeed, significant supplies of (p. 707)
precision tools, semi-automated machinery, and fabrication opportunities were estab
lished between multinational and national firms in Penang in the 1980s and 1990s. The
Action Plan for Industrial Technology Development was launched in 1991 to stimulate in
dustrial upgrading. Several organizations were introduced to solve collective action prob
lems and to promote the innovation essential to make Malaysia a developed country by
2020 (Malaysia, 1991). The government then launched the second STI policy (NSTP2)
(2002–10) and the third NSTP (2013‒20) to stimulate technological upgrading.
In the 1990s the government launched the Human Resources Development Council
(HRDC), the Malaysian Technology Development Corporation, the Multimedia Develop
ment Corporation, the Malaysia Industry–Government Group for High Technology, and
the Multimedia Super Corridor (MSC) to support structural transformation of industry
from low to high value-added activities. The HRDC collects 2 per cent of the payroll from
firms with 50 or more employees, which the firms can only reclaim through approved
training expenditure. Collaborative Research in Engineering, Science and Technology
(CREST) was formed in 2012 to strengthen R&D collaboration between universities, gov
ernment, and industry.13 R&D expenditure in GDP also increased from 2009 (Figure
24.8).
Several R&D funds have been introduced since the 1990s: Commercialization of R&D
(1996); Technology Acquisition (1996); Biotechnology Acquisition (2006); Biotechnology
Commercialization (2006); Industrial Technical Assistance (1990); the Techno (2006); the
E-Content (2006); the Demonstrator Application Grant Scheme (2006); the MSC Malaysia
R&D Grant Scheme (1997); the Science Fund (2006); the Agro-Biotechnology R&D Initia
tive (2006); the Genome and Molecular Biology R&D Initiative (2006); the Pharmaceutical
and Nutraceutical R&D Initiative (2006); the Fundamental Research Grant Scheme
(2006); the Long-Run Research Grant Scheme (2009); and High Impact Research (2009).
However, despite the long-standing role of Malaysia’s government in funding R&D pro
grammes, there is no systematic appraisal and monitoring mechanism (Rasiah, 2018).
The Philippines’ low GDP growth rates are largely a consequence of the country’s special
ization in low-value-added activities. The Philippine Development Plan 2011–16 launched
strategies for using STI to boost productivity and competitiveness in agriculture and
small businesses (including in manufacturing), especially in geographical areas dominat
ed by poor, vulnerable, and marginalized residents. The Harmonized Agenda for Science
Page 24 of 35
and Technology 2002‒20 sought to promote the establishment of five government-funded

centres of excellence by 2020 in (p. 708) biotechnology, nanotechnology, genomics, semi
conductors, and electronics design. The Philippines Genome Centre, which operates two
core facilities for DNA sequencing and bioinformatics, was built at the University of the
Philippines Diliman. The Advanced Device and Materials Testing Laboratory, located at
the Department of Science and Technology’s compound in Bicutan in Taguig City, started
operations in 2013 with three laboratories in surface, thermal, chemical, and metallurgi
cal analysis. The Electronics Product Development Centre is due to be built at the Depart
ment of Science and Technology in Bicutan, and will provide cutting-edge design, proto
typing, and testing facilities for printed circuit boards.
The government supported an expansion in the number of graduates, including doctoral

graduates, between 2009 and 2013. Hence, while the Philippines only invested 0.3 per
cent of GDP in higher education in 2013, tertiary enrolment experienced a rise from 2.6
million in 2009 to 3.2 million in 2013. The number of doctoral graduates more than dou
bled from 1,622 in 2009 to 3,305 in 2013 (UNESCO, 2015). However, not only has GERD
in GDP remained low (Figure 24.8), but neither the research undertaken in universities
nor the human capital produced by them have participated actively in the commercializa
tion of manufactured products in the Philippines.
24.6.4 Singapore
Through the EDB, Singapore has systematically stimulated technological upgrading in the
country, leveraging its world-class infrastructure, efficient civil service, and provision of
incentives and grants in return for continuous technological upgrading by foreign TNCs.
Science parks and R&D have been supported through the development of STI infrastruc
ture to finance strategic technologies in knowledge-based industries. By 2015, the two
science parks that acted as R&D hubs for companies housed more than 350 organizations
and companies.
Singapore’s 2014 GERD-to-GDP ratio of 2.3 per cent was far below the 4.5 per cent and
4.1 per cent enjoyed by Israel and Korea respectively (UNESCO, 2015; Figure 24.8). The
government has invested heavily in the development of science and technology at
Singapore’s leading universities—the National University of Singapore (NUS) and
Nanyang Technical University—with cutting-edge research facilities, including R&D labs,
machinery, and equipment, and has opened employment in the country to world-class sci
entists and engineers. Biopolis opened in 2003 to promote biomedical research, while Fu
sionopolis was established in 2008 to promote ICT research. The National Framework for
Innovation and Enterprise enjoyed a total allocation of SG$4.4 billion during 2008–12.
Singapore has emerged as a leader of incremental innovation activity in South East Asia
owing to its world-class basic infrastructure, integration in global markets, and connectiv
ity and coordination between firms and organizations. Laboratories at NUS have been a
major platform for petrochemical innovations, which support over a hundred firms in Sin
gapore (Singapore, 2019a; ASMI, 2019). Similar support is also enjoyed by the shipbuild
Page 25 of 35
ing (p. 709) industry, which includes strong research linkages between firms, and labs in
NUS and Nanyang University (Singapore, 2019b).
24.6.5 Thailand
Since the 1990s, the Thai government has promoted technology diffusion and innovation,
starting with the National Science and Technology Development Agency, which estab
lished the Industrial Consultancy Services in 1992 to promote alliances between local and
foreign technical consultants. The agency launched a Software Park to stimulate innova
tion in start-up firms. The BOI also developed the Unit for Industrial Linkage Develop
ment programme to strengthen linkages among small and medium-sized manufacturers
(UNCTAD, 2005). However, the lack of human capital, the absence of R&D grants to stim
ulate design and R&D, and the lack of electronics-based research in universities and oth
er laboratories drove American chip manufacturing out of Thailand from the 1980s (Rasi
ah, 2009). Nevertheless, designing of ICs related to automotive systems has emerged
with collaborative links with the universities of Chulalongkorn, Mongkut, and Chiang Mai
(Intarakumnerd et al., 2015). The founding of the Hard Disk Drive (HDD) Institute helped
provide scientific infrastructure for the HDD industry by establishing a central laboratory
and networks of government laboratories. National firms, such as Hana Microelectronics,
Stars Microelectronics Thailand, and Silicon Craft Technology, began designing cus
tomized integrated circuit packaging (Intarakumnerd et al., 2015).
While Thailand is South East Asia’s leading producer of disk drives and automobiles, a
transition to higher value-added activities would require strong R&D funding, which,
however, fell from 1.1 per cent of GDP in 2002 to 0.7 per cent in 2012 (Figure 24.8). Nev
ertheless, the National Science and Technology Development Agency has become an an
chor for stimulating R&D, which in 2015 employed over 7 per cent of the country’s full-
time researchers in four organizations: the National Centre for Genetic Engineering and
Biotechnology; the National Electronics and Computer Technology Centre; the National
Metal and Materials Technology Centre; and the National Nanotechnology Centre. How
ever, although it is part of the ten-year National Science and Technology Action Policy
(NSTAP), 2004–13, targeted at developing the STI infrastructure, it has made little
progress. The 2012–21 NSTAP attempted to focus on infrastructure development, capaci
ty building, regional science parks, industrial technology assistance, and tax incentives
for R&D. While it set a target of 1 per cent GERD of GDP by 2021, it still only offered tax
exemptions for R&D without any grants up front.
Page 26 of 35
Figure 24.8 GERD in GDP, South East Asian market

economies, 1996–2014 (%)
Source: UNESCO (2015).
Except for the Asian financial crisis period of 1997‒98, all the South East Asian market
economies showed an intellectual property right (IPR) TB inferior to that of South Korea
from 1981 to 2018. Singapore has managed to close the gap with South Korea since
2013, while Malaysia has improved since 2015 (Figure 24.9). Although there were im
provements in some years, Indonesia, the Philippines, and Thailand have remained
strongly dependent on foreign intellectual property. The IPR TB figures for (p. 710)
Malaysia, Indonesia, Thailand, and the Philippines in 2018 were −0.73, −0.94, −0.96, and
−0.96, respectively. This hugely negative coefficient partly explains why firms in these
countries have not been able to upgrade into higher value-added activities to enjoy posi
tive de-industrialization.
Figure 24.9 Intellectual property trade balance, se

lected economies, 1981–2018
Source: World Bank (2016).
Page 27 of 35
24.6.6 Summary
Overall, except for Singapore, the STI infrastructure of the South East Asian market
economies have serious shortcomings. A combination of low R&D funding, especially in
Indonesia, the Philippines, and Thailand, and poor governance instruments in Malaysia,
has undermined catching up and leapfrogging in these countries. Consequently, Malaysia,
the Philippines, Indonesia, and Thailand have no cutting-edge manufacturing operations.
This stark reality is reflected in heavy reliance on foreign technology as the IPRs of these
countries are well below those of South Korea and Singapore (Figure 24.9). Despite not
having national manufacturing firms at the technology frontier, Singapore has managed
to continuously renew its manufacturing structure through ingeniously leveraging on its
endowments, and attracting TNCs and foreign researchers to support high value-added
manufacturing activities. Singapore has also managed to close the IPR trade gap with
South Korea since 2015, following grants offered to biotechnology labs and firms.
(p. 711) 24.7 Conclusions

Manufacturing has played a major economic role in raising GDP growth, reducing unem
ployment, and alleviating poverty in the South East Asian market economies, especially
since the 1960s. Although ISI began in Indonesia, Malaysia, the Philippines, Singapore,
and Thailand in the 1950s, the lack of performance standards and initiatives to pressure
relevant firms to export undermined the competitiveness of import-replacing industries.
Ambitious heavy industry promotion in Indonesia, Malaysia, and the Philippines also
largely failed. As IS industries drained the economies, BoP, debt service, and rising unem
ployment problems drove the South East Asian market economies to introduce EOI. For
eign direct investment arrived in the form of giant TNCs from developed countries relo
cating low-end stages of production to stimulate exports from all these economies. In
deed, FDI has been the main engine of export manufacturing growth in all the South East
Asian market economies.
Indonesia, Malaysia, Thailand, and the Philippines have all experienced premature dein
dustrialization, with the value added in manufacturing gross output falling before reach
ing developed status. Singapore has managed to become developed, focusing on incre
mental innovation by continuously renewing its composition of cutting-edge industries us
ing a leveraging strategy that offers incentives to attract leading foreign firms. In addi
tion to shipbuilding and petrochemical industries, Singapore has also evolved a strong
bio-pharma industry. Malaysia began to develop its STI infrastructure in 1991 but a lack
of stringent appraisal mechanisms in the selection and monitoring of firms for support
caused a massive dissipation of rents. Ethnically coloured policies also prevented indus
tries from attracting world-class diaspora members to return and lead both national firms
and the organizations created to solve collective action problems (Rasiah, 2011).
Whereas the South East Asian market economies largely prevented the twin troubles of
‘fallacy of composition’ and ‘Dutch Disease’ from dragging them to the debt precipice, the
Page 28 of 35
lack of effective policy coordination to avert such problems cost Indonesia, Malaysia, the
Philippines, and Thailand considerable resources. Singapore not only managed to prevent
these problems from becoming chronic, but its sophisticated petrochemical, biotechnolo
gy, and shipbuilding industries have evolved to become world leaders. The lack of scale,
and related research in its universities, have contributed to Singapore reaching its criti
cal limits in electronics manufacturing. However, the remaining four market economies
have populations and land space exceeding that of South Korea and Taiwan, and there
fore the contraction of the manufacturing sector since the turn of the millennium sug
gests that they are facing premature de-industrialization. While Singapore’s textile and
clothing, and electricals and electronics, and vehicle manufacturing sectors have de
clined, the ‘mission-oriented state’ (Mazzucato, 2013) has successfully removed incen
tives from them, rapidly moving them away from the country and gradually replacing
them with new compatible (p. 712) industries, such as biotechnology, while retaining the
shipbuilding and petrochemical industries. Indonesia, Malaysia, the Philippines, and Thai
land have not managed to create an integrated STI infrastructure that can stimulate
transformation of the manufacturing sector to the technology frontier as South Korea and
Taiwan have done. Singapore, however, has managed to stay ahead of these countries
with a sophisticated STI infrastructure, and using a flexible strategy that has continuous
ly renewed the industries required to support a rise in value added.
References
Abramovitz, Moses (1956) ‘Resource and Output Trends in the United States since 1870’,
American Economic Review 46(2): 5–23.
Akrasanee, Narongchai (1979) Economic Development of Thailand and ASEAN Economic

Cooperation. Canberra: Australian University Press.
Alavi, Rokiah (1996) Industrialization in Malaysia: Import Substitution and Infant Indus
try Performance. London: Routledge.
Aminullah, Erman (2015) ‘Learning, R&D Intensity and Economic Prosperity in Low R&D
Countries (LRDCs): Envisioning the Indonesian Future’. Keynote speech presented at the
12th International ASIALICS conference, Jogjakarta, 15–17 September.
Amsden, Alice (1993) ‘Diffusion of Development: The Late-Industrializing Model and

Greater East Asia’, American Economic Review 81(2): 282–6.
Anwar, Arsjad (1979) Trade Strategies and Industrial Development in Indonesia. Canber
ra: Australian National University Press.
ASMI (2019) ‘Singapore Marine Industry Milestones’, Association of Singapore Marine

Industries (ASMI). Available at http://www.asmi.com/index.cfm?GPID=237.
Page 29 of 35
Bautista, Romeo (1983) ‘Industrial Policy and Development in the ASEAN Countries’,
Philippines Institute for Development Studies, Monograph No. 2, Manila.
Best, Michael (2001) The New Competitive Advantage. Oxford: Oxford University Press.
Chang, Ha-Joon (1994) The Political Economy of Industrial Policy. Basingstoke: Palgrave
Macmillan.
Chang, Shan-Fang, Rajah Rasiah, and Wei-Meng Chan (2016) ‘Understanding Innovations
in Malaysia’s Construction Industry: A Study of Four Large National Firms, Asian Journal
of Technology Innovation 24(3): 275–92.
Chenery, Hollis, Sherman Robinson, and Moshe Syrquin (1986) Industrialization and
Growth: A Comparative Study. Washington, DC: World Bank.
Dosi, Giovanni (1988) ‘The Nature of the Innovative Process’, in Giovanni Dosi, Christo
pher Freeman, Richard R. Nelson, Gerald Silverberg, and Luc Soete (eds) Technical
Change and Economic Theory. London: Pinter Publishers, pp. 221–38.
Edwards, Christopher (1975) ‘Protection, Profits and Policy: An Analysis of Industrializa

tion in Malaysia’, PhD thesis, University of East Anglia.
Gerschenkron, Alexander (1952) ‘Economic Backwardness in Historical Perspective’, in

Bert F. Hoselitz (ed.) The Progress of Underdeveloped Areas. Chicago, IL: University of
Chicago Press, pp. 3–29. (p. 713)
Hill, Hal (1996) The Indonesian Economy since 1966: Southeast Asia’s Emerging Giant.
Intarakumnerd, Patarapong and Peera Chaoroenporn (2013) ‘The Roles of Intermediaries

and the Development of their Capabilities in Sectoral Innovation Systems: A Case Study
of Thailand’, Asia Pacific Business Review 21(S2): 99–114.
Intarakumnerd, Patarapong, Pun-Arj Chairatana, and Preeda Chayanajit (2015) ‘Global

Production Networks and Host-Site Industrial Upgrading: The Case of the Semiconductor
Industry in Thailand’, Asia Pacific Business Review 22(2): 289–306.
Johnson, Chalmers (1982) MITI and the Japanese Miracle. Stanford, CA: Stanford Univer
sity Press.
Jomo, Kwame Sundaram and Christopher Edwards (1993) ‘Malaysian Industrialization in

Historical Perspective’, in Kwame Sundaram Jomo (ed) Industrialising Malaysia: Policy,
Performance, Prospects. London: Routledge, pp. 14–39.
University Press.
Kim, Lin-Su (1997) From Imitation to Innovation: The Dynamics of Korea’s Technological
Learning. Cambridge, MA: Harvard Business School Press.
Page 30 of 35
Maikaew, Piyachart (2019) ‘Automotive Industry at a Turning Point: Regional Competition

and the Shift to EVs Will Test Local Competitiveness in Cars’, Bangkok Post, 7 January.
Available at https://www.bangkokpost.com/business/1606570/automotive-industry-
at-a-turning-point.
Malaysia (1965) First Malaysia Plan 1966–1970. Kuala Lumpur: Government Printers.
Malaysia (1971) Second Malaysia Plan 1971–1975. Kuala Lumpur: Government Printers.
Malaysia (1981) Fourth Malaysia Plan 1981–1985. Kuala Lumpur: Government Printers.
Malaysia (1986) Fifth Malaysia Plan 1986–1990. Kuala Lumpur: Government Printers.
Malaysia (1991) Action Plan for Industrial Technology Development. Kuala Lumpur: Gov
ernment Printers.
Menardo, Amelia A. (2004) ‘Tariff Reforms in the Philippines’. Available at https://

www.mofa.go.jp/policy/economy/apec/conference/present0409/philippines.pdf.
OECD (2013) ‘Innovation in Southeast Asia—Vietnam’. Available at http://www.oecd.org/

sti/inno/innovation-in-southeast-asia-2012-9789264128712-en.htm.
Oey-Gardiner, Mayling (2011) ‘Study of the Role of the Indonesian Institute of Science
(LIPI) in Bridging between Research and Development Policy’. Final report submitted to
Insan Hitawasana Sejahtera, Jakarta. Available at https://dfat.gov.au/about-us/publica
tions/Documents/indo-ks5-lipi.pdf.
Ofreneo, Rene (2015) ‘Growth and Employment in De-industrializing Philippines’, Journal

of the Asia Pacific Economy 20(1): 111–29.
Penrose, Edith (1959) The Theory of the Growth of the Firm. New York: John Wiley &
Sons.
Phang, Eng-Fong and Hui-Heng Tan (1981) ‘Employment and Export-led Industrialization:
The Experience of Singapore’, in Rashid Amjad (ed.) The Development of Labour-Inten
sive Industry in ASEAN Countries. Bangkok: International Labour Organization, pp. 141–
73.
Phongpaichit, Pasuk and Chris Baker (1996) Thailand: Economy and Politics. Oxford: Ox
Polanyi, Michael (1966) ‘The Logic of Tacit Inference’, Philosophy 41(155): 1‒18. (p. 714)
Poot, Huib (1981) ‘The Development of Labour-intensive Industries in Indonesia’, in

Rashid Amjad (ed.) The Development of Labour-intensive Industry in ASEAN Countries.
Bangkok: ILO Asian Employment Programme, pp. 25–35.
Page 31 of 35
Prawiro, Radius (1998) Indonesia’s Struggle for Economic Development: Pragmatism in

Action. Oxford: Oxford University Press.
Problems’. Economic Commission for Latin America. New York: United Nations Depart
ment of Economic Affairs.
Rasiah, Rajah (1988) ‘The Semiconductor Industry in Penang: Implications for the New
International Division of Labour Theories’, Journal of Contemporary Asia 18(1): 24–46.
Rasiah, Rajah (1989) ‘Competition and Restructuring in the Semiconductor Industry: Im
plications for Technology Transfer and Absorption in Penang’, Southeast Asian Journal of
Social Sciences 17(2): 41–57.
Rasiah, Rajah (1993) Pembahagian Kerja Antarabangsa: Industri Separa Konduktor di Pu
lau Pinang. Kuala Lumpur: Malaysian Social Science Association.
Rasiah, Rajah (1995) Foreign Capital and Industrialization in Malaysia. Basingstoke:

Macmillan.
Rasiah, Rajah (2009) ‘Expansion and Slowdown in Southeast Asian Electronics Manufac
turing’, Journal of the Asia Pacific Economy 14(2): 123–37.
Rasiah, Rajah (2010) ‘Industrialization in the Second-tier NIEs’, in Rajah Rasiah and Jo
hannes Dragsbaek Schmidt (eds) The New Political Economy of Southeast Asia.
Cheltenham: Edward Elgar, pp. 44–102.
Rasiah, Rajah (2011) Malaysian Economy: Structure and Social Change. Shah Alam: Ox
Rasiah, Rajah (2018) ‘Innovation Policy, Inputs and Output’, in Masahito Ambashi (ed.) In
novation Policy in ASEAN. Jakarta: Economic Research Institute for ASEAN and East
Asia, pp. 277–320.
Rasiah, Rajah and Kamal Salih (eds) (2019) Driving Development: Revisiting Razak’s Role
in Malaysia’s Economic Progress. Kuala Lumpur: University of Malaya Press.
Rasiah, Rajah and Xiao-Shan Yap (2016) ‘Institutional Support, Technological Capabilities
and Domestic Linkages in the Semiconductor Industry in Singapore’, Asia Pacific Busi
ness Review 22(1): 180‒92.
Rasiah, Rajah and Xiao-Shan Yap (2019) ‘How Much of Raymond Vernon’s Product-cycle
Thesis is Still Relevant Today: Evidence from the Integrated Circuits Industry’, Interna
tional Journal of Technological Learning, Innovation and Development 11(1): 56‒77.
London: Anthem Press.
Page 32 of 35
Schumpeter, Joseph (1934) Theory of Economic Development. Cambridge, MA: Harvard

University Press.
Schumpeter, Joseph (1943) Capitalism, Socialism and Democracy. Cambridge, MA: Har
vard University Press.
Singapore (2019a) ‘Energy and Chemicals, Economic Development Board (EDB)’. Avail
able at https://www.edb.gov.sg/en/our-industries/energy-and-chemicals.html.
Singapore (2019b) ‘Key Milestones, Maritime and Port Authority of Singapore (MPA)’.
Available at https://www.mpa.gov.sg/web/portal/home/maritime-companies/re
search-development/milestones-projects/key-milestones. (p. 715)
Singer, Hans (1950) ‘The Distribution of Gains between Investing and Borrowing Coun
tries’, American Economic Review 40(2): 473–85.
Smith, Adam (1776) The Wealth of Nations. London: Strahan and Cadell.
Tambunlertchai Somsak and Chetsada Loohawenchit (1981) ‘Labour Intensive and Small
Scale Manufacturing Industries in Thailand’, in Rashid Amjad (ed.) The Development of
Labour Intensive Industry in ASEAN Countries. Bangkok: ILO Asian Employment Pro
gramme.
Tee, Kian-Wee (2006) ‘Policies for Private-sector Development in Indonesia’. ADB Insti
tute Working Paper No. 46, Tokyo.
UNCTAD (1996) Trade and Development Report. Geneva: United Nations Conference for
Trade and Development.
UNCTAD (2005) ‘Transfer of Technology for Successful Integration into the Global Econo
my: A Case Study of the Electronics Industry in Thailand’. Working Paper No. UNCTAD/
ITE/IPC/2005/6. Geneva: United Nations Conference for Trade and Development. Avail
able at http://unctad.org/en/docs/iteipc20056_en.pdf.
UNESCO (2015) UNESCO Science Report: Towards 2030. Geneva: United Nations Educa
tional, Scientific and Cultural Organization (UNESCO).
Wade, Robert H. (1990) Governing the Market: Economic Theory and the Role of Govern
Wong, Poh-Kang (2001) ‘Leveraging Multinational Corporations, Fostering Technopre

neurship: The Changing Role of S&T Policy in Singapore’, International Journal of Tech
nology Management 22(5/6): 539‒67.
World Bank (2016) World Development Indicators. Washington, DC: World Bank.
World Bank (2019) World Development Indicators. Washington, DC: World Development
Institute.
Page 33 of 35
WTO (2016) ‘WTO Trade Review: Singapore’. Available at http://www.intracen.org/

BB-2016-07-28-WTO-Trade-Policy-Review-Singapore/.
Yap, Xiao-Shan and Rajah Rasiah (2017) Catching Up and Leapfrogging: The New Late
comers in the Integrated Circuits Industry. Abingdon: Routledge.
38(152): 527‒42.
Zhang, Miao, Rajah Rasiah, and John Kean-Yew Lee (2017) ‘Navigating a Highly Protect
ed Market: China’s Chery Automobile in Malaysia’, Journal of Contemporary Asia 47(5):
774–91.
Notes:
(1) Computed using data from World Bank (2019).
(2) We have excluded the transition economies of Cambodia, Lao PDR, Myanmar, Viet
nam, Timor-Leste, and Brunei from South East Asia only because the first five economies
were integrated into the capitalist world economy much later (starting with Vietnam
since 1986) and Brunei has been an oil- and gas-driven economy that has not implement
ed any industrial policies.
(3) Johnson (1982) and Wade (1990) addressed significant elements of industrial policy
implemented in Japan and Taiwan respectively, but did not link it to manufacturing being
a critical engine of growth.
(4) We would argue that the abrupt liberalization since the 1980s, but especially following
the Plaza Accord of 1985, denied the South Korean government the space to insulate the
national economy from private borrowings from abroad and exchange rate volatility.
(5) Singapore’s ISI lasted from 1959 till 1965 (Phang and Tan, 1981).
(6) ‘International’ was added to the Ministry of Trade and Industry in the 1980s (Rasiah,
1995).
(7) See Rasiah and Yap (2016) for a detailed study of Singapore’s inferior location in the
semiconductor technology trajectory.
(8) Singapore’s low corporate tax base of 14 per cent in 2018, along with the FTAs, make
it an attractive regional headquarters for TNCs.
(9) All uncited statistics on manufacturing composition and growth rates were computed
from World Bank (2019).
(10) We have avoided using total factor productivity and labour productivity measures to
examine performance owing to serious problems associated with them.
Page 34 of 35
(11) We avoided using revealed comparative advantage (RCA) owing to the lack of reliable
information on value-added share in gross output, which is seriously compromised be
cause of varying degrees of protection in these countries. Also, it does not distinguish
specific industries within the categories based on their technological sophistication.
(12) Communication with Erman Aminullah on 10 October 2016 in Bandung.
(13) Interview conducted by the author on 12 December 2015 in Georgetown.
Rajah Rasiah
Rajah Rasiah currently holds the position of distinguished professor of economics at

the University of Malaya. His previous positions were professor of industrial organi
zation at National University of Malaysia, and senior research fellow at UNU-IN
TECH, Maastricht. He obtained his doctorate in economics from the University of
Cambridge in 1992. He was awarded the Celso Furtado Prize by the World Academy
of Sciences in 2014 for advancing social science thought (economics).
Page 35 of 35
National Champions, Reforms, and Industrial Policy in China
National Champions, Reforms, and Industrial Policy in

China
Chen Li and Muyang Chen
China’s economic reforms since the late 1970s have been characterized by a combination
of market-oriented transition and persistent practices of statist industrial policy. A key di
mension of China’s industrial policy is to transform and consolidate the backbone of its
large state-owned enterprises sector into a group of globally competitive big businesses.
The rise of China’s ‘national champions’, defined here as the giant central state-con
trolled corporations, business groups, and financial institutions in the ‘commanding
heights’ sectors, has reshaped the landscape of China’s business system. This chapter re
views the origin and evolution of China’s national champions industrial policy. It examines
how the Chinese state has restructured the core assets and enterprise units from the so
cialist command economy into a set of large state-controlled corporations and business
groups viable in a predominantly market-oriented environment. This process has been
driven by continuous organizational learning and capability building at both the govern
ment and enterprise levels. The current governance regime of China’s national champi
ons can be characterized as a networked hierarchy that interweaves multiple mecha
nisms of institutional bridging and reciprocal control. Despite their remarkable size and
growth, China’s national champions are still at an early stage of developing their interna
tional competitiveness.
Keywords: national champions, reforms, industrial policy, China, commanding heights, governance, networked hi
erarchy, institutional bridging, reciprocal control
25.1 Introduction
IN the 1980s and 1990s, massive and rapid privatization-cum-liberalization was the core
strategy of industrial reform prescribed by the mainstream transition economics for so
cialist countries. It was argued that ‘the government is in a race against time’ and the
achievement of reforms cannot be maintained unless privatization occurs ‘quickly and on
a vast scale’ (Sachs, 1991: 40‒1). The existing industrial system inherited from the com
munist past was seen as an inefficient fossil to be phased out, and the challenge of re
Page 1 of 47
structuring was primarily ‘to efficiently close much of the old structure and allow for a
rapid expansion of a new one’ (Blanchard et al., 1991: 65). Many argued that transition
economies should not imitate the industrial policies of Japan and South Korea, and in
stead should shift their state‒business relations towards an ‘arm’s-length’ type under
competitive market structures and private ownership.
China’s reforms since the 1990s, however, have been characterized by a combination of
market-oriented transition and the persistent practice of industrial policies. A key dimen
sion of China’s industrial policy has been to transform and consolidate the core of its
large state-owned enterprises sector into a batch of state-controlled big businesses with
international competitiveness. The rise of China’s ‘national champions’, defined here as
the giant central state-controlled corporations, business groups, and financial institutions
in the ‘commanding heights’ sectors, has reshaped the landscape of China’s business sys
tem. Now China has the second-largest number of Fortune Global 500 companies in the
world. In 2018, there were a total of 107 large firms from Mainland China in the Fortune
Global 500 by revenue, among which fifty-eight are central (p. 717) state-controlled na
tional champions (apart from seven large financial institutions; the rest are under the su
pervision of China’s State-owned Assets Supervision and Administration Commission,
hereafter SASAC). Orchestrated by state-led restructuring, each of these national champi
ons occupies a leading position in China’s domestic markets in their respective business
areas and many of them, such as CNOOC, Sinopec, China State Construction Engineering
Corporation, COSCO, CITIC Group, Bank of China and Industrial & Commercial Bank of
China, have started to build significant operations outside China, going increasingly glob
al.
This chapter examines the origin and evolution of China’s national champions industrial
policy in the context of economic reforms. It reviews three key analytical perspectives on
this issue: (1) the mainstream transition economics view; (2) the state capitalism view;
and (3) the late industrialization and developmental state view. It then examines how
state industrial policy has shaped the rise of China’s national champions in both the in
dustrial and financial sectors. It shows that the central plank of China’s national champi
ons industrial policy is to restructure the core assets and enterprise units of the socialist
command economy into a batch of state-controlled corporations, business groups, and fi
nancial institutions viable in a predominantly market-oriented environment. China has
chosen to reinvent the core of its ‘old structure’ by innovatively combining the existing
bureaucratic institutions and resources with new governance forms and practices learnt
from the advanced economies’ capital markets and big businesses. This process has in
volved continuous organizational learning and capability building at both government and
enterprise levels and has led to some unique features of state‒business relations in Chi
na. The current governance regime of China’s national champions can be characterized
as a networked hierarchy that interweaves multiple mechanisms of institutional bridging
and reciprocal control. This chapter concludes by evaluating the position of China’s na
tional champions in the global business system. It shows that despite their remarkable
Page 2 of 47
size and growth, China’s national champions are still at a relatively early stage of devel
oping their international competitiveness.
25.2 Industrial Policy, Big Business, and

China’s Economic Reforms
The period since the early 1990s has witnessed China’s persistent national champions in
dustrial policy consolidating the core of China’s existing industrial ministries and finan
cial system into a batch of large central state-controlled corporations, business groups,
and financial institutions (Nolan, 2001a, 2001b; Sutherland, 2003; Li, 2015). However, un
til recently, this aspect of China’s industrial policy had been relatively overlooked by the
mainstream literature on China’s economic reforms and industrial development. This sec
tion reviews three key analytical perspectives which provide (p. 718) different frameworks
to understand China’s national champions industrial policy and reforms: (1) the main
stream transition economics view; (2) the state capitalism view; (3) the late industrializa
tion and developmental state view. They reflect different perspectives on the roles of in
dustrial policy and big business in economic development and transition.
25.2.1 The Mainstream Transition Economics View
Mainstream transition economics pays limited attention to the roles of oligopolistic big
business in the economic and technical progress of advanced capitalist countries, tending
to assume that countries grow prosperous through free markets and perfect competition.
It interprets the ‘East Asian miracle’ primarily through the lens of market forces, getting
the prices right and following comparative advantage (World Bank, 1993; Stiglitz, 1996;
Nolan, 2001a; Lin et al., 2003). Under Communist planning, China and the former USSR
each had a small number of relatively large plants in many industrial sectors such as iron
and steel, power equipment, aerospace, and defence industries, but neither had real mod
ern multi-plant corporations with substantial vertical integration and diversification like
Exxon Mobil, General Electric, and Siemens (Nolan, 2001a; Shmelev and Popov, 1989;
Popov, 2019; Popov, Chapter 27 in this volume). In advising on the industrial reforms of
transition economies, mainstream economics focuses primarily on the privatization of in
dividual plants and disregards the feasibility and desirability of merging and developing
these plants to produce large, globally competitive multi-plant firms (Nolan, 2001a,
2001b). Consistent with the Washington Consensus, the ‘transition orthodoxy’ dismisses
industrial policy as government-induced distortions that undermine market efficiency and
fairness. National champions are typically viewed as either powerful interest groups influ
encing the government’s policymaking for their own benefit or the state’s instruments for
achieving political goals, such as maintaining control and social stability.
In the mainstream transition economics literature, China’s reforms are often character
ized as ‘the retreat of the state, and the advance of the market’, their dominant logic de
scribed as the gradual establishment of a free and competitive private enterprise system
by changing the government‒business relationship to one of an arm’s-length type (Tian
Page 3 of 47
and Liang, 1999; Qian, 1999, 2000; Qian and Wu, 2000; Lin et al., 2003; Wu, 2005; Stein
feld, 2010). The key to understanding China’s rapid economic growth and industrial de
velopment during the reforms is, it is argued, China’s shift towards an industrial struc
ture that reflects its comparative advantage. Small and medium-sized enterprises (SMEs)
are regarded as the most dynamic part of China’s industrial economy (World Bank, 1997;
Lin et al., 2003). In assessing China’s industrial policies of targeting the perceived ‘pillar
industries’ and supporting selected large state-owned enterprises (SOEs) in the 1990s,
World Bank (1997: 39) argues that ‘it is difficult to say what this policy really means’ and
concludes that ‘it is quite likely they work at cross purposes and in unintended ways that
distort development’. When China (p. 719) implemented the strategy of ‘grasping the large
and letting go of the small’ in the late 1990s and early 2000s, the central government’s
policy focus was on ‘grasping the large’, that is, concentrating resources and policy sup
port on targeted large SOEs through the privatization of local small and medium-sized
SOEs (Wang, 1998; Huang, 2008). However, neo-liberal analyses of China’s industrial re
forms tend to focus exclusively on the ‘letting go of the small’ (Li, 2015). Around China’s
WTO entry in 2001, it was widely predicted that state‒business relations in the country
would converge towards an arm’s-length type compatible with the norms of Western mod
ern market economies and ‘unless privatized, the SOEs have no chance of surviving’ in
market competition (Tian and Liang, 1999: 81; Qian and Wu, 2000). With regard to
China’s state control over large SOEs, Qian (2001) interpreted it as a second-best transi
tional arrangement which would soon be phased out as the market reforms deepen. It
was argued that in the early stage of its development, China’s industrial economy should
be based on small-scale, labour-intensive industries rather than using industrial policy to
support the growth of big businesses in capital-intensive and technology-intensive indus
tries (Lin et al., 2003; Lin and Chang, 2009). China’s industrial policymaking to support
national champions is criticized as a departure from economic rationality and a symptom
of the ‘partial reform traps’ that lead to capital misallocation and rent seeking (Steinfeld,
2004; Huang, 2008).
25.2.2 The State Capitalism View
Recent studies on the rise of China’s ‘state capitalism’ or ‘state-led capitalism’, however,
suggest that there have been fundamental changes in the nature of China’s industrial re
forms since the early 1990s. Some even argue there was a ‘great reversal’ of reforms as
the state-controlled system regained the policy upper hand over the entrepreneurial pri
vate sector (Bremmer, 2009; Huang, 2008, 2011). While earlier studies along these lines
tend to label China as a unitary and yet only vaguely defined model of ‘state capitalism’,
the recent political-economy literature has produced more nuanced perspectives and
many important insights into the common features and sectoral variation in state control
and regulation, corporate governance, ownership patterns, and managerial practices of
China’s national champions (Lin and Milhaupt, 2013; Liebman and Milhaupt, 2015; Mil
haupt, 2017; Milhaupt and Zheng, 2015; Naughton and Tsai, 2015).
Page 4 of 47
The focus of this ‘state capitalism’ literature is not on the aggregate ownership structure
of China’s industrial system, but on the significance of China’s industrial policy, the devel
opment of a robust state sector in combination with a predominantly market-oriented
economy and emerging capital markets, and in particular, the rise of China’s national
champions during recent decades (Lin, 2011; Szamosszegi and Kyle, 2011; Naughton and
Tsai, 2015; Lardy, 2019). The ‘state capitalism’ narratives emphasize that a crucial part of
China’s industrial system has been evolving along a heterodox path defying the main
stream model of neo-liberal transition and institutional (p. 720) convergence (Li, 2015). In
particular, Pearson (2015) argues that China’s economy has developed a hierarchical,
three-tiered structure, with large state-controlled enterprises in strategic ‘commanding
heights’ industries in the top tier. As China’s party-state becomes disengaged from direct
management and ownership of enterprises in non-strategic economic sectors, it has up
graded its interests and capability in developing national champions in strategic indus
tries. Emphasizing the role of China’s central government, Lin (2011) argues that China
has been evolving towards ‘centrally managed capitalism’. There is now also a growing
literature that analyses the central role of the Chinese Communist Party (CCP) in govern
ing China’s national champions (Brødsgaard, 2012; Lin and Milhaupt, 2013; Liebman and
Milhaupt, 2015; Li, 2016, 2019; Leutert, 2018a, 2018b).
25.2.3 The Late Industrialization and Developmental State View
The literature of late industrialization and the developmental state associates the degree
of state intervention in industrialization with the level of backwardness as well as the tim
ing and international environment for latecomers seeking to catch up with advanced in
dustrial economies. Gerschenkron’s (1962) backwardness thesis contends that the more
backward a country is (vis-à-vis early developers), the more state-led its process of indus
trialization tends to be. The rise of the communist party-state and command economy in
the USSR and China can be seen as extreme forms of state intervention to promote late
industrialization. In the 1920s, emulating the organizational structures of modern capital
ist big businesses, the Soviet communist party had grouped major industrial enterprise
units into a number of ‘trusts’. Then as the USSR pushed forward rapid industrialization
through the First Five-Year Plan under Stalin, these ‘trusts’ were reorganized as various
ministerial departments, absorbed by the party and state’s administrative hierarchies as
the foundation of the Stalinist system of industrial ministries (Nove, 1977; Harrison and
Markevich, 2008). Pursuing heavy industries-oriented catch-up, China developed similar
institutional arrangements in the 1950s and 1960s (Lin et al., 2003; Wu, 2005; Li, 2015).
The rapid industrialization of Japan and South Korea, with their extensive use of industri
al policy, has provided a different paradigm from both the neo-liberal Washington Consen
sus and socialist central planning. Japan and South Korea have been considered quintes
sential examples of late development, in which the developmental state coordinated eco
nomic development through rational industrial policies designed and implemented by an
elite bureaucracy (Johnson, 1982; Amsden, 1989, 2001; Amsden and Singh, 1994; Chang,
1994; Kasza, 2018; Lee, Chapter 23 in this volume).
Page 5 of 47
The late industrialization and developmental state literature has emphasized the complex
relations and crucial roles of the state and big business in modern economic development
and technological innovation (Gerschenkron, 1962; Chandler, 1977, 1990; Chandler et al.,
1997; Nolan, 2001a, 2001b; Colpan et al., 2010). With economies (p. 721) of scale and
scope, big businesses are ‘the fertile learning ground for technological, managerial, and
organizational knowledge for an entire economy’ (Chandler et al., 1997: 25; Penrose,
2009). Lee et al. (2013) have shown that big businesses have a significant and positive ef
fect on economic growth and are positively associated with stability in economic growth.
Many developing countries have fewer big businesses than predicted by their size and
thus failed to escape the middle-income trap (Lee, 2019). Apart from responding to mar
ket failures (or institutional void), large business groups in developing countries can
serve as the organizational device for economic catch-up under state activism (Lee, 2006,
2010; Hahn and Lee, 2006; Khanna and Yafeh, 2007; Lee and Jin, 2009). In particular,
large business groups in both Japan (kereitsus) and South Korea (chaebols) featured
prominently in their economic miracles. Both Japanese and South Korean governments
used various industrial policies to support their domestic big businesses to gain interna
tional competitiveness, including actively encouraging mergers between their leading in
digenous firms in strategic industries, but they also encouraged oligopolistic rivalry to
avoid monopoly, using exports and international market shares as the performance goals
to evaluate big businesses supported by government policy (Amsden and Singh, 1994;
Chang, 1994; Amsden, 2001; Nolan, 2001a, 2001b; Lee, Chapter 23 in this volume).
China’s national champions industrial policy has drawn inspiration from non-mainstream
economic theory and the policy practices of other late industrializing countries as well as
from political considerations of nationalism and power (Nolan, 2001a). In particular, it
has been heavily influenced by selective policy learning from Japan and South Korea
(Heilmann and Shih, 2013). As Nolan (1995) pointed out in the 1990s, it is conceivable
that the communist party bureaucracy in China could adapt and be effectively trans
formed into a new type of developmental state. Rather than to ‘efficiently close much of
the old structure’, an alternative approach for enterprise reform could be to learn from
the industrial policies used by Japan and South Korea during their catch-up phase of
growth, so the enterprise system built up under the command economy could be upgrad
ed instead of demolished. Although it was largely overlooked in the World Bank’s policy
prescriptions for transition economies, China has in effect adopted this heterodox ap
proach over the past three decades (Nolan and Wang, 1999; Nolan, 2001a, 2001b; Suther
land, 2003; Li, 2015). There is now a growing literature of firm-level empirical studies ex
amining China’s industrial policies and the performances of China’s national champions
since the early 1990s, such as the pioneering contributions of Nolan (2001a, 2001b, 2012,
2014) on the development of China’s national champions in multiple sectors; as well as a
number of single-sector studies, including, for example, the petroleum and petrochemical
industry (Zhang, 2004, 2015, 2018), the aerospace and aviation industry (Liu, 2012), the
automobile industry (Sutherland, 2003; Lu and Feng, 2005; Thun, 2006), the shipbuilding
industry (Moore, 2002), the coal mining industry (Nolan and Rui, 2004; Rui, 2005), the
steel industry (Sun, 2005, 2007; Nolan and Rui, 2007), and banking (Nolan, 2015).
Page 6 of 47
China’s industrial policy is broadly similar to those of Japan and South Korea in terms of
performing the functions of stimulating investment and structural changes (p. 722) with
the development of large indigenous business groups under the guidance of an elite eco
nomic bureaucracy. However, China’s model also fundamentally differs from Japan and
South Korea in several key aspects, especially in terms of direct state ownership and con
trol, the institutional configuration of state bureaucracy, and the openness to foreign di
rect investment (FDI) (Naughton and Tsai, 2015). Moreover, China’s national champions
are typically not as diversified across a range of industrial sectors as Japanese and South
Korean business groups. They tend to be more vertically integrated, each focusing on a
core sector, such as CNPC and Sinopec in oil and gas, ICBC in commercial banking, Chi
na Mobile in telecom, and CRRC Group in railway rolling-stock manufacturing (Nolan,
2001a, 2001b; Steinfeld, 2004; Lin and Milhaupt, 2013). Fierce competition in China’s do
mestic markets with high penetration by foreign multinational corporations tend to re
strict excessive diversification of domestic firms (Lee and Woo, 2002). China’s central
government has also directed that its national champions should focus on their core busi
ness areas rather than pursuing unrelated diversification (Li, 2015). Being less diversi
fied, China’s national champions might be more vulnerable to firm-specific or sector-spe
cific risks than Japanese kereitsus and Korean chaebols (Lee and Woo, 2002), but this is
mitigated by common strategic state ownership in these national champions. From this
perspective, China’s industrial policy has generated innovative features that adapt to its
own specific political and economic conditions as well as a different international policy
environment from that which Japan and South Korea encountered in their catch-up
(Nolan and Wang, 1999).
25.3 Industrial Policy and the Rise of China’s

National Champions
China’s national champions industrial policy practices reflect its unorthodox approach of
reforming its former central planning system and promoting industrial catch-up. The his
torical origin of China’s national champions can be traced back to the mid-1950s, when
all large-scale industrial production and finance in Mainland China were consolidated in
to a handful of vertical administrative hierarchies led by the Chinese Communist Party
bureaucracy. With decades of reform experiments, China has selectively imitated the gov
ernance forms, corporate structures, and practices of advanced industrial countries, and
combined them with the existing institutional arrangements and resources in its bureau
cratic structures. In particular, since the early 1990s, a key goal of China’s industrial poli
cy has been to nurture and consolidate selected large enterprise units into a number of
modern large corporations and business groups as China’s ‘main force’ to catch up and
compete with Western multinational big businesses. With continuous organizational
learning and capability building at the levels of both government bureaucracy and enter
prises, the core productive assets of China’s system of industrial ministries were trans
formed into a (p. 723) group of national champions under SASAC. Parallel restructuring
has occurred in the state-controlled financial system which has given rise to a group of in
Page 7 of 47
creasingly modernized, giant financial institutions as China’s national champions in fi

nance.
25.3.1 The Rise of China’s National Champions in Industrial Sectors
China’s policy experiment of restructuring the Stalinist system of central industrial min
istries into certain forms of large multi-plant business organizations under the state’s
control has a long history. It can be traced back to the early 1960s when President Liu
Shaoqi proposed that China should establish a number of giant national ‘corporate
trusts’ (tuolasi) as socialist China’s counterparts of Western big businesses. In the 1960s,
the State Economic Commission (SEC) in China was tasked with studying how modern
big businesses were organized in advanced industrial countries such as the United
States, the United Kingdom, France, and Japan, notably including some relatively de
tailed analyses of General Electric and Westinghouse Electric, as well as earlier Soviet
practices under Lenin. While this policy approach was halted by political campaigns
launched by Chairman Mao, the long-term impact of this essential idea recurred in differ
ent forms. In the 1980s, China had two competing industrial policy approaches. On the
one hand, China established a set of large central state-controlled zonggongsi (literally
translated as ‘general corporation’), such as Sinopec, as national administrative corpora
tions to emulate Western big businesses. On the other hand, the dominant approach of in
dustrial reforms promoted decentralization and enhancing ‘enterprise autonomy’ for indi
vidual plants. Many large industrial plants or enterprise units previously controlled by
central industrial ministries were transferred to local governments (primarily at the lev
els of provinces and cities). Challenged by decentralization reforms, some industry-wide
zonggongsi were dissolved, while others devolved managerial authority to their subordi
nate plants, with varied outcomes across sectors (Li, 2015, 2016).
Since the early 1990s, however, industry policy of nurturing national champions has been
a high priority endorsed with strong political commitments by China’s leadership. Build
ing a number of central state-controlled oligopolistic big businesses in the form of large
corporations and large business groups has been at the core of China’s industrial policy
making over the past three decades. These national champions are supposed not only to
dominate the ‘commanding heights’ of China’s national economy, but also to gradually de
velop international competitiveness. Indeed, as China deepens its integration with the
global economy, domestic industrial consolidation is regarded as imperative for national
champions to survive and compete in international markets (Nolan, 2001a, 2001b; Suther
land, 2003; Li, 2015; Naughton, 2015). This policy logic is illustrated by the following
quote from Wu Bangguo, former vice premier of China’s State Council in charge of indus
trial reforms in the late 1990s and early 2000s:
(p. 724)
In our world today, economic competition among nations is in fact among each
nation’s large enterprises and business groups. A nation’s economic might is con
centrated and manifested in the economic power and international competitive
Page 8 of 47
ness of its large enterprises and business groups…the United States, for example,
relies on General Motors, Boeing, Du Pont and a batch of other multinational cor
porations. Japan relies on its six large business groups and Korea relies on its ten
large commercial conglomerates. In the same way now and in the next century
our nation’s position in the international economic order will be to a large extent
determined by the position of our nation’s large enterprises and business groups…
If we let our strong large-scale enterprises and business groups all fight alone,
everyone will still find it difficult in the ever-intensifying domestic and internation
al competition to compete on equal terms with large international companies. We
must therefore unite and rise together, develop economies of scale and scope and
nurture a ‘national team’ capable of entering the world’s top 500. (Jingji Ribao, 8
January 1998, quoted from IIECASS, 1998: 124)
During this period, instead of creating industry-wide monopolies or pursuing comprehen

sive decentralization, China’s central government selected the core productive assets and
enterprise units in sectors perceived as strategic, carved them out from their respective
ministries, consolidated them into large central state-controlled corporate groups, and
promoted oligopolistic rivalries between them. Those assets and enterprise units deemed
outside this category were decentralized and privatized as the reforms proceeded. For ex
ample, the Ministry of Electricity Industry in China was first reorganized into the State
Electricity Industry Corporation as an industry-wide zonggongsi, then split up into sepa
rate corporate groups; the ministries of coal, chemicals, machinery, and metallurgy indus
tries were downgraded to state bureaus and later reorganized as industrial associations;
the Ministry of Post & Telecommunications and Ministry of Electronics Industry were
merged. All of them had their selected core assets carved out and reorganized as large
corporate groups (Li, 2015).
By 2003, most of the key large designated ‘backbone’ state enterprise units in China’s
system of central industrial ministries had been carved out and consolidated into 196
large central state-controlled business groups. They constitute China’s ‘national team’ in
industrial sectors. SASAC was established to exercise state ownership and supervise
these business groups on behalf of the party and state. Far from being a passive share
holder, SASAC has actively shaped the strategies and structures of its subordinate big
businesses. As stated by Li Rongrong, head of SASAC between 2003 and 2010, the goal of
SASAC is that ‘China must nurture its own multinationals to challenge the dominance of
foreign corporations’ and SASAC was to ‘vigorously pursue a strategy of creating major
corporate conglomerates’ and ‘accelerate the strategic adjustment of China’s economic
structure’ (Buckley, 2005). Without taking the boundaries of its subordinate firms as giv
en, SASAC directed that the business groups under its supervision should be consolidated
into eighty to one hundred competitive big businesses in targeted strategic industries,
among which thirty to fifty were supposed to become large corporate groups with strong
international competitiveness (SASAC, 2009). (p. 725) As Naughton (2005) commented,
‘increasingly, the SASAC formulates its agenda in terms that a Wall Street investment
bank would understand’; under SASAC’s supervision, each of these large business groups
were supposed to become No. 1, 2, or 3 in the markets of its core business areas, or it
Page 9 of 47
would be discarded through state-directed mergers, acquisitions, or even outright bank

ruptcy. Guided by this overarching policy goal, SASAC has initiated comprehensive re
form programmes to restructure and strengthen the national team. At the industry level,
it has clarified the definition of strategic industries targeted by the state to maintain con
trol. According to SASAC’s designation, central state-controlled business groups should
operate in three categories of sectors: (1) key industries concerning national security and
the lifeline of the national economy; (2) basic and pillar industries; and (3) other targeted
industries. The first category of key industries comprises seven sectors: defence, power
generation and distribution, oil and gas and petrochemicals, telecommunications, coal,
aviation, and shipping. The state maintains absolute control over these sectors, promot
ing a number of large ‘backbone’ firms within them as first-class global companies (Li,
2015).
At the firm level, transforming traditional state-owned enterprises into corporations is at

the core of SASAC’s policy agenda. Corporatization clarifies the relations between own
ers and managers, the scope of managerial authority and responsibility, and incentive
mechanisms (Naughton, 2015). Establishing modern corporate governance structures
and listing their minority stakes in stock markets is critical for the rise of China’s national
champions, with the government filtering potential candidates for public listing on do
mestic and international stock markets and prioritizing the needs of targeted large ‘key’
enterprises. In particular, the late 1990s saw the rise of international stock market list
ings for China’s central state-controlled corporate groups carved out from various indus
trial ministries or the State Council. The pioneering deals included the initial public offer
ings (IPOs) of China Mobile, PetroChina (the listing company of CNPC), and Sinopec.
Each of these companies took over a large chunk of industrial assets and enterprise units
previously managed under their old industrial ministries. It was these international list
ings that started to bring China’s emerging ‘national team’ onto the global stage.
Since its establishment, SASAC has been a major force shaping the use of stock markets
to support national champions. The listing strategy preferred by SASAC is the holistic
IPO, which corporatizes each entire business group as a whole and lists its minority
stakes, rather than separately listing the subsidiary enterprise units of the larger busi
ness group. For example, China Railway Group as a whole listed a minority stake of 40
per cent in Shanghai and Hong Kong stock markets for RMB43.4 billion in 2007; China
State Construction Engineering Corporation as a whole listed a minority stake of 40 per
cent in Shanghai for around RMB50.2 billion in 2009. After their IPOs, national champi
ons can raise further capital through various secondary offerings and refinancing without
diluting the central government’s corporate control (Walter and Howie, 2006, 2011; Li,
2015).
Under its industrial targeting framework, SASAC has led several rounds of industrial re
structuring and consolidation. Most of China’s national champions started as relatively
loosely amalgamated groupings. With a core ‘parent company’ serving as the (p. 726)
group’s headquarters, each group incorporated many ‘son companies’, ‘grandson compa
nies’, and even ‘great-grandson companies’ engaged in a wide spectrum of disparate
Page 10 of 47
business activities, many of which have nothing to do with the group’s designated core
business areas. While diversification might be profitable from the perspective of the indi
vidual group, from the central government’s point of view it was considered detrimental
to the overall goal of strengthening China’s ‘national team’. To rein in ‘undisciplined’ di
versification, SASAC has designated ‘core businesses’ for each of the national champions.
The approach of SASAC is to consolidate national champions’ designated core businesses,
gradually spinning off their existing non-core businesses and discouraging new invest
ments in non-core business areas. To achieve this goal, SASAC strengthened its supervi
sion of national champions’ investment management process and required them to obtain
its approval before making substantial investments in non-core business areas. It also
pushed national champions to streamline their internal management hierarchies and im
prove their core competences in the designated key business areas.
To strengthen the international competitiveness of China’s ‘national team’ as a whole,

SASAC has conducted a series of mergers and consolidation programmes among China’s
national champions. For instance, SASAC consolidated China’s aviation industry in 2008
with the merger of the Aviation Industry Corporation of China I and II. Over the past
decade, SASAC has consolidated a variety of sectors including transportation infrastruc
ture, machinery building, energy, steel and non-ferrous metallurgy industries, with the
rate of state-led mergers accelerating in recent years. For example, in 2015, SASAC
merged CNR Group and CSR Group into CRRC, which has overtaken Alstom and Siemens
to become the world’s largest manufacturer of railway equipment. In 2016, SASAC
merged Baosteel Group and Wuhan Iron & Steel Group to create the second-largest steel
producer in the world. In 2017, it further merged China Guodian Corporation and Shen
hua Group into China Energy Investment Corporation (China Energy), forming the
world’s largest producer of coal, thermal power, wind power, coal-to-liquid fuels and coal
chemicals. The guiding principle of SASAC-led restructuring is to transform the national
team into a small number of large corporations and business groups capable of dominat
ing their core business areas in the domestic markets and competing against top incum
bent multinationals in their respective industries. By the end of 2018, the original 196
business groups under SASAC had been restructured and consolidated into ninety-six
groups. Between 2003 and 2017, the total assets of these business groups under SASAC
increased from RMB8.3 trillion to 54.5 trillion, growing around 14 per cent per annum on
average. Their combined revenues and profits increased from RMB4.5 trillion and 0.3 tril
lion in 2003 to RMB26.4 trillion and 1.4 trillion respectively in 2017. The total taxes they
paid to the state also grew from RMB0.36 trillion to over 2 trillion during this period
(SASAC, 2004, 2017). Occupying the ‘commanding heights’ of China’s strategic sectors
such as oil and gas, power generation, telecommunications, aerospace, aviation, automo
bile, mining, airlines, and nuclear, these business groups are a key pillar of China’s indus
trial economy.
Page 11 of 47
(p. 727) 25.3.2 The Rise of China’s National Champions in Finance
Between the 1950s and the 1980s, China’s financial system was organized as a ‘mono-
bank’ that amounted to no more than a bureaucratic appendage of China’s Ministry of Fi
nance. Driven by a combination of market-oriented reforms and industrial policies since
the 1980s, China has now developed a complex network of financial institutions, with a
handful of large state-controlled commercial and policy banks, asset management compa
nies, and insurance companies, and two stock exchanges at the apex of the system.
China’s financial reforms have given rise to its national champions in finance, which also
sustain critical support for the reforms and growth of the industrial-sector national cham
pions as they channel financial resources to support the targeted firms.
The focus of China’s financial reforms in the 1990s was to deal with the soft-budget con
straints and moral hazard inherent in China’s enterprise system. The reforms aimed at
carving out financial institutions from the government bureaucracy and providing them
with more operational autonomy. In particular, China established three new state-owned
policy banks: China Development Bank, the Export-Import Bank of China, and Agriculture
Development Bank of China. The idea was to reduce the policy burdens facing China’s
large commercial banks by spinning off their policy loan responsibilities to the three new
policy banks. However, this separation of policy banks and commercial banks did not re
duce the importance of large commercial banks in China’s industrial policy. While China
Development Bank and the Export-Import Bank of China have played increasingly impor
tant roles in financing targeted large enterprises, large commercial banks have consis
tently been the chief source of funding for China’s industrial restructuring and large
state-owned enterprises reform. In particular, between 1991 and 1998, the five largest
commercial banks generated well over RMB300 billion of non-performing loans to their
SOE clients. The accumulation of these non-performing loans was partly due to inefficien
cy and lax internal control within these banks, as well as interference by various levels of
local government, but also due to absorbing the costs of implementing the central
government’s industrial policy (Li, 2015).
In the context of the Asian financial crisis and China’s entry into the WTO, it became in
creasingly clear that China’s commercial banks had profound weaknesses. With massive
non-performing loans, weak risk management, backward technologies and operational
mechanisms, together with deeply problematic corporate governance, their prospects of
competing with the giant global banks on the ‘global level playing field’ of the WTO were
bleak (Nolan, 2015). In the late 1990s, China’s central government implemented major
banking reforms to mitigate financial-sector fragility and improve the competitiveness of
its large commercial banks. The state arranged a massive capital injection of RMB270 bil
lion into the four largest state-owned commercial banks (Yi, 2009). By spinning off staff
from the commercial banks, China Orient, Cinda, Great Wall, and Huarong were estab
lished in 1999 as the ‘Big Four’ asset management companies (AMCs) under the Ministry
of Finance. Funded by the (p. 728) Ministry of Finance (MoF) and the People’s Bank of
China, they took on around RMB1.4 trillion of problematic assets at face value from the
China Construction Bank (CCB), Bank of China (BOC), Industrial and Commercial Bank of
Page 12 of 47
China (ICBC), Agriculture Bank of China (ABC), and China Development Bank (CDB) in
1999–2000. While these asset acquisitions cleaned up the big banks’ balance sheets, they
fell short of significantly strengthening their risk-management capabilities. Indeed, as
their lending grew in subsequent years, non-performing loans rapidly accumulated again
to reach a staggering level of around RMB2.1 trillion by the end of 2002. This made the
big banks vulnerable against the backdrop of China’s WTO entry, as China was expected
to open its financial services sector to foreign financial institutions after 2006.
Following intense policy debates in 2002‒03, China announced the goal of developing ‘in
ternationally competitive modern financial enterprises’ and decided to continue to re
strict the roles of foreign financial institutions in China, limiting their permitted owner
ship shares in China’s indigenous financial institutions. China’s central government also
decided not to split up the large state-owned commercial banks, but to reform them as
‘complete corporate entities’. This entailed further removals of non-performing loans; cre
ating a strategic partnership with leading global banks to facilitate learning; investing
heavily in advanced information technology; upgrading their risk management, human re
sources, and operational mechanisms; and eventually floating their minority shares on in
ternational and domestic stock markets. To implement this strategy, China’s central gov
ernment established the Central Huijin Investment Corporation in 2003 as the key organi
zational vehicle to facilitate the reform of targeted financial institutions. The Big Four
AMCs were directed to acquire a second batch of non-performing loans between 2004
and 2005 from big commercial banks, totalling around RMB1.6 trillion (Walter and
Howie, 2011). The MoF also stepped in directly and made an innovative arrangement es
tablishing ‘co-managed funds’ with ICBC and ABC respectively in 2005 and 2008. Funded
by the issue of MoF debts, the central government further offloaded around RMB910 bil
lon of non-performing loans from ICBC and ABC into the co-managed funds. By 2010,
China’s five largest commercial banks had all completed their public listings in Shanghai
and Hong Kong Stock Exchanges. All but ABC had introduced leading global financial in
stitutions as foreign ‘strategic investors’ to acquire some portions of their minority
stakes. Public listings on stock markets exposed these banks to the scrutiny of global in
vestors and provided them with market-based capital-raising channels and discipline
(Pan, 2012; Nolan, 2015). While these public listings amount to partial privatization,
China’s central government has maintained absolute majority ownership after their IPOs;
it remains the single largest shareholder in all of the five largest commercial banks and
still fully owns the three policy banks. During the restructuring process, China’s central
government prohibited foreign financial institutions from acquiring control of any major
domestic financial institutions in China. Global banks such as Citigroup and HSBC had to
build up their own branch networks step by step in China. By 2010, when all the five
largest state-controlled commercial banks had completed their public listings, foreign
banks only accounted for around 2 per cent of China’s total financial assets (Walter and
Howie, 2011; Nolan, 2015; Li, 2015). Over the past decade, China’s national champion
banks have ranked among the global top banks in terms of revenue, profits, market capi
talization, and capital adequacy ratios, while the market shares of foreign banks remain
Page 13 of 47
negligible in China despite the fact that China’s government has removed many of its pol
icy restrictions.
Guided by the state’s industrial policy, there has been close strategic cooperation be
tween China’s national champions in industrial sectors and in finance. Large state-con
trolled banks were crucial in channelling credit to support corporate restructuring in the
industrial sectors. For example, in the 1990s, inspired by the Japanese experience,
China’s central government designated ‘main banks’ to guarantee credit supply to 300
(later expanded to 512) key enterprises. In the early 2000s, the central government ap
proved extensive debt/equity swaps for central state-owned business groups, with a total
of around RMB377 billion in debts owed to big banks converted into equities held by the
Big Four AMCs and China Development Bank, which profoundly relieved the financial
burdens facing those enterprises (SASAC, 2005: 62). Guided by the industrial policy to
upgrade China’s aircraft industry, Aviation Industry Corporation of China (AVIC) estab
lished various strategic partnership agreements with China’s banking champions, gaining
RMB282 billion in credit lines from them in 2009. Commercial Aircraft Corporation of
China (COMAC) received over RMB230 billion in credit-line support from these banks by
2011 (Li, 2015).
Both state-controlled commercial banks and policy banks have played crucial roles in fi
nancing the overseas business growth of China’s industrial national champions. For ex
ample, China Development Bank and the Export-Import Bank of China are the most im
portant strategic partners of China’s oil national champions (Sinopec, CNPC, and
CNOOC) in their efforts to secure long-term supplies of oil and gas overseas, with various
‘loan for oil’ and ‘loan for gas’ deals with countries such as Turkmenistan, Kazakhstan,
Ghana, Venezuela, Brazil, and Russia totalling around US$120 billion (Jiang and Sinton,
2011; Downs, 2011; Jiang and Ding, 2014; Zhang, 2018).
25.4 State‒Business Relations and the Gover

nance of China’s National Champions
As shown earlier in the chapter, the central plank of China’s national champions industri
al policy is to transform the core assets and enterprise units of the former command
economy into large state-controlled corporations and business groups that are competi
tive in a predominantly market-oriented environment. Rather than ‘efficiently closing
much of the old structure’, China has chosen to reinvent the core of its old industrial
structure. This process has taken advantage of the existing institutions and resources of
the party-state, and has innovatively combined them with new (p. 730) institutional ele
ments, governance forms and practices learnt from advanced economies. It involves con
tinuous capability building, learning, and organizational changes at both the government
and enterprise levels, which have also led to some special features of state‒business rela
tions in China.
Page 14 of 47
25.4.1 Government Capability and Learning
In the 1980s and 1990s, the ‘transition orthodoxy’ focused on excluding communist bu
reaucrats from participating in the reforms of transition economies. It was argued that
‘the collapse of communist one-party rule was the sine qua non for an effective transition
to a market economy’; ‘it is naïve to think of existing bureaucracy as equipped, profes
sionally and temperamentally, to implement sophisticated policies based on Western-style
theories of the “second best”’; and ‘the bureaucracy cannot be relied upon for efficiency
in regulating monopoly prices, promoting infant industries, or implementing industrial
policy’ (Lipton and Sachs, 1990: 87–8). In China, however, industrial reforms have been
consistently led by the existing communist party bureaucracy with emphases on organiza
tional learning and capability building to implement industrial policy.
Following Deng Xiaoping’s cadre modernization scheme, the educational level and techni
cal competence of China’s communist party cadres and economic bureaucrats has im
proved dramatically since the 1980s (Lee, 1991; Li, 2001; Zheng, 2004, 2010). Large
state-owned enterprises and central industrial ministries also served as a major reservoir
from which the Party could draw its top leadership talent. China’s industrial policymaking
primarily reflects the training, work experience, and ideological outlook of these elite
cadre-bureaucrats. From industrial ministries to modern big businesses, the transforma
tion of organizational forms, coordination mechanisms, and personnel roles undoubtedly
requires the acquisition of new knowledge, skills, capabilities, and behaviours. As China’s
State Council put it, ‘to restructure government administrative ministries into firm-like
economic organizations is a vital experiment for the economic system reform, and we
need a process of exploration and learning’ (The State Council, 1982).
Over the past four decades, China’s central government has undertaken multiple rounds
of administrative reforms to streamline and upgrade the bureaucracy’s capabilities and
mitigate coordination failures, creating new pivotal agencies such as SASAC and MIIT
(Ministry of Industry and Information Technology) to strengthen its industrial policy prac
tices with increasingly professionalized, efficient, and focused modes of intervention
(Pearson, 2005; Jung, 2011; Naughton and Tsai, 2015). Using SASAC and Central Huijin
as its key vehicles, China’s central government has also learnt to increasingly use finan
cialized means of managing assets and governing its national champions (Wang, 2015). In
particular, the stock markets have been used as key industrial policy instruments to nur
ture national champions (Walter and Howie, 2006, 2011). The Chinese Communist Party
also upgraded its cadre training and personnel management system as national champi
ons became corporatized and listed in stock markets (Brødsgaard, 2012; Li, 2019).
25.4.2 Enterprise Capability and Learning
Developing countries typically suffer from firm-level capability failure and size failure
(Lee, 2019). It is difficult for them to generate indigenous firms with sufficient research
and development (R&D) or technological capabilities and of sufficient size to achieve
economies of scale. Under the traditional industrial ministries system, the very definition
Page 15 of 47
of ‘firm’ or ‘enterprise’ was ambiguous in China. Between the 1950s and 1980s, most of
the ‘large state enterprises’ were merely large-sized individual plants (or projects, facto
ries, mines, and research institutes, etc.) which had no distinct organizational boundaries
from their governing industrial ministries and corresponded only to the lowest levels of
production entity in a typical modern big business in an advanced industrial economy
(Komiya, 1987a, 1987b; Nolan, 1996, 2001a, 2001b). The coordination of these state en
terprises’ activities was fragmented among various government bureaucratic agencies.
They hardly qualify as autonomous organizations of production, let alone fully fledged
business firms.
China’s corporatization reforms and state-led restructuring to create national champions

have involved a crucial learning process about how to define the boundaries of firms and
to develop organizational capabilities at the level of large multi-plant firms and business
groups. In particular, the corporatization and initial public offerings of China Mobile,
Sinopec, and CNPC on the eve of China’s WTO entry demonstrated the feasibility of Chi
na consolidating core assets and enterprise units of traditional industrial ministries into
giant corporate groups in a relatively short period of time (Zhang, 2004, 2017). Guided by
this approach, the state-controlled corporate groups and listed companies became the
new units and organizational devices of firm-level catch-up. Similarly, China’s banking
sector restructuring was also designed to promote organizational learning and capability
building at the level of large nationwide banks.
Overall, listing the minority stakes of China’s national champions in stock markets is not
just about raising capital, but more about absorbing modern corporate governance prac
tices, management structures and market-based disciplinary mechanisms, and creating
partnership with international banks, law firms, auditors, accountants, regulators, and in
stitutional investors to facilitate such learning. The strategies employed by China’s indus
trial policymakers and national champions thus involve building linkages with incumbent
multinational corporations and global capital markets, targeting resources for leverage
which are imitable and transferable, and enhancing organizational learning to catch up
(Matthews, 2002). Starting from relatively loose amalgamations of enterprise units
carved out from former bureaucratic hierarchies, these business groups and financial in
stitutions have gradually become fully fledged modern (p. 732) firms, with coherent corpo
rate structures coordinating and performing various business functions including strate
gic planning, operations, financing, procurement, marketing and sales, human resources,
and inhouse R&D.
25.4.3 Networked Hierarchy, Institutional Bridging, and Reciprocal

Control
The governance of China’s national champions in industrial sectors is embedded in a

complex organizational architecture, which can be seen as a gigantic pyramid. With the
Party centre and the State Council at the apex, tiers of national champions are organized
under the central government and strategically linked to each other (Naughton, 2015).
Lin and Milhaupt (2013) proposed the notions of networked hierarchy and institutional
Page 16 of 47
bridging to characterize China’s unique model of industrial organization: a hierarchical

system in terms of state ownership and control, the formulation and implementation of in
dustrial policy, and the internal structures of these firms. China’s industrial national
champions are typically vertically integrated groups with multiple layers of member en
terprise units linked by ownership arrangements (typically including a parent holding
company, one or a few core publicly listed subsidiaries, some internal research institutes
and finance companies and several tiers of operational enterprise units). SASAC exercis
es shareholder’s rights on behalf of the state, but its control rights are incomplete as the
Party maintains ultimate control of key corporate appointments (such as chairman and
CEO) in these firms (Brødsgaard, 2012; Li, 2016). A dense structure of inter-organization
al networks connects firms with the Party and various government agencies as well as
other firms. These institutional bridging networks facilitate the flow of information, col
laboration, and relationship building in the system (Lin and Milhaupt, 2013; Li, 2015).
The governance regime of China’s national champions in finance shares similar features,
with the Ministry of Finance and Central Huijin as the most important shareholders, and
the Party retains control over key personnel appointments in major state-owned commer
cial banks, policy banks, insurance companies, asset management companies, and stock
exchanges (Heilmann, 2005; Pistor, 2012; Pistor et al., 2016).
The Chinese Communist Party’s nomenklatura control and personnel management system
are crucial in the governance of China’s national champions. The Party controls the ap
pointments and career prospects of top-level corporate leaders in national champions as
well as key officials in SASAC, MoF, and all financial regulators. It aims to establish a
meritocratic system to identify and promote cadres with both strong political loyalty and
professional competence. It often rotates and transfers cadres among different national
champions, regulatory agencies, and even local government leadership positions. Candi
dates for key leadership positions in the few dozen most important national champions
are groomed and monitored directly by the Party’s Central Organization Department
(COD) (Lin, 2011; Brødsgaard, 2012; Pistor, 2012; Li, 2016, 2019; Leutert, 2018a, 2018b).
These appointments concern not only each specific national champion, but also the needs
of the system as a whole. It is often the case that a key change of personnel will call for,
and be accompanied by, a series of top personnel reshuffles, involving multiple institu
tional units apparently independent of each other. Like the former central industrial min
istries, China’s national champions have served as a talent pool for the Party to recruit its
top political elites and a pathway for ambitious cadres to advance their political careers
(Li, 2011; Lin, 2017). These linkages tend to generate what Lin (2011: 73) calls ‘synchro
nized incentives and mobility’: top personnel are motivated by both political and profes
sional achievement, and can be moved back and forth across political hierarchy and state-
controlled corporate hierarchy under the Party’s personnel management system. Lin and
Milhaupt (2013) argue that this structure amounts to an ‘encompassing organization’ and
policy coalition whose members have an incentive to promote the prosperity of the over
all system (Olson, 1982).
Page 17 of 47
A core policy challenge of late industrialization is how to exert discipline over the interac
tions between the state and firms. The market paradigm relies on competitive market
structures and creative destruction induced by technological changes serving as discipli
nary forces, while the command economy paradigm relies on direct state planning and
monitoring, which is typically characterized by soft budget constraints. Successful indus
trial policies typically have mechanisms to establish certain forms of reciprocal control in
state‒business relations, with performance standards and disciplinary conditions at
tached to the state support that firms receive (Amsden, 2001). The rise of China’s nation
al champions has relied on extensive state supports such as subsidies, domestic market
protection, preferential access to finance, state-orchestrated mergers, and bailouts. How
to strengthen financial discipline and harden budget constraints has been a critical policy
challenge.
Over the past three decades, China has gradually developed its own approach to recipro
cal control in governing its national champions. First, it builds on the Party’s personnel
management and disciplinary system to shape competition for promotion among cadres
(Lin, 2011; Brødsgaard, 2012; Pistor, 2012; Li, 2016, 2019). Second, while erecting high
entry barriers and nurturing the rise of giant business groups in strategic sectors,
China’s central government has promoted oligopolistic rivalries among national champi
ons and encouraged them to compete in international markets, although export competi
tiveness is not a key performance criterion. Moreover, China has developed and increas
ingly resorted to various financialized policy tools and channels to impose discipline, cre
ating the unique structure of a ‘shareholding state’ (Wang, 2015). On the one hand, the
corporatization and public listings of national champions have rendered these firms in
creasingly sensitive to shareholder value. As a controlling shareholder, SASAC has devel
oped a sophisticated performance evaluation system for national champions, setting vari
ous benchmarks and targets with an array of indicators such as profits, economic value
added, technical progress, return on equity, return on asset, leverage ratios, and corpo
rate social responsibility, and factoring in the different industrial environments within
which different national champions operate. In (p. 734) some respects, the relations be
tween SASAC and national champions now resemble the ways private equity funds man
age their portfolio companies.
On the other hand, the restructuring of China’s banking sector has also made its large
state-controlled commercial banks more effective disciplinarians. These banks now face
strong pressure from the central government (as the controlling shareholder) to create
shareholder value, strict capital adequacy regulatory requirements consistent with global
standards and more robust internal risk management systems. Between 1999 and 2008,
China’s central government had offloaded a total of around RMB4 trillion non-performing
loans from the big banks through various vehicles of the People’s Bank of China, MoF,
and the Big Four AMCs. While the big banks were recapitalized by the state with the MoF
issuing new government bonds to fund the restructuring, they were required to pay out
sufficient dividends from their profits in the subsequent years for the MoF to pay off the
bonds. This imposes additional discipline on banks’ operational performance. Overall,
China’s Ministry of Finance has been a fairly prudent ‘gatekeeper’ and disciplinarian dur
Page 18 of 47
ing the industrial and banking reforms of the past two decades (Walter and Howie, 2011;
Pan, 2012; Jiang, 2019).
However, it remains too early for China to declare success in developing an effective sys
tem of governing China’s national champions. The Party’s personnel management and
cadre disciplinary system is prone to complex principal‒agent problems, often inter
twined with cronyism and factionalism. Given the informational constraints, it is difficult
for the Party to rely on personnel control to effectively monitor and discipline politically
well-connected business leaders who possess substantial managerial autonomy in daily
business operations. As the recent anti-corruption campaigns reveal, corruption in
China’s national champions often involves hidden political networks stretching across the
Party and government hierarchy (Brødsgaard, 2012; Li, 2016). SASAC’s ownership con
trol rights are also incomplete and it faces strong resistance from the national champions
(Naughton, 2015). To what extent and how China’s big banks and capital markets can im
pose effective discipline remains an open question as the Chinese economy shifts to a
‘new normal’ of slower growth and China’s financial system further liberalizes. Above all,
industrial policy is a highly politicized process of resource allocation, creating winners
and losers with different interests and political voices. Now China’s national champions
industrial policy is facing increasing criticisms and external pressure from international
organizations and Western advanced countries (World Bank, 2012; Wu, 2016).
25.5 International Competitiveness of China’s

National Champions
Over the past four decades, as China has deepened its reform and opening up, the busi
ness systems of the advanced countries have also experienced profound (p. 735) restruc
turing which was arguably no less revolutionary than the rise of modern big businesses
during the late nineteenth and early twentieth century (Nolan, 2001a, 2001b; Steinfeld,
2004; Nolan, 2012). The rise of giant global firms, the emergence of big-business-centred
production networks and the explosion of mergers and acquisitions dramatically re
shaped the landscape of global business competition. The commanding heights of the
global business system were almost entirely occupied by firms from high-income coun
tries with superior technologies and powerful brands. These constituted the ‘system inte
grators’ at the apex of global value chains. As they consolidated their leading positions,
the system integrator firms, with enormous procurement expenditures, exerted intense
pressures on their supply chains to minimize costs and stimulate technical progress. This
established the global context within which China’s central government designed and
practised its national champions industrial policy (Nolan, 2001a, 2001b; Zhang, 2004;
Nolan, et al., 2007; Nolan, 2012, 2014).
While giant multinational companies headquartered in advanced economies have increas

ingly penetrated the Chinese economy since China’s WTO entry, China’s national champi
ons still occupy the apex of China’s strategic sectors. They have built up corporate struc
tures that increasingly resemble the global leading big businesses of their respective in
Page 19 of 47
dustries. They have developed significant technological capabilities and well-trained, so
phisticated managerial leadership, floated their minority shares on domestic and interna
tional stock markets, established joint ventures or strategic partnerships with global lead
ing big businesses, and achieved a scale unprecedented in China’s business history
(Nolan, 2012; Li, 2015). In 2009, there were only a total of forty-three Mainland Chinese
firms in the Fortune Global 500, among which thirty-eight were central state-controlled
national champions. By 2018, however, the number had increased to 107. Among them,
fifty-eight were national champions which generated a total of US$4.4 trillion combined
revenue and US$221 billion combined profit in 2017/18 (Table 25.1). Among the best per
forming were ICBC whose profits alone reached US$42.3 billion, roughly equivalent to
the combined profits of JP Morgan Chase and Bank of America; China Mobile’s profits
reached over US$10.9 billion, compared with US$9.9 billion for Softbank and US$3.9 bil
lion for Deutsche Telekom. China’s State Grid, Sinopec, and CNPC all ranked among the
world’s five largest energy companies by revenue. Among the world’s ten largest aero
space and defence companies, five are China’s national champions: China North Indus
tries Group, AVIC, China South Industries Group, China Aerospace Science and Technolo
gy Corporation, and China Aerospace Science and Industry Corporation. Among the
world’s ten largest engineering and construction companies, there are also five Chinese
national champions: China State Construction Engineering, China Railway Engineering
Group, China Railway Construction, China Communications Construction, and PowerChi
na.
Page 20 of 47
Table 25.1 China’s National Champions in Fortune Global 500, 2018
Ranking Company Name Revenue (million Profit (million Core business

US$) US$)
2 State Grid Corpora 3,48,903 9,533 Power grid

tion of China
(SGCC)
3 Sinopec Group 3,26,953 1,538 Petroleum and

petrochemicals
4 China National Pe 3,26,008 −691 Petroleum and nat

troleum Corpora ural gas
tion (CNPC)
23 China State Con 1,56,071 2,675 Construction engi

struction Engineer neering
ing Corporation
(CSCES)
26 Industrial & Com 1,53,021 42,324 Banking

mercial Bank of
China
31 China Construction 1,38,594 35,845 Banking

Bank
Page 21 of 47
40 Agricultural Bank of 1,22,366 28,550 Banking

China
42 China Life In 1,20,224 267 Insurance

surance
46 Bank of China 1,15,423 25,509 Banking
53 China Mobile 1,10,159 10,932 Telecom
56 China Railway 1,02,767 1,170 Infrastructure con

Group (CRECG) struction & equip
ment manufactur
ing
58 China Railway Con 1,00,855 1,309 Infrastructure con

struction Corpora struction & equip
tion (CRCC) ment manufactur
ing
65 Dongfeng Motor 93,294 1,400 Automobile

Corporation
86 China Resources 82,184 3,152 Diversified

Group
Page 22 of 47
87 China National Off 81,482 3,019 Petroleum and nat

shore Oil Corpora ural gas
tion (CNOOC)
91 China Communica 79,417 1,545 Infrastructure con

tions Construction struction & equip
Company (CCCC) ment manufactur
ing
98 Sinochem Group 76,765 753 Energy and chemi

cals
101 China Energy In 75,522 2,495 Energy and chemi

vestment Corpora cals
tion (CHN Energy)
109 China Minmetals 72,997 −211 Metallic minerals

Corporation
110 China Southern 72,787 1,938 Power Grid

Power Grid (CSG)
113 China Post Group 72,197 4,961 Postal service and

logistics
Page 23 of 47
117 The People’s In 71,579 2,382 Insurance

surance Company
of China (PICC)
122 COFCO Group 69,669 394 Diversified
125 China FAW Group 69,524 2,856 Automobile
140 China North Indus 64,646 858 Defence

tries (Norinco
Group)
141 China Telecommu 63,974 1,820 Telecom

nications
149 CITIC Group 61,316 3,225 Diversified
161 Aviation Industry 59,263 363 Aviation and de

Corporation of Chi fence
na (AVIC)
162 China Bauwu Steel 59,255 22 Steel

Group
167 China National 57,989 −739 Chemicals

Chemicals Corpora
tion (ChemChina)
Page 24 of 47
168 Bank of Communi 57,711 10,390 Banking

cations
182 Power Construction 53,870 947 Energy and infra

Corporation of Chi structure
na (PowerChina)
194 China National 51,844 690 Healthcare

Pharmaceutical
Group (Sinopharm)
222 Aluminum Corpora 46,684 −429 Aluminium

tion of China (Chal
co)
242 China South Indus 44,785 740 Defence and auto

tries Group mobile
243 China National 44,701 15 Building materials

Building Material
Group (CNBM)
245 China Shipbuilding 44,431 716 Shipbuilding

Industry Corpora
tion (CSIC)
Page 25 of 47
256 Sinomach Group 42,638 472 Machinery Manu

facturing
273 China United Net 40,664 63 Telecom

work Communica
tions
280 China Merchants 39,971 4,039 Diversified

Group
289 China Huaneng 38,872 216 Power generation

Group
312 China Poly Group 37,002 1,153 Diversified
322 China Everbright 35,840 1,895 Diversified

Group
333 China Energy Engi 35,048 372 Energy engineering

neering Group (En
ergy China)
335 China Cosco Ship 34,668 1,404 Shipping and logis

ping Corporation tics
Page 26 of 47
343 China Aerospace 34,254 2,225 Aerospace

Science & Technol
ogy Corporation
346 China Aerospace 34,073 1,607 Aerospace

Science & Industry
Corporation
369 China Electronics 31,990 167 Electronics

Group
371 China National Avi 31,942 402 Aviation fuel

ation Fuel Group
(CNAF)
381 Xinxing Cathay In 31,078 440 Diversified

ternational (XXCIG)
385 CRRC Corporation 30,634 1,598 Rail Transit Equip

ment
388 China Electronic 30,176 1,774 Electronics

Technology Group
Corporation (CETC)
Page 27 of 47
393 China State Ship 29,797 371 Shipbuilding

building Corpora
tion Limited (CSSC)
395 State Power Invest 29,727 199 Power Generation

ment Corporation
(SPIC)
397 China Huadian Cor 29,612 333 Power Generation

poration
428 Ansteel Group 27,792 61 Steel
465 China Taiping In 25,598 449 Insurance

surance Group
468 China Datang Cor 25,299 342 Power Generation

poration
Source: Fortune Global 500 (2018).
Page 28 of 47
However, China’s national champions overall are still far from catching up with the global
leading big businesses in terms of international competitiveness. Most of China’s national
champions’ sales and profits come from China’s domestic markets where they still enjoy
varying degrees of protection by the state’s industrial policies against competitive pres
sures from both foreign multinational companies and domestic (p. 736) (p. 737) (p. 738)
challengers. In particular, they benefit from government procurements and buying prod
ucts from each other. The profitability of China’s national team is highly concentrated and
sensitive to government regulatory and price policies (Nolan, 2012; Naughton, 2015; Li,
2015). Apart from their remarkable success in building infrastructure in developing coun
tries, the export competitiveness of China’s national champions remains limited. In terms
of foreign investment, despite being supported by the ‘going out’ strategy of China’s cen
tral government, China’s national champions are still at an early stage of international ex
pansion. In 2016, there were only two Mainland Chinese firms (both are national champi
ons under SASAC) in the world’s top 100 non-financial multinational enterprises ranked
by foreign assets: CNOOC and COSCO. CNOOC had a total of US$66.7 billion foreign as
sets, less than 20 per cent of Royal Dutch Shell’s foreign assets. In 2016, China’s total
outward FDI stock was less than the combined foreign assets of five top multinational
companies from the advanced countries (Royal Dutch Shell, Toyota, BP, Total, and An
heuser-Busch InBev) (UNCTAD, 2017). Moreover, China’s national champions face
tremendous barriers in conducting major cross-border mergers and acquisitions. Their at
tempts to gain control over established firms in high-income countries have often ended
in failure. Even their efforts to acquire substantial minority stakes have encountered for
midable political and regulatory resistance from high-income countries (Nolan, 2012;
Naughton, 2015; Wu, 2016; Gordon and Milhaupt, 2019). While China’s five largest com
mercial banks have expanded their international operations through a combination of or
ganic growth and acquisitions, none of them entered the world’s top fifteen transnational
financial corporations ranked by the extent of their geographical spread in 2017 (UNC
TAD, 2018).
As successors of former central industrial ministries, China’s national champions have

long concentrated most of China’s indigenous technological capabilities in their respec
tive strategic sectors. They have achieved remarkable technological progress in indus
tries such as aerospace and defence, oil and gas, high-speed railway, telecoms, and elec
tronics as well as power generation and related equipment. However, the gap in R&D ca
pabilities between most of China’s national champions and the global leading big busi
nesses in their respective sectors remains huge. According to the EU’s 2018 Industrial
R&D Scoreboard, the world’s top 2,500 companies ranked by R&D expenditure together
invested around €736.4 billion, approximately 90 per cent of the world’s business-funded
R&D in 2017/2018. They include 778 US companies for 37 per cent of the total invest
ment, 577 EU companies for 27 per cent, 339 Japanese companies for around 14 per
cent, 438 Chinese companies for 10 per cent and 368 from the rest of the world for 12
per cent (European Commission, 2018). Among the 100 largest Chinese companies
ranked by R&D expenditure, twenty are China’s central state-controlled national champi
ons (or their member firms), which together invested a total of around €14.9 billion in
Page 29 of 47
R&D, compared with €13.4 billion by Samsung alone and €13.1 billion by Volkswagen.
Nonetheless, these twenty national champions contributed around 21 per cent of the total
R&D invested by the 438 Chinese companies ranked among the global top 2,500. As
shown in Table 25.2, the R&D intensity of China’s national champions is generally low
and the investment is heavily concentrated in oil and gas, (p. 739) construction and mate
rials, and industrial engineering (primarily infrastructure building and related equip
ment).
Following China’s entry to the WTO, the Chinese government has actively maintained its
industrial policies to secure technology transfers from global leading big businesses to its
national champions. The results of this strategy have been mixed. In some sectors, no
table success has been achieved: for example, in the high-speed railway and metro sector,
China’s national champions have rapidly absorbed technologies from leading multination
als such as Alstom, Siemens, and Kawasaki Heavy Industries, and have produced substan
tial indigenous innovation. In other sectors such as automobiles, the strategy has been far
less successful, if not a failure. For example, China’s national champions in the automo
bile sector are now heavily dependent upon the technologies and brands of global leading
automobile companies that they have partnered with in joint ventures: FAW relies heavily
upon Volkswagen and Toyota, and Dongfeng relies heavily upon Nissan and Peugeot SA
(Lu and Feng, 2005; Nolan, 2012; Li, 2015). Overall, despite their dominant sizes in
China’s domestic markets, China’s national champions are still at an early stage of devel
oping their international competitiveness in terms of building up the necessary capabili
ties, technologies, brands, and global production networks to compete globally, especially
in the markets of high-income countries.
Page 30 of 47
Table 25.2 National Champions in the top 100 companies ranked by R&D in mainland China
World Compa Indus R&D Net R&D Capex Capex Operat Em
rank ny try 2017/1 sales intensi (€mn) intensi ing ployees
8 (€mn) ty (%) ty (%) profits
(€mn) (€mn)
86 China Con 1,585.9 1,32,85 1.2 2,230.4 1.7 9,053.5 2,70,46

State struc 7.8 7
Con tion &
struc Materi
tion En als
gineer
ing Cor
poration
88 PetroCh Oil & 1,578.0 2,58,13 0.6 29,354. 11.4 9,304.3 4,94,29
ina Gas 7.7 4 7
Produc
ers
99 China Con 1,421.8 88,198. 1.6 1,677.5 1.9 3,945.1 2,83,63

Railway struc 4 7
tion &
Materi
als
Page 31 of 47
102 CRRC Indus 1,343.5 26,512. 5.1 1,009.4 3.8 1,984.4 1,76,75
trial En 3 4
gineer
ing
105 China Con 1,331.4 86,183. 1.5 3,871.1 4.5 3,508.7 2,61,33
Railway struc 2 3
Con tion &
struc Materi
tion als
Corpo
ration
126 China Con 1,104.9 61,538. 1.8 1,692.8 2.8 3,871.6 1,16,89
Commu struc 5 3
nica tion &
tions Materi
Con als
struc
tion
Compa
ny
Page 32 of 47
144 Power Con 972.7 33,518. 2.9 9,365.5 27.9 2,259.1 1,31,09
Con struc 1 1
struc tion &
tion Materi
Corpo als
ration
of China
166 Sinopec Oil & 822.5 3,02,22 0.3 8,136.5 2.7 12,198. 4,46,22
Gas 6.2 6 5
Produc
ers
200 Metal General 683.3 30,848. 2.2 522.4 1.7 1,887.5 97,771
lurgical Indus 6
Corpo trials
ration
of China
219 China Banks 607.1 28,274. 2.1 1,575.7 5.6 11,611. 72,530
Mer 4 7
chants
Bank
237 Dongfe Automo 546.7 16,008. 3.4 418.6 2.6 4.0 1,46,84
ng Mo biles & 5 3
tor Parts
Page 33 of 47
251 Baosha Indus 506.4 36,619. 1.4 1,700.1 4.6 3,320.5 57,154
n Iron & trial 6
Steel Metals
& Min
ing
271 China Con 447.6 30,011. 1.5 538.4 1.8 1,776.7 1,30,29
Energy struc 4 5
Engi tion &
neering Materi
Group als
290 Chongqi Automo 413.6 9,658.6 4.3 423.7 4.4 −29.8 39,138
ng biles &
Changa Parts
n
303 China Indus 392.2 4,835.8 8.1 186.3 3.9 145.7 36,315
Ship trial En
building gineer
ing
465 AviChi Indus 237.8 4,174.1 5.7 245.8 5.9 353.9 49,672
na In trial
dustry Trans
& Tech porta
nology tion
Page 34 of 47
475 China Con 231.4 7,419.0 3.1 195.6 2.6 505.1 41,588
Nation struc
al tion &
Chemi Materi
cal En als
gineer
ing
Group
496 CNOOC Oil & 222.7 23,867. 0.9 6,112.4 25.6 5,928.3 19,030
Gas 5
Produc
ers
501 China Con 221.0 13,409. 1.6 1,095.7 8.2 1,032.5 41,241
Gezhou struc 3
ba tion &
Group Materi
als
537 NARI Soft 205.7 3,048.3 6.7 153.1 5.0 556.3 5,865
Technol ware &
ogy De Com
velop puter
ment Ser
vices
Page 35 of 47
Source: The EU Industrial R&D Investment Scoreboard (2018) and company reports.
Page 36 of 47

China’s national champions industrial policy practices reflect its unorthodox approach to
reforming the former socialist command economy. It also represents China’s adaptive re
sponses to the challenges of late industrialization in a new global business environment.
The ‘commanding heights’ sectors are typically characterized by high capital intensity,
technology complexity, asset specificity, and substantial economies of scale. Developing
countries face tremendous barriers to build viable indigenous firms with sufficient capa
bilities and sizes to compete with incumbents from advanced countries in the same sec
tors. The institutional limitations of the socialist command economy were not only about
the absence of properly functioning markets, but also the absence of viable modern big-
business firms. It should not be assumed that the deficiencies of firm-level organizational
capabilities in transition economies and developing countries can be spontaneously re
solved by marketization.
Moreover, the industrial governance of transition economies comprises complex inter-or

ganizational linkages which, in the old socialist command economy structure, connected
various government agencies, enterprise units, and financial organizations that were a
reservoir of talent, knowledge, and experience. There are substantial complementarities
and valuable resources in these organizations. Instead of wasting existing resources and
capabilities, the rise of China’s national champions suggests that it is possible to inte
grate and restructure the core assets and enterprise units inherited (p. 740) (p. 741)
(p. 742) from the command economy into viable big-business firms as organizational de
vices to coordinate indigenous industrial development in a predominantly market-orient

ed economic environment. Effective industrial reform should recognize and take advan
tage of such institutional legacies.
Under appropriate industrial policies, the danger of ‘ossification’ can be turned into a
spur to new adaptation and growth. The key is, first, to promote continuous organization
al learning and capability building at both the government and enterprise levels, strategi
cally coordinating complementary reforms. Second, it is important to establish workable
governance arrangements that adapt to a country’s own institutional conditions to facili
tate learning and to discipline the recipients of state policy support. In particular, the
state can innovatively use corporatization, commercial and policy banks, capital markets,
and other financial means as industrial policy tools. The rise of China’s national champi
ons is the result of its adaptive policy responses to its own specific political and economic
conditions as well as the changing global business environment. Taking advantage of its
large and continuously growing domestic markets, China’s industrial policy has created
strategic space and opportunities for its national champions to acquire new capabilities
and achieve remarkable growth. However, they are still at an early stage of developing in
ternational competitiveness in terms of building their global footprints and capabilities.
Page 37 of 47
References
Amsden, Alice H. (2001) The Rise of ‘The Rest’: Challenges to the West from Late-Indus
trializing Economies. Oxford: Oxford University Press.
Amsden, Alice H. and Ajit Singh (1994) ‘The Optimal Degree of Competition and Dynamic
Efficiency in Japan and Korea’, MPRA Paper 38(3–4): 941–51.
Blanchard, Olivier, Rudiger Dornbusch, Paul Krugman, Richard Layard, and Lawrence
Summers (1991) Reform in Eastern Europe. Cambridge, MA: MIT Press.
Bremmer, Ian (2009) ‘State Capitalism Comes of Age—the End of the Free Market’, For
eign Affairs 88: 40.
Brødsgaard, Kjeld Erik (2012) ‘Politics and Business Group Formation in China: The Party
in Control?’, The China Quarterly 211: 624–48.
Buckley, Chris (2005) ‘China’s “Nanny” Overhauls State Enterprises’, The New York
Times, 1 June.
Chandler, Alfred D. (1977) The Visible Hand: Managerial Revolution in American Busi
ness. Cambridge, MA: Belknap Press of Harvard University Press.
Chandler, Alfred D. (1990) Scale and Scope: The Dynamics of Industrial Capitalism. Cam
Chandler, Alfred D., Franco Amatori, and Takashi Hikino (1997) Big Business and the
Wealth of Nations. Cambridge, MA: Cambridge University Press.
Chang, Ha-Joon (1994) The Political Economy of Industrial Policy. New York St. Martin’s
Press. (p. 743)
Colpan, Asli M., Takashi Hikino, and James R. Lincoln (2010) The Oxford Handbook of
Business Groups. Oxford: Oxford University Press.
Downs, Erica Strecker (2011) Inside China, Inc.: China Development Bank’s Cross-border
Energy Deals. Washington, DC: The John L. Thornton China Center at Brookings.
European Commission (2018) The 2018 EU Industrial R&D Investment Scoreboard. Lux
embourg: Publications Office of the European Union.
Fortune (2018) ‘Global 500’. Available at http://money.cnn.com/magazines/fortune/

global500/index.html.
Gerschenkron, Alexander (1962) Economic Backwardness in Historical Perspective: A

Book of Essays. Cambridge, MA: Belknap Press of Harvard University Press.
Page 38 of 47
Gordon, Jeffrey N. and Curtis J. Milhaupt (2019) ‘China as a National Strategic Buyer: To
ward a Multilateral Regime for Cross-border M&A’, Columbia Business Law Review 1:
192.
Hahn, Donghoon and Keun Lee (2006) ‘Chinese Business Groups: Their Origins and De
velopment’, in Sea-Jin Chang (ed.) Business Groups in East Asia: Financial Crisis, Restruc
turing, and New Growth. Oxford: Oxford University Press, pp. 207–31.
Harrison, Mark and Andrei Markevich (2008) ‘Hierarchies and Markets: The Defense In
dustry under Stalin’, in Mark Harrison (ed.) Guns and Rubles: The Defense Industry in the
Stalinist State. New Haven, CT: Yale University Press, pp. 50–77.
Heilmann, Sebastian (2005) ‘Regulatory Innovation by Leninist Means: Communist Party

Supervision in China’s Financial Industry’, The China Quarterly 181: 1–21.
Heilmann, Sebastian and Lea Shih (2013) ‘The Rise of Industrial Policy in China, 1978–
2012’. Available at http://www.harvard-yenching.org/features/hyi-working-paper-se
ries-sebastian-heilmann-and-lea-shih.
Huang, Yasheng (2008) Capitalism with Chinese Characteristics: Entrepreneurship and

the State. Cambridge: Cambridge University Press.
Huang, Yasheng (2011) ‘Rethinking the Beijing Consensus’, Asia Policy 11: 1–26.
IIECASS (Institute of Industrial Economics of Chinese Academy of Social Sciences) (1998)

Zhongguo Gongye Fazhan Baogao (China Industrial Development Report). Beijing: Econo
my & Management Publisher.
Jiang, Jianqing (2019) Zhongguo Daxing Shangye Yinghang Gugaishi (History of China’s
Large Commercial Bank Shareholding Reforms). Beijing: China Finance Publisher.
Jiang, Julie and Chen Ding (2014) Update on Overseas Investments by China’s National
Oil Companies. Achievements and Challenges since 2011. Paris: International Energy
Agency.
Jiang, Julie and Jonathan Sinton (2011) Overseas Investments by China’s National Oil
Companies: Assessing the Drivers and Impacts. Paris: OECD/International Energy
Agency.
cy: 1925–1975. Stanford, CA: Stanford University Press.
Jung, Joo-Youn (2011) ‘Reinvented Intervention: The Chinese Central State and State-
owned Enterprise Reform in the WTO Era’, in Jean Chun Oi (ed.) The Politics of Corporate
Restructuring and System Reform: Going Private in China. Stanford, CA: Walter H.
Shorenstein Asia-Pacific Research Center, pp. 119–34.
Page 39 of 47
Kasza, Gregory J. (2018) ‘Gerschenkron, Amsden, and Japan: The State in Late Develop
ment’, Japanese Journal of Political Science 19(2): 146–72.
Khanna, Tarun and Yishay Yafeh (2007) ‘Business Groups in Emerging Markets: Paragons
or Parasites?’, Journal of Economic Literature 45(2): 331–72. (p. 744)
Komiya, Ryutaro (1987a) ‘Japanese Firms, Chinese Firms: Problems for Economic Reform
in China, Part I’, Journal of the Japanese and International Economies 1(1): 31–61.
Komiya, Ryutaro (1987b) ‘Japanese Firms, Chinese Firms: Problems for Economic Reform
in China, Part II’, Journal of the Japanese and International Economies 1(2): 229–47.
Lardy, Nicholas (2019) The State Strikes Back: The End of Economic Reform in China?
Washington, DC: Peterson Institute for International Economics.
Lee, Hong Yung (1991) From Revolutionary Cadres to Party Technocrats in Socialist Chi
na. Berkeley, CA: University of California Press.
Lee, Keun (2006) ‘Business Groups as an Organizational Device for Catch-up’, in Junji
Nakagawa (ed.) Managing Development: Globalization, Economic Restructuring and So
cial Policy. London: Routledge, pp. 217–33.
Lee, Keun (2010) ‘Business Groups in China’, in Asli M. Colpan, Takashi Hikino, and
James R. Lincoln (eds) The Oxford Handbook of Business Groups. Oxford: Oxford Univer
sity Press, pp. 210–36.
Lee, Keun (2019) The Art of Economic Catch-up: Barriers, Detours and Leapfrogging in
Lee, Keun and Xuehua Jin (2009) ‘The Origins of Business Groups in China: An Empirical
Testing of the Three Paths and the Three Theories’, Business History 51(1): 77–99.
Lee, Keun, and Wing T. Woo (2002) ‘Business Groups in China: Compared with Korean
Chaebols’, in The Post-Financial Crisis Challenges for Asian Industrialization, edited by R.
Hooley and J. Yoo, 721–47. New York, JAI Press.
Lee, Keun, Byung-Yeon Kim, Young-Yoon Park, and Elias Sanidas (2013) ‘Big Businesses
and Economic Growth: Identifying a Binding Constraint for Growth with Country Panel
Analysis’, Journal of Comparative Economics 41(2): 561–82.
Leutert, Wendy (2018a) ‘Firm Control: Governing the State-owned Economy under Xi Jin
ping’, China Perspectives 1–2: 27–36.
Leutert, Wendy (2018b) ‘The Political Mobility of China’s Central State-owned Enterprise
Leaders’, The China Quarterly 233: 1–21.
Li, Chen (2015) China’s Centralized Industrial Order: Industrial Reform and the Rise of
Centrally Controlled Big Business. London: Routledge.
Page 40 of 47
Li, Chen (2016) ‘Holding “China Inc.” Together: The CCP and the Rise of China’s Yangqi’,
The China Quarterly 228: 927–49.
Li, Chen (2019) ‘China’s Central State Corporatism: The Party and the Governance of
Centrally Controlled Businesses’, in Lance L. P. Gore and Yongnian Zheng (eds) The Chi
nese Communist Party in Action: Consolidating Party Rule. London: Routledge, pp. 220–
39.
Li, Cheng (2001) China’s Leaders: The New Generation. Lanham, MD: Rowman & Little
field Publishers.
Li, Cheng (2011) China’s Midterm Jockeying: Gearing up for 2012 (Part 4: Top Leaders of
Major State-Owned Enterprises). Washington, DC: Brookings Institution.
Liebman, Benjamin L. and Curtis J. Milhaupt (2015) Regulating the Visible Hand? The In
stitutional Implications of Chinese State Capitalism. Oxford: Oxford University Press.
Lin, Justin Yifu and Ha‐Joon Chang (2009) ‘Should Industrial Policy in Developing Coun
tries Conform to Comparative Advantage or Defy It? A Debate between Justin Lin and Ha‐
Joon Chang’, Development Policy Review 27(5): 483–502.
Lin, Justin Yifu, Cai Fang, and Li Zhou (2003) The China Miracle: Development Strategy
and Economic Reform. Hong Kong: Chinese University of Hong Kong Press. (p. 745)
Lin, Li-Wen (2017) ‘Reforming China’s State-owned Enterprises: From Structure to Peo
ple’, The China Quarterly 229: 107–29.
Lin, Li-Wen and Curtis J. Milhaupt (2013) ‘We Are the (National) Champions: Understand
ing the Mechanisms of State Capitalism in China’, Stanford Law Review 65: 697–759.
Lin, Nan (2011) ‘Capitalism in China: A Centrally Managed Capitalism (CMC) and its Fu
ture’, Management and Organization Review 7(1): 63–96.
Lipton, David and Jeffrey Sachs (1990) ‘Creating a Market Economy in Eastern Europe:
The Case of Poland’. Brookings Papers on Economic Activity. No. 1: 75–133. Washington,
DC: Brookings Institution.
Liu, Chunhang (2012) Multinationals, Globalisation and Indigenous Firms in China. Lon
don: Routledge.
Lu, Feng and Kaidong Feng (2005) Fazhan Woguo Zizhu Zhishi Chanquan Qiche Gongye
De Zhengce Xuanze (The Policy Choice of Developing Chinese Automobile Industry with
Independent Intellectual Property Rights). Beijing: Peking University Press.
Matthews, John (2002) ‘Competitive Advantages of the Latecomer Firm: A Resource-

based Account of Industrial Catch-up Strategies’, Asia Pacific Journal of Management 19:
467–88.
Page 41 of 47
Milhaupt, Curtis J. (2017) ‘Chinese Corporate Capitalism in Comparative Context’, in

Weitseng Chen (ed.) The Beijing Consensus? How China Has Changed the Western Ideas
of Law and Economic Development. Cambridge: Cambridge University Press, pp. 275–
300.
Milhaupt, Curtis J. and Zheng Wentong (2015) ‘Beyond Ownership: State Capitalism and
the Chinese Firm’, Georgetown Law Journal 103: 665–722.
Moore, Thomas G. (2002) China in the World Market: Chinese Industry and International
Sources of Reform in the Post-Mao Era. Cambridge: Cambridge University Press.
Naughton, Barry (2005) ‘SASAC Rising’, China Leadership Monitor 14(7). Available at
http://media.hoover.org/sites/default/files/documents/clm14_bn.pdf.
Naughton, Barry (2015) ‘The Transformation of the State Sector: SASAC, the Market
Economy and the New National Champions’, in Barry Naughton and Kellee S. Tsai (eds)
State Capitalism, Institutional Adaptation and the Chinese Miracle. New York: Cambridge
Naughton, Barry and Kellee S. Tsai (2015) State Capitalism, Institutional Adaptation and
the Chinese Miracle. New York: Cambridge University Press.
Nolan, Peter (1995) China’s Rise, Russia’s Fall: Politics, Economics and Planning in the
Transition from Stalinism. Basingstoke: Macmillan.
Nolan, Peter (1996) ‘Large Firms and Industrial Reform in Former Planned Economies:
The Case of China’, Cambridge Journal of Economics 20(1): 1–29.
Nolan, Peter (2001a) China and the Global Business Revolution. London: Palgrave.
Nolan, Peter (2001b) China and the Global Economy: National Champions, Industrial Poli
cy and the Big Business Revolution. London: Palgrave.
Nolan, Peter (2012) Is China Buying the World? Cambridge: Polity Press.
Nolan, Peter (2014) Chinese Firms, Global Firms: Industrial Policy in the Era of Globalisa
tion. London: Routledge.
Nolan, Peter (2015) ‘The West and China: Globalisation and Competition in Financial Ser
vices’, Journal of Chinese Economic and Business Studies 13(2): 87–104. (p. 746)
Nolan, Peter and Huaichuan Rui (2004) ‘Industrial Policy and Global Big Business Revolu
tion: The Case of the Chinese Coal Industry’, Journal of Chinese Economic and Business
Studies 2(2): 97–113.
Nolan, Peter and Huaichuan Rui (2007) ‘The Global Industrial Consolidation and the Chal
lenge for China: The Case of the Steel Industry’, in Peter Nolan (ed.) Integrating China:
Towards the Coordinated Market Economy. London: Anthem Press, pp. 71–94.
Page 42 of 47
Nolan, Peter and Xiaoqiang Wang (1999) ‘Beyond Privatization: Institutional Innovation
and Growth in China’s Large State-owned Enterprises’, World Development 27(1): 169–
200.
Nolan, Peter, Jin Zhang, and Liu Chunhang (2007) The Global Business Revolution and the
Cascade Effect: Systems Integration in the Aerospace, Beverages and Retail Industries.
Basingstoke: Palgrave Macmillan.
Nove, Alec (1977) The Soviet Economic System. London: Allen & Unwin.
Olson, Mancur (1982) The Rise and Decline of Nations: Economic Growth, Stagflation,
and Social Rigidities. New Haven, CT: Yale University Press.
Pan, Gongsheng (2012) Dahangdiebian: Zhongguo Daxing Yinhang Fuxing Zhilu (Meta
morphose: The Rejuvenation of Chinese Big Banks). Beijing: China Financial Publisher.
Pearson, Margaret (2005) ‘The Business of Governing Business in China: Institutions and
Norms of the Emerging Regulatory State’, World Politics 57(2): 296–322.
Pearson, Margaret (2015) ‘State-owned Business and Party–State Regulation in China’s

Modern Political Economy’, in Barry Naughton and Kellee S. Tsai (eds) State Capitalism,
Institutional Adaptation, and the Chinese Miracle. New York: Cambridge University Press,
pp. 27–45.
Press.
Pistor, Katharina (2012) ‘The Governance of China’s Finance’, in Joseph Fan, Randall
Morck, and Bernard Yeung (eds) Capitalizing China. Chicago, IL: University of Chicago
Press, pp. 35–60.
Pistor, Katharina, Guo Li, and Zhou Chun (2016) ‘The Hybridization of China’s Financial
System’, in Benjamin L. Liebman and Curtis J. Milhaupt (eds) Regulating the Visible
Hand? The Institutional Implications of Chinese State Capitalism. Oxford: Oxford Univer
Popov, Vladimir (2019) ‘Successes and Failures of Industrial Policy: Lessons from Transi
tion (Post-Communist) Economies of Europe and Asia’. Available at https://mpra.ub.uni-
muenchen.de/id/eprint/95332.
Qian, Yingyi (1999) ‘The Institutional Foundations of China’s Market Transition’. Annual
World Bank Conference on Development Economics, Working Paper No. 47582, Washing
ton, DC: World Bank. Available at https://pdfs.semanticscholar.org/
7611/8aa97e37b0fb73f7c7a22bf2c62331eea92e.pdf.
Qian, Yingyi (2000) ‘The Process of China’s Market Transition (1978–98): The Evolution
ary, Historical, and Comparative Perspectives’, Journal of Institutional and Theoretical
Economics 156(1): 151–71.
Page 43 of 47
Qian, Yingyi (2001) ‘Government Control in Corporate Governance as a Transitional Insti

tution: Lessons from China’, in Joseph E. Stiglitz and Shahid Yusuf (eds) Rethinking the
East Asian Miracle. Washington, DC: World Bank, pp. 295–322.
Qian, Yingyi and Wu Jinglian (2000) ‘China’s Transition to a Market Economy: How Far
across the River?’. SCID Working Paper No. 69. Stanford, CA: Stanford Institute for Eco
nomic Policy Research. (p. 747)
Rui, Huaichuan (2005) Globalization, Transition and Development in China: The Case of
the Coal Industry. London: Routledge.
Sachs, Jeffrey (1991) ‘Poland’s Big Bang: A First Report Card’, The International Economy
5(1): 40–3.
SASAC (2004) Zhongguo Guoyou Zichan Jiandu Guanli Nianjian (China State-Asset Super
vision and Administration Yearbook). Beijing: China Economy Publisher.
SASAC (2009) Guowuyuan Guoziwei 2008 Nian Huigu (the State Council SASAC’s 2008
Review). Beijing: China Financial & Economic Publishing House.
Shmelev, Nikolai and Vladimir Popov (1989) The Turning Point: Revitalizing the Soviet
Economy. New York: Doubleday.
Steinfeld, Edward S. (2004) ‘China’s Shallow Integration: Networked Production and the
New Challenges for Late Industrialization’, World Development 32(11): 1971–87.
Steinfeld, Edward S. (2010) Playing our Game: Why China’s Economic Rise Doesn’t
Threaten the West. Oxford: Oxford University Press.
Stiglitz, Joseph E. (1996) ‘Some Lessons from the East Asian Miracle’, The World Bank
Research Observer 11(2): 151–77.
Sun, Pei (2005) ‘Industrial Policy, Corporate Governance, and the Competitiveness of
China’s National Champions: The Case of Shanghai Baosteel Group’, Journal of Chinese
Economic and Business Studies 3(2): 173–92.
Sun, Pei (2007) ‘Is the State-led Industrial Restructuring Effective in Transition China?
Evidence from the Steel Sector’, Cambridge Journal of Economics 31(4): 601–24.
Sutherland, Dylan (2003) China’s Large Enterprises and the Challenge of Late Industriali
sation. London: Routledge.
Page 44 of 47
Szamosszegi, Andrew and Cole Kyle (2011) An Analysis of State-owned Enterprises and
State Capitalism in China. Washington, DC: US–China Economic and Security Review
Commission.
The State Council of China (1982) ‘Guowuyuan Guanyu Chengli Zhongguo Chuanbo
Gongye Zonggongsi De Tongzhi (the State Council’s Notice on the Establishment of China
Shipbuilding Industry Corporation)’. Available at http://www.gov.cn/gongbao/shuju/
1982/gwyb198210.pdf.
Thun, Eric (2006) Changing Lanes in China: Foreign Direct Investment, Local Govern
ments and Auto Sector Development. Cambridge: Cambridge University Press.
Tian, Guoqiang and Liang Hong (1999) ‘What Kind of Privatization?’, in Andrew J. Nathan
(ed.) Dilemmas of Reform in Jiang Zemin’s China. London: Lynne Rienner, pp. 114–34.
UNCTAD (2017) World Investment Report. Geneva: United Nations Conference on Trade
and Development.
UNCTAD (2018) World Investment Report. Geneva: United Nations Conference on Trade
and Development.
Walter, Carl E. and Fraser J. T. Howie (2006) Privatizing China: Inside China’s Stock Mar
kets. Singapore: John Wiley & Sons.
Walter, Carl E. and Fraser J. T. Howie (2011) Red Capitalism: The Fragile Financial Foun
dation of China’s Extraordinary Rise. Singapore: John Wiley & Sons. (p. 748)
Wang, Haibo (1998) Zhonghua Renmin Gongheguo Gongye Jingjishi 1949.10–1998 (the
Industrial Economic History of the People’s Republic of China, 1949.10–1998). Taiyuan:
Shanxi Economy Publisher.
Wang, Yingyao (2015) ‘The Rise of the “Shareholding State”: Financialization of Economic
Management in China’, Socio-Economic Review 13(3): 603–25.
World Bank (1993) The East Asian Miracle: Economic Growth and Public Policy. New
World Bank (1997) China 2020. Washington, DC: World Bank.
World Bank (2012) China 2030: Building a Modern, Harmonious and Creative High-In
come Society. Washington, DC: World Bank.
Wu, Jinglian (2005) Understanding and Interpreting Chinese Economic Reform. Singa
pore: Thomson.
Wu, Mark (2016) ‘The “China, Inc.” Challenge to Global Trade Governance’, Harvard In
ternational Law Journal 57: 1001–63.
Page 45 of 47
Yi, Gang (2009) Zhongguo Jinrong Gaige Sikaolu (On the Financial Reforms of China).
Beijing: The Commercial Press.
Zhang, Jin (2004) Catch-up and Competitiveness in China: The Case of Large Firms in the
Oil Industry. London: Routledge.
Zhang, Jin (2015) ‘China’s Oil Industry: Investment in Developing Countries’, in Carla P.
Freeman (ed.) Handbook on China and Developing Countries. Cheltenham: Edward Elgar,
pp. 287–317.
Zhang, Jin (2017) ‘Western Ideas of Corporate Governance: China’s Reform of Large
State-owned Enterprises’, in Malcolm Warner (ed.) The Diffusion of Western Economic
Ideas in East Asia. London: Routledge, pp. 360–73.
Zhang, Jin (2018) ‘State Structures, Business–State Relations, and Multinational Corpo
rate Behaviours: A Case Study of Chinese Multinational Oil Companies’, in Xiaoke Zhang
and Tianbiao Zhu (eds) Business, Government and Economic Institutions in China. Bas
Zheng, Yongnian (2004) Globalization and State Transformation in China. Cambridge:

Cambridge University Press.
Zheng, Yongnian (2010) The Chinese Communist Party as Organizational Emperor: Cul
ture, Reproduction and Transformation. London: Routledge.
Chen Li
Chen Li is an assistant professor at the Centre for China Studies, The Chinese Uni
versity of Hong Kong (CUHK). He is also an assistant professor (by courtesy) at
CUHK’s Lau Chor Tak Institute of Global Economics and Finance (IGEF). He has re
searched and written extensively on issues related to China’s economic development,
business environment, and public policies, such as China’s state-owned enterprise re
form, financial regulatory reform, government‒business relations, and industrial and
regional development. He received his PhD and MPhil in development studies from
the University of Cambridge and dual bachelor’s degrees in law and economics from
Peking University.
Muyang Chen
Muyang Chen is an assistant professor at the School of International Studies, Peking

University. Her research interests include infrastructure finance, development bank
ing, international development, and state‒market relations. She has been a visiting
scholar at the National Graduate Institute for Policy Studies (Japan) and a pre-doctor
al fellow at the Global Development Policy Center of Boston University. She received
her PhD from the University of Washington, an MA from University of California,
Berkeley, and BAs from Peking University and Waseda University.
Page 46 of 47
Page 47 of 47
Industrial Policies in the BRICS

Fernando Santiago
This chapter discusses industrial policy in the BRICS. It contributes to the literature in
three ways. First, by approaching the BRICS as a single entity, it documents the gradual
buy-in to the concept of BRICS by the participating countries, and their efforts at
strengthening collaboration, including on industrial development matters. Second, it cor
roborates that differences in individual development paths influence the expected contri
bution of each member to advancing a joint industrial development agenda, while it is
early to dismiss their ability to consolidate as a major player in global economic dynam
ics. Third, the BRICS response to the Fourth Industrial Revolution builds on their tradi
tional proactive stance around industrial policy, while their expected collective collabora
tion with third-party regions, particularly Africa, reflects cumulative interests at the indi
vidual country level.
Keywords: BRICS, integration, industrial policy, industrial development, Fourth Industrial Revolution, Africa
26.1 Introduction
THIS chapter discusses industrial policies in the BRICS (Brazil, Russia, India, China, and
South Africa). The discussion corroborates that the emergence of systems of innovation
and the dynamics of industrial development and structural change in BRICS cannot be ex
plained without considering the active role of national governments, which have support
ed specific sectors, industries, and even firms to grow from national champions to major
global manufacturing and technological players (Vértesy, 2011; Scerri and Lastres, 2013;
Kahn, Martins de Melo, and Pessoa de Matos, 2014). This active industrial policy ap
proach is likely to continue at both the individual and the collective level.
The chapter moves on from previous contributions to the literature, where the notion of
BRICS is used as a reference to analyse individual country experiences, and to draw com
mon elements and the main differences in policy models followed by BRICS (Di Maio,
2015; Brigante Deorsola et al., 2017). Instead, the chapter interrogates how the BRICS
notion has been endorsed by the countries involved, to the extent that they are undertak
Page 1 of 34
ing efforts to strengthen integration and collaboration. Industrial development in general,

and industrial policy in particular, are relatively recent, yet already integral topics in the
BRICS integration process.
The analysis also sheds light on the extent to which the recent industrial performance of
each BRICS country influences their participation in group dynamics. Differences in in
dustrial development paths within BRICS suggest widening industrialization gaps, with
China increasingly differentiated from the rest. This is likely to raise tensions in the inte
gration process. Looking to the future, the chapter explores how the BRICS are respond
ing to the opportunities and challenges associated with the Fourth (p. 750) Industrial Rev
olution (4IR) (Schwab, 2016; Hallward-Driemeier and Nayyar, 2017). This new wave of
technological change has the potential to alter industrial leadership, affecting prospects
and conditions for catching up and forging ahead (Liu et al., 2017; Lee et al., 2019).
Finally, we briefly introduce an international dimension, looking into BRICS interaction

with third-party countries and regions. While each BRICS is a recognized economic and
political powerhouse in their respective geographical regions, Africa seems to constitute
a place of convergence for all of them.
26.2 The Tradition of Industrial Policy in the

BRICS
Common among the BRICS is that recent industrial performance has accompanied
processes of sustained capability accumulation and continuous policy reform to underpin
substantive economic and institutional transformations. These experiences lend support
to those who advocate active public interventions as the basis for industrialization, in
cluding conditions for resources mobilization, and more generally, for social and econom
ic agents to thrive.
Cimoli, Dosi, and Stiglitz (2009: 2) argue that as ‘intrinsic fundamental ingredients of all
development processes’, industrial policies encompass interventions across multiple poli
cy domains (Di Maio, 2015; Stiglitz, 2017):
• Science and technology and innovation—including intellectual property rights (IPRs);

• Trade policies;
• Policies to foster demand for manufacturing, including public procurement;
• Policies to attract foreign direct investments (FDI) and other financial resources;
• Policies to foster development of specific industries—including but not limited to the
promotion of infant industries;
• Competition policies and regulation, among others.
From the end of the Second World War onwards, several of these policy domains have
been integral components of each of the BRICS’ industrial policy programmes. In addi
Page 2 of 34
tion to direct contributions to the building of domestic systems of innovation (Scerri and
Lastres, 2013), BRICS governments have fostered the articulation of industrial policies to
gether with innovation, education, and other policies to underpin technological and pro
ductive capability building. Hence, interlinkages across distinct policy domains have been
instrumental for the industrial development trajectories of BRICS (Di Maio, 2015; Liu et
al., 2017; Dominguez Lacasa et al., 2019). Such interplay characterizes co-evolutionary
processes whereby investments in science, technology, (p. 751) and innovation (STI) capa
bilities on the one hand, and industrialization and structural change on the other hand,
have fed—or failed to feed—on each other historically (Nelson, 1994; Scerri and Lastres,
2013).
The scope of industrial policies has varied over time and across countries according to
changes in the prevailing needs and conditions of sectors, industries, or firms. External
influences on the role of government generally, and industrial policy in particular, are im
portant in fostering development. Periods of strong government intervention—import sub
stitution, for example—alternate with more liberal stances—inspired largely by the Wash
ington Consensus—towards the rule of markets in the economy.
BRICS have actively used trade policies in the early stages of industrialization. Regarding
FDI policies, Andreff (2015) documents that unlike Brazil and India, where outward FDI
mainly responds to economic purposes, China and Russia also target foreign policy, diplo
macy, and even state ideology goals. Similarly, the author comments that the larger share
of state-owned companies (SOCs) among Chinese and Russian multinationals means
tighter state controls over corporate strategic decisions.
A constant across the board is support to national champion firms, including through di
rect investment or through the creation of SOCs who then lead the development of specif
ic industrial sectors, encouraging them to become major global manufacturing and tech
nological players (Rodriguez-Arango and Gonzalez-Perez, 2016; Santiago, 2015; Liu et al.,
2017; Di Maio, 2015). Interventions to assist domestic champions to weather the effects
of global economic crises or shocks in relevant markets for their products or services (Si
machev et al., 2014; Vértesy, 2011), or mobilization of national banks and SOCs to lever
age private investment in strategic sectors or activities are also common (Di Maio, 2015;
Simachev et al., 2014).
BRICS support for champion firms remains debatable. Cui, Jiao, and Jiao (2016) argue
that in BRICS where government ownership is comparatively high, firms tend to show a
lower probability of engaging in innovation; they have strong incentives to capitalize on
monopoly positions and protection, and willingness to tackle emerging market opportuni
ties through entrepreneurship and innovation is low. Liu et al. (2017) illustrate this
through the case of innovations in the growing online payments business, which is often
associated with booming e-commerce in China. While none of the four largest state-
owned banks was willing to develop such systems, the Alibaba Group took the initiative to
develop Alipay, which has become a world-renowned model since its launch in 2004.
Page 3 of 34
Turning attention to individual country experiences, in the case of Brazil, Di Maio (2015)
asserts that, setting aside controversies over the results of the period of import substitu
tion (1950s to the 1980s), during those years the Brazilian government undertook great
commitments to enhance domestic technological capabilities and economic diversifica
tion. It actively guided the direction and pace of industrialization, often as a direct in
vestor and owner/manager of firms (Di Maio, 2015). However, after the collapse of the im
port-substitution model (see Ocampo and Porcile, Chapter 28 in this volume) during the
1990s, disappointment with active industrial policies led the Brazilian government,
(p. 752) like those of other Latin American countries, to endorse a programme of econom
ic reforms inspired by the Washington Consensus (Di Maio, 2015). The dismantling of
trade barriers, the widespread privatization of publicly owned assets, efforts to restruc
ture the economy, and a generalized opening of domestic markets to multinationals char
acterized this period. More recently, although the emphasis on structural change has fad
ed, the government’s involvement in industrial development and capability building con
tinues, with an emphasis on STI, support to small and medium-size enterprises (SMEs),
and the promotion of clusters and investment in specific sectors, for example, automo
biles (Di Maio, 2015).
Russia has experienced difficulties managing the transition towards a market-driven

economy, inspired mainly by the Washington Consensus and the corresponding hiatus in
the government’s ability to implement active industrial policy during the 1990s (Di Maio,
2015). One structural characteristic is the severe lack of transparency in state‒business
interactions and the capture of industrial policymaking by lobbyist and interest groups
with differing views on how to govern industrial policy implementation (Simachev et al.,
2014). In recent years, Di Maio (2015) notes a ‘slide to the past’, with a growing partici
pation of the Russian government in economic activities. State control of a large share of
productive assets is characteristic of the current economic model, heavily influenced by
an industrial policy oriented towards fostering import substitution, employment, and a se
lective and tightly controlled FDI policy. At the same time, Simachev et al. (2014) assert
that over the last decade or so, Russia has sought to harmonize industrial policy with STI
policy as the basis for economic diversification, development of technology-driven sec
tors, and continued growth; this is occurring amidst a somewhat adverse international en
vironment and a heavy burden imposed by political affairs at home and abroad. In 2017,
the Russian government introduced an ambitious digital economy programme, which in
tends to foster the institutional structure and infrastructure necessary for digitalization
and the use of data at a large scale to increase competitiveness, economic growth, and
sovereignty (Government of the Russian Federation, 2017).
Similar to the other BRICS, India’s rapid industrialization starts with an import-substitu
tion model adopted by the newly independent country in the early 1950s. Industrial devel
opment has been guided by five-year plans that give the Indian government great lever
age to intervene in economic matters, including by identifying areas suitable for state mo
nopoly and others where private investors could participate (Di Maio, 2015). The break
down of the import-substitution model and the ensuing reforms of the early 1990s grant
ed greater freedom to private investors and expanded the scope of partnerships with ex
Page 4 of 34
ternal economic agents by reducing controls on FDI, allowing foreign majority ownership,
and facilitating technology transfer. Liberalization and privatization of manufacturing are
tangible in the structure of Indian manufacturing today. The country’s specialization in in
formation and communication technologies (ICTs) benefited from accumulated invest
ments in education and research institutions, including several private-sector-run initia
tives, which also contribute to (p. 753) explaining the rapid expansion of services (Di Maio,
2015). Lee (2019a) argues that such expansion has been mostly at the expense of agricul
ture, while the share of manufacturing in total GDP remains constant. According to the
author, this process, led by three giants, Infosys, Tata Consultancy Services (TCS), and
Wipro, suggests that India has bypassed the stage of manufacturing-led growth,
leapfrogged into service-led growth, and then back to promote manufacturing (Lee,
2019a).
China’s outstanding industrial and economic performance coincides with an ambitious pe
riod of economic reforms that started during the 1970s. The country’s leadership commit
ted to giving greater weight to the markets, without renouncing active government influ
ence and the steering of the economic system. This pragmatic approach to industrial poli
cy has been dominant since the 1980s. China has supported structural transformation to
wards a market-driven economy, combined with heavily selective support, including
through FDI and compulsory—including coercive—technology transfer, to strategic activi
ties and sectors (Di Maio, 2015; Santiago, 2015). Investment in large infrastructure
projects is noticeable; the provision of physical and digital infrastructure has been a pri
ority, particularly over the last twenty to twenty-five years (Liu et al., 2017); and preferen
tial credit and fiscal treatment has been given particularly to manufacturing and non-agri
cultural raw materials. China is getting close to or is already at the end of a rapid catch
ing-up phase (Liu et al., 2017; Tourk and Marsh, 2016). She searches for the next policy
cycle, away from top-down approaches to industrial development, and increasingly under
the rule of markets, private entrepreneurship, and distinct institutional conditions for
economic agents to operate (Liu et al., 2017; OECD, 2017).
In South Africa, the democratic transition that started in 1994 immediately led to efforts
to reinsert the country into the international scene. The new leadership committed to dis
mantling industrial initiatives characteristic of the apartheid regime, including a large
privatization programme. In parallel, the government adopted the Black Empowerment
Programme to redress profound inequalities in access to skills, employment, and econom
ic opportunities for black South Africans. Accession to the WTO and the signing of free-
trade agreements with the EU and the Southern African Development Community (SADC)
signalled the government’s intention to promote internationalization of large domestic
firms, particularly in mining (Di Maio, 2015). The evidence suggests that the positive in
come effects associated with this modernization programme may have started to wear off,
as industrial competitiveness seems stagnant. Industrial policy measures also face diffi
culties boosting exports and diversifying manufacturing (Di Maio, 2015). In recent years,
a relevant change in the post-apartheid orientation of industrial policy has been the shift
Page 5 of 34
towards an emphasis on skills and innovation, and from a sectoral focus to functional ap
proaches (Di Maio, 2015).
Considering this heterogeneous past engagement with industrial policy, how do we ex
plain the recent emergence of BRICS as a novel political and economic entity in the glob
al landscape?
(p. 754) 26.3 To BRICS or not to BRICS

The literature, particularly from a political science perspective, documents debates on the
epistemological, methodological, and geopolitical underpinnings of BRICS, and the impli
cations of the term for the rest of the world, particularly least-developed regions such as
Africa. Considering the great heterogeneity characteristic of BRICS, is the term a valid
geopolitical configuration? What are the strengths and limitations of the study of BRICS
as an academic enterprise? Are BRICS a legitimate alternative to contemporary global,
West-dominated economic and political governance, or are they merely an extension of it
(Gray and Murphy, 2013; Gray and Gills, 2017; Mittelman, 2016)?
While addressing these debates is beyond the scope of this chapter, in what follows we of
fer food for thought from an industrial policy perspective. Notwithstanding efforts at the
highest possible political level towards a common BRICS development agenda, differ
ences in recent individual industrial performances have introduced tensions into the
BRICS integration process.
26.3.1 From BRIC to BRICS
From 2001, Jim O’Neill and colleagues at Goldman Sachs released several papers propos
ing the acronym BRIC to acknowledge the growing power and influence of the large and
dynamic economies of Brazil, Russia, India, and China over global political and economic
dynamics (O’Neill, 2001, 2018; O’Neill, Wilson, and Purushothaman, 2005). Wilson and
Purushothaman (2003) argued that given the pace of their GDP growth, income per capi
ta, and currency movements, the BRIC could become a major driver of the world econo
my by 2050. They could overtake several of the G6 economies—G7 minus Canada—in US
dollar terms by 2025 (Wilson and Purushothaman, 2003). The changed geography of the
ten largest economies in the world, in GDP terms, would result in a more complex and
more diversified scenario (Wilson and Purushothaman, 2003).
O’Neill (2001) advocated an ‘upgraded’1 G7—the United States, Japan, the United King
dom, Germany, France, Italy, and Canada—to include the BRIC, or at least some of them,
granting them a stronger say in global policymaking. While China appeared a natural can
didate, the potential influence of the other three was undeniable, on a par with Canada or
Italy. At the same time, O’Neill cautiously wondered whether the BRIC would want to in
tegrate and embrace their potential counterbalancing role in global economic and politi
cal affairs (O’Neill, 2001, 2018). After several informal meetings, at Russia’s initiative the
first formal meeting of the BRIC foreign ministers took place in Yekaterinburg, Russia on
Page 6 of 34
18 May 2008. The first summit of the BRIC heads of state (p. 755) followed in June 2009,
also in Yekaterinburg. The induction of South Africa in 2011 expanded the group to form
BRICS as we know it today (BRICS, 2011). Ever since, BRICS has sought to tighten col
laboration in mutually beneficial ways.
Today, BRICS cooperation mechanisms include the annual summits, which convene the
heads of state of each member country, the most recent of which took place in Brasilia in
November 2019. A diversity of meetings on specific topics involve representatives from
all sorts of public, private, and academic organizations and hierarchies. Meetings regular
ly sideline those of multilateral organizations such as the United Nations, the G20, or the
Bretton Woods Institutions. Several expert working groups contribute to a broader collab
oration agenda (BRICS, 2019).
26.3.2 BRICS to Build an Economic Block?
Economic collaboration among BRICS evolves gradually. Framed against the global finan
cial and economic crises, negotiations at the early leaders’ summits centred on financial,
trade, and sustainable growth issues (BRIC, 2009, 2010). As interactions intensified and
diversified, largely influenced by individual approaches to industrial and other related
policies, the agenda began to incorporate broader STI issues—including IPRs—employ
ment creation, connectivity and ICT, and a Framework for BRICS e-commerce Coopera
tion.
At the end of the Fortaleza Meeting in 2014, the BRICS leaders agreed to develop a
BRICS Economic Cooperation Strategy and a Framework of BRICS Closer Economic Part
nership (BRICS, 2014). The former was adopted at the Ufa Summit (BRICS, 2015a) and,
together with the BRICS Action Plan on Economic and Trade Cooperation (BRICS, 2015b,
2017a), guides actions intended to strengthen economic collaboration. Progress in the im
plementation of the Strategy for the BRICS Economic Partnership is reviewed every five
years, or earlier as necessary (BRICS 2015b, 2017a).
The BRICS Economic Cooperation Strategy sets out guidelines to increase, expand, and
promote intra-trade and investment, manufacturing and minerals processing, energy, STI,
connectivity, and ICT among other areas (BRICS, 2015b). The Action Plan calls for joint
economic- and trade-related initiatives, including on industrial and technical upgrading,
to foster economic complementarity and diversification (BRICS, 2017a). It also targets a
reduction in the digital divide and the associated economic and social implications both
within and beyond BRICS. Collaboration on e-commerce should contribute to industrial
development and inclusive growth in the BRICS and elsewhere (BRICS, 2017a).
A similarly incremental approach characterizes the BRICS’ search for collaboration on in
dustrial development and industrial policy. These topics were absent from the joint state
ments at the first two BRIC leaders’ summits. The term ‘industrialization’ appeared for
the first time in the Sanya Declaration of 2011, in reference to the BRICS’ interest in sup
Page 7 of 34
porting industrialization in Africa (BRICS, 2011). The associated (p. 756) action plan pro
posed cooperation in pharmaceuticals as part of STI collaboration (BRICS, 2011).
The term ‘manufacturing’ appeared in the Strategy for the BRICS Economic Partnership
at the 7th BRICS leaders’ meeting in 2015. The Strategy acknowledges the significant
contribution of manufacturing to structural change, the creation of quality jobs, and eco
nomic growth (BRICS, 2015b). The joint agenda proposes to expand manufacturing and
minerals processing (BRICS, 2015a) and to foster such industries as mining and metals,
chemicals and petrochemicals. BRICS committed to intensifying industrial production ca
pabilities and promoting industrial parks and clusters, technology parks, and engineering
centres; similarly, they intend to support specialized training for engineering and techni
cal personnel and managers, while encouraging innovation and the development of high-
tech industries (BRICS, 2015a, 2015b).
Investment in railways, roads, ports, and airports is also envisaged. At the 2018 BRICS
leaders’ summit in Johannesburg, they recognized energy efficiency as a factor condition
ing the achievement of economic targets such as industrial competitiveness, economic
growth and job creation, and environmental sustainability (BRICS, 2018). Advancing col
laboration in the ocean economy, including through coastal industrial zone development,
is also on the agenda (BRICS, 2018).
26.3.3 Industrial Policy Coordination
The launch of the BRICS Industry Ministers Meetings institutionalized collaboration on

industrial development and industrial policy matters. The first meeting took place in 2015
in Moscow, under the Russian presidency (BRICS Industry Ministers, 2015). Intra-BRICS
cooperation expects to boost trade and sustainable economic growth, strengthening com
prehensive industrial ties, promoting technology transfer and innovation, and improving
investment climates and job creation (BRICS Industry Ministers, 2015). Joint training and
skill development programmes are proposed, as are common research and business op
portunities in several broadly defined industrial areas (BRICS Industry Ministers, 2015).
The scope of collaboration was subsequently refined with the adoption of a seven-point
Action Plan at the end of the second ministerial meeting in Hangzhou, China (BRICS In
dustry Ministers, 2017). The Action Plan acknowledges the emergence of 4IR and two
other major transformations in global manufacturing: the increasing interdependence of
manufacturing and manufacturing-related services, with the latter increasingly driving
economic development; and the transformational power of concepts such as digitaliza
tion, networking, and ‘intellectualization’ and their significance for production and busi
ness models with the potential to create new industries (BRICS Industry Ministers, 2017).
The Action Plan reaffirms the BRICS commitment to enhance collaboration in industrial
capacity building, SME development, industrial policy coordination, standards, develop
ment of new industrial infrastructure, and technology- and innovation-related projects.
Page 8 of 34
Collaboration in industrial policy matters reaches out to the UN system, for exam
(p. 757)
ple, to the United Nations Industrial Development Organisation (UNIDO) in establishing

the BRICS Consolidated Technology Platform, with a mandate to boost cooperation in the
upgrading and/or development of new high-tech engineering industries and innovation,
and in the development of capital goods and machinery related to manufacturing and
minerals processing (BRICS, 2017c; BRICS Industry Ministers, 2017).
Consonant with the BRICS long-standing tradition of supporting large firms, the eThekwi
ni Declaration encourages existing SOCs to explore cooperation, exchange of informa
tion, and best practices (BRICS, 2013). Simultaneously, each country’s ministries and
agency responsible for local SMEs is exploring opportunities to collaborate, particularly
in international trade, innovation, and joint R&D (BRICS, 2013).
The BRICS approach to collaboration is consistent with individual traditions of creating

interdependence and complementarity between industrial policies and other policy areas.
This recurs in several of the declarations and action plans across different policy areas.
Common interests include technology transfer, support for joint research agendas, inten
tion to leverage on initiatives around training, skilling, and upskilling, firm registration,
and other crosscutting or horizontal policy interventions. Successful integration of differ
ent policy areas into a single BRICS industrial development programme should have ma
jor implications for manufacturing development in each individual BRICS.
26.4 The Challenge of Heterogeneous Industri

al Performances
The ability of the BRICS to overcome their distinct individual economic, political, and so
cial conditions and traditions, and to collectively contribute to global economic and politi
cal checks and balances is the subject of debate. Degaut and Meachan (2015), Di Maio
(2015), and Dominguez Lacasa et al. (2019) perceive BRICS as a heterogeneous collec
tion of industrial upgrading models, each of them with distinct perspectives and ap
proaches to the accumulation of technological, productive, and other capabilities. Howev
er, as we discuss later, China’s dominance in the group dynamic poses major challenges
to the BRICS integration. Kejin (2014) argues that with Chinese choices and capabilities
continuing to frame the scope of cooperation, attempts to enhance economic cooperation
and build consensus are encountering difficulties.
26.4.1 Contribution to Global Economic Activity
The BRICS’ growing global economic importance is evident (Figure 26.1). In 2017, they
represented 23.3 per cent of global GDP, a three-fold increase on 1990 (7.9 per cent).
(p. 758) Similarly, they contributed about a third of total global manufacturing value
added in 2017. Despite this enhanced collective presence, the individual shares are het
erogeneous. China largely drives the dynamics in terms of both total GDP and manufac
turing value added, although the rate of expansion has receded slightly in recent years.
Page 9 of 34
The most dramatic gains appear in manufacturing value added. In around a decade,
China’s contribution rose from about 10 per cent (2005) to about a quarter of the world
total in 2017. The shares of the rest of the BRICS are either growing less rapidly (India),
are stagnant (Russia and Brazil), or are shrinking (South Africa). India’s share in both
GDP and manufacturing value added has overtaken that of Brazil and Russia. Differences
in individual contributions to both group and global economic dynamics should continue
to deepen in the short run (Mbele, 2018).
Figure 26.1 BRICS’ growing but heterogeneous con

tribution to global economic dynamics
Source: Author, based on World Bank Development

Indicators.
The BRICS’ export performance over the period 2005‒15 further illustrates the hetero
geneity within the group. China reports the largest and fastest-growing share of manufac
turing in total exports. Brazil records a steady increase in the share of exports of agricul
tural and mining products and, to a lesser extent, services. In Russia and South Africa,
manufacturing exports are giving way to mining and services, while services account for
44 per cent of Indian exports.
The short-term dynamics should conform to Mbele (2018), who qualifies BRICS growth
trajectories as singling out distinct subgroups. China will continue to upgrade her posi
tion in global supply chains, consolidating as a major global manufacturing centre and
boosting her share in manufacturing exports within the BRICS and from the BRICS to the
world. Brazil, Russia, and South Africa, in turn, should consolidate as exporters of natural
resources-based products, while India continues to bounce between a manufacturing- and
a predominantly services-based economy (Amirapu and Subramanian, 2015). (p. 759)
26.4.2 Structural Change and Capability Accumulation
Industrial development implies structural change, or the transformation from traditional

agriculture or natural resource-based economies into economies led by increasingly com
plex manufacturing activities (Haraguchi, Cheng, and Smeets, 2017). There are three di
mensions to structural change: technological, industrial, and organizational. It involves
deliberate and decisive efforts to learn and accumulate different capabilities (UNIDO,
2002, 2005). Liu et al. (2017) argue that building innovation capabilities is a precondition
Page 10 of 34
to overcoming middle-income traps; they assert that the lack of capabilities, more than
political institutions, may be the most significant binding constraint for middle-income
countries.
Countries achieve productive and technological upgrading by balancing different frame

work conditions to support capabilities to absorb and use different kinds of knowledge, to
mobilize investments, and to accumulate the necessary capital to undertake structural
change, sustain catch-up, and eventually, leapfrog (Abramovitz, 1986). At least two kinds
of capability accumulation processes are possible (Lee, 2019b). First, production capabili
ties, which refers to the accumulation of advanced physical capital and the associated hu
man capital required to run productive facilities at given levels of efficiency. Second, in
novation capabilities assist economic agents to master and change the technology they al
ready use, eventually allowing them to create new technology. They also help latecomers
to close the gap relative to the technological frontier and, under certain conditions, seek
to overtake incumbents and assume leadership in furthering the technological frontier
(Lee and Malerba, 2017). By tapping into emerging market opportunities, through ex
ports for example, latecomers may reduce dependence on global value chains, steadily
capturing market shares, and increasing domestic value addition (Lee, 2019b).
However, because capability building is path dependent, it may constrain the ability to
identify and capture emerging windows of opportunity to further upgrading (Lee and
Malerba, 2017) and limit the flexibility to accommodate changes in technological and pro
ductive trajectories (Abramovitz, 1986). Globalization and international conventions
around IPRs also restrict the scope of technology transfer, and limit the possibilities of
imitating and reproducing technologies through R&D (Dominguez Lacasa et al., 2019;
Lee, 2019a). Latecomers should focus on developing small but incremental innovations
while avoiding confrontation with incumbents and their IPRs (Hobday, Rush, and Bessant,
2004).
26.4.2.1 Productive Capabilities

Compared to other developing countries, and thanks to a process broadly described as
economic modernization, BRICS productive competencies tend to feature greater techno
logical content and complexity relative to peer countries. Moreover, each of the BRICS
hosts several of the largest and most important multinational firms in the developing
world (Rodriguez-Arango and Gonzalez-Perez, 2016; Andreff, 2015). (p. 760) Several of
those firms are challenging, or have attained, global leadership in areas traditionally re
served for firms from highly industrialized countries, including aerospace, transport,
pharmaceuticals, and advanced ICT applications (Daudt and Willcox, 2016; Santiago,
2015).
However, modernization is insufficient to ensure long-term sustainability and growth

(Cimoli, Dosi, and Stiglitz, 2009). Notwithstanding progress in specific sectors or individ
ual firms, the BRICS’ economic structures are yet to achieve sustainable catch-up. They
face overexposure to external shocks and risk dependence on external technology suppli
Page 11 of 34
ers; moreover, differences in their endowments, technological progress, and integration

into global flows of trade and investment remain significant (Di Maio, 2015).
Figure 26.2 illustrates these points. It plots industrial and economic performance of each
of the BRICS over the last twenty to twenty-five years along two dimensions. First, the x-
axis plots long-term industrial performance, as measured by each individual country’s
score on UNIDO’s Competitive Industrial Performance Index (CIPIndex), which bench
marks a country’s ability to produce and export manufactured goods competitively (UNI
DO, 2019). The index helps to identify the occurrence of structural change towards high
value-added technology-intensive industrial sectors, and assess the impact of a country’s
industrial production on the world market. Second, the y-axis plots the value of gross na
tional income (GNI) per capita measured in constant US dollars of 2010.
Figure 26.2 BRICS: structural change and catching

up, 1990‒2015
Notes: Constant thousand US$ (2010=100). CIP In

dex: the earliest date available for CIP Index for Chi
na is 1995.
Source: Author’s based on UNIDO CIP database and

World Bank Development Indicators.
Heterogeneous performance within BRICS is evident. China stands out, as it has man
aged to combine and sustain structural change with a rapid expansion in GNI per capita.
According to CIPIndex scores, China’s increased industrial competitiveness was fuelled
mainly by a rapid expansion in her ability to generate manufacturing value (p. 761) added
and a strong export-led growth model that seems to have reached a tipping point (Tourk
and Marsh, 2016). Positive industrial performance has accompanied a five-fold increase in
GNI per capita in about two decades.
By contrast, the CIP Index scores of Russia, Brazil, and South Africa suggest processes of
gradual industrial recession; in the best-case scenario, these countries are stuck in a peri
od of industrial stagnation. Russia, in particular, is slowly recovering from the negative ef
fects on industrial competitiveness that represented the end of the Soviet era. While GNI
per capita continues to grow, albeit at a slower pace relative to China’s, the country’s CIP
Index score remained constant for most of the 1990‒2015 period. Similar conclusions can
apply to Brazil, although the descent in CIP Index rankings seems to have accelerated
during the 2010s. By contrast, South Africa’s gains in income per capita have seldom ac
companied improvements in its industrial competitiveness. This loss results from weak
Page 12 of 34
production skills and digital infrastructure, poor access to finance and high energy costs,
while firms that have addressed such challenges have continuously upgraded their capa
bilities and invested in innovative technologies (UNIDO, 2020). Finally, India records
much steadier growth in terms of both industrial competitiveness and GNI per capita, but
from a much lower base relative to South Africa, Brazil, or Russia. Overall, the data show
that relative to the other BRICS, China is rapidly closing the gap in terms of GNI per capi
ta while her relative distance in terms of industrial competitiveness is forging fast ahead.
26.4.2.2 Innovation Capabilities

Dominguez Lacasa et al. (2019) argue that innovation, broadly defined as a process that
includes but is not limited to R&D, is at the core of technological upgrading. While im
provements in infrastructure and the institutional environment have generally enhanced
innovation in the BRICS, particularly in China, performance remains modest relative to
developed-country standards (Franco and Oliveira, 2017). Innovation performance cor
roborates the very distinct development paths followed by individual BRICS.
Building on patent data to study technological upgrading in the BRICS over the period
1980‒2015, Dominguez Lacasa et al. (2019) propose a statistical framework, which dis
tinguishes between three interrelated dimensions: intensity of technology upgrading,
structural change, and global interaction. The authors find multiple unique paths of tech
nology upgrading in the BRICS, with differences across the three dimensions. Generally,
the BRICS have increased their innovation capabilities, while reducing dependence on
foreign actors and external knowledge transfer in order to catch up. They have progres
sively developed the ability to carry out technological frontier-pushing activities—with the
sole exception of Russia. As a result, the structure of technological knowledge—measured
through the share of high-tech and knowledge-intensive patent applications—has in
creased across countries (Dominguez Lacasa et al., 2019). China and Russia are the only
ones to have augmented, or at least kept constant, the intensity of behind-the-frontier
technological activities. China is unique in this regard, as she has rapidly enhanced inno
vation capabilities, diversified technology knowledge bases and entered into dynamic
frontier-pushing areas. (p. 762) The country is fostering structural change and global in
teractions towards technological frontier-expanding activities—a process that, according
to Dominguez Lacasa et al. (2019), emulates Korea or Taiwan during earlier periods of
technological upgrading and catching up (see Rasiah, Chapter 24 in this volume). In con
trast, foreign knowledge still tends to crowd out domestic behind-the-frontier technologi
cal efforts in Brazil, India, and South Africa.
Massive R&D investments evidence China’s stronger commitment to the building of pro
ductive and particularly technological capabilities relative to other BRICS. In 2011, China
became the world’s second-largest investor in R&D in volume terms, second only to the
United States (UIS, 2019). While in GDP terms China still ranks behind most developed
countries, it tends to outperform the other BRICS. The R&D funding and execution struc
ture shows that business enterprises account for about three-quarters of the total, a
Page 13 of 34
structure similar to advanced countries and in stark contrast with the general situation of
government-driven R&D in most other BRICS and other developing countries.
26.4.3 Integration through Trade and Investment
Integration through value addition incorporated in manufacturing exports within BRICS

has grown steadily over the past decade, but bilateral relationships are rather mixed (Ta
ble 26.1). With the exception of China, and to a lesser extent Russia, the BRICS have en
hanced their backward integration with others in the group. For instance, the share of
value addition of Brazilian manufacturing exports coming from the BRICS grew from 1.2
per cent in 2005 to 2.7 per cent in 2015. The figures for India and South Africa are more
significant, from 3.2 per cent to 5.2 per cent and from 2.1 per cent to 6.9 per cent, re
spectively. These two countries show the strongest level of backward linkage with other
BRICS. At less than 1.5 per cent, Chinese exports maintain the lowest level of backward
integration.
Page 14 of 34
Table 26.1 Composition of gross manufacturing exports by country of origin of value added, 2005–15
20 20 20 20 20 20 20 20 20 20 20 20 20 20 20
05 10 15 05 10 15 05 10 15 05 10 15 05 10 15
Brazil China India Russia South Africa
Bra — — — 0.3 0.6 0.4 0.2 0.4 0.4 0.2 0.2 0.1 0.3 0.4 0.7
zil
Chi 0.6 1.1 2.0 — — — 1.5 2.7 3.7 0.7 1.8 2.0 1.2 2.2 4.7
na
In 0.2 0.3 0.3 0.4 0.4 0.3 — — — 0.1 0.2 0.2 0.4 0.7 1.1
dia
Rus 0.3 0.3 0.3 0.8 0.6 0.5 0.8 0.9 0.6 — — — 0.2 0.3 0.4
sia
Sou 0.1 0.1 0.1 0.2 0.3 0.2 0.7 0.8 0.5 0.1 0.2 0.1 — — —
th
Afri
ca
Page 15 of 34
Do 86. 86. 83. 71. 77. 81. 74. 66. 72. 85. 86. 85. 77. 76. 70.
me 3 9 8 6 4 3 8 5 7 9 3 9 8 1 1
stic
val
ue
add
ed
BRI 1.2 1.8 2.7 1.7 1.9 1.4 3.2 4.8 5.2 1.1 2.4 2.4 2.1 3.6 6.9
CS
Res 12. 11. 13. 26. 20. 17. 21. 28. 22. 13. 11. 11. 20. 20. 23.
t of 5 4 4 7 9 4 9 7 0 1 4 7 0 3 0
the
wor
ld
Notes: BRICS minus the specific member country reporting exports.
Source: Author’s based on OECD’s Trade in Value Added (TiVA) database. http://oe.cd/tiva.
Page 16 of 34
The situation is less straightforward if one considers the industry of origin of value added
in each of the BRICS’ manufacturing exports. Generally, Chinese manufacturing value
added contributes 50 per cent or more of the manufacturing exports of other BRICS. In
contrast, with the exception of India, which contributes about 39.5 per cent of Chinese
manufacturing exports, the rest of the BRICS have seen declining shares of local manu
facturing value addition incorporated into Chinese manufacturing exports. In 2015, the
figures were 19.9 per cent for Brazil, 22.5 per cent for Russia and 15.3 per cent for South
Africa. Long-term trends show a decline in the intra-BRICS value-added content of Chi
nese manufacturing exports, while supplies of mining and quarrying products are the
main input to Chinese manufacturing exports. While South African exports tend to incor
porate a more significant share of manufacturing supplies from other BRICS, mining and
quarrying products are the main source of South African value added incorporated into
other BRICS manufacturing exports. This intra-BRICS (p. 763) (p. 764) pattern of trade
should continue to influence the bargaining power of individual group members, particu
larly vis-à-vis China.
The BRICS have emerged as major destinations and sources of FDI. In 2018 they repre
sented 20 per cent of the FDI inflows and around 10 per cent of the inward FDI stock in
the world (UNCTAD, 2019). Despite this growing importance at the global level, and
notwithstanding positive growth trends in intra-BRICS FDI in the early 2000s (UNCTAD,
2013), the level of integration within the group remains low (Table 26.2). The strongest
linkages appear in inward FDI stock from China in South Africa (5.0 per cent), followed
by Russia (1.6 per cent), while the share of South Africa and Russia in Chinese inward
FDI stock is considerably lower. South African investments in India (1.2 per cent) stand
out among the rest of the BRICS, which generally show very limited integration; individ
ual shares hardly exceed 0.3 per cent of total inward FDI stocks in each individual BRICS.
Overall, South Africa is the country with the largest share of inward FDI stocks from
BRICS (5.7 per cent), followed by India (2.3 per cent) and Russia (2.0 per cent). The data
also corroborate the greatest importance of circular FDI flows (round-tripping) for China
and Russia. Andreff (2015) attributes this to the tradition of BRICS multinationals target
ing tax havens or, in the case of China, benefiting from the favourable tax conditions of
fered to foreign investors in mainland China.
Page 17 of 34
Table 26.2 UNCTAD FDI estimates by ultimate investor, share in inward FDI stock, 2017
Recipient
Investor Brazil China India Russia South Africa
Brazil 0.0 0.1 0.0 0.0 0.0
China 0.7 7.8 0.8 1.6 5.0
India 0.1 0.2 1.6 0.3 0.4
Russia 0.2 0.3 0.3 6.5 0.2
South Africa 0.1 0.5 1.2 0.1 0.1
BRICS 1.1 1.1 2.3 2.0 5.7
Notes: UNCTAD estimates for 108 recipient countries, corresponding to 93 per cent of the value of global inward
FDI stock. Details in Casella (2019).
Source: UNCTAD (2019).
Page 18 of 34
The low level of integration through FDI can largely be explained by BRICS multination
als’ predominantly market-seeking strategy, which shows a preference for investing with
in regional value chains—including through trans-border mergers and acquisitions—and
in tax havens or markets in (tax-friendly) developed countries (UNCTAD, 2013; Andreff,
2015). China, and to a lesser extent Russia, show a stronger inclination to diversify geo
graphically, including to developing regions—notably Africa and in the case of China, also
Latin America (Andreff, 2015).
According to Andreff (2015), BRICS investment in developed countries mainly

(p. 765)
takes the form of mergers and acquisitions rather than greenfield investments. Regarding
sectoral composition, the evidence suggests a resource-seeking behaviour with relevant
shares in primary-sector activities, followed by services in overall outward FDI.
26.5 BRICS and the 4IR

An analysis of patent and trade data conducted by (UNIDO, 2020), and involving the ad
vanced digital technologies usually associated with the 4IR—industrial Internet of things,
big-data analytics, advanced robotics, artificial intelligence, cloud computing, and addi
tive manufacturing—characterized the global landscape of production and the use of
those new technologies. About 167 economies were divided into four distinct groups ac
cording to their level of engagement with the new technologies.
Looking at the BRICS, only China made it to the frontrunners’ group, which includes the
top ten countries—mostly developed—in terms of innovation in advanced digital produc
tion technologies. These economies account for more than 90 per cent of all patent appli
cations and about 70 per cent of the exports of goods associated with these technologies
(UNIDO, 2020). The other four BRICS appear in the follower group, economies that ac
tively engage with the new technologies, producing and selling in international markets,
but to a much lesser extent than the frontrunners. Some differences persist within this
group. While Brazil, Russia, and India appear as producers and exporters of 4IR technolo
gies, South Africa remains mainly a user—importer—of them (UNIDO, 2020).2
Lee et al. (2019) question how much the 4IR represents a window of opportunity for up
grading or whether, on the contrary, this is a process likely to reinforce the risk of becom
ing stuck in middle-income traps. Similar questions influence strategic thinking and poli
cymaking within BRICS, and efforts are already ongoing towards a BRICS joint 4IR devel
opment agenda. The emerging evidence suggests that each country is positioning itself
differently depending on individual industrial development trajectories and individual ap
proaches to industrial policy.
Generally, the BRICS are among the few developing countries where a dedicated strategy,
or at the very least, some explicit policy efforts, are geared to addressing the 4IR (UNI
DO, 2020; Santiago, 2018). BRICS aspire to foster innovation-driven economies, away
from commodities and traditional industrial products and moving increasingly into higher
value-added sectors. China, in particular, is steadily moving towards industrial (p. 766) de
Page 19 of 34
velopment strategies that seek to capitalize on her increasing ability to reproduce and
produce new technologies; the search for value addition and enhanced technological con
tent is superseding traditional cost advantage strategies (Li, 2017).
Similar to the case of advanced countries, the BRICS national strategies around the 4IR
are a blend of policy realms and approaches, including industrial policies (China and
South Africa) or STI plans (Brazil and South Africa), digitalization strategies (Russia) or
national digital agendas (China), or standalone documents using such terms as ‘Industry
4.0’ and ‘advanced manufacturing’, or referring to specific technologies (UNIDO, 2020).
South Africa exemplifies the latter approach, as it has recently introduced a strategy
around additive manufacturing (CSIR and DST, 2016). Alternatively, India is leveraging on
existing industrial development plans, such as Make in India, strengthening linkages to
the digitalization of industry.
An interesting feature of 4IR is its ability to underscore the need for novel approaches to
strategy setting. In his study of industrial policy in the BRICS countries, Di Maio (2015)
asserts that a recent trend in industrial policy design is the adoption of principles of the
so-called new industrial policy, stressing multi-stakeholder participatory processes and
the contribution of public‒private dialogues to policy design and implementation. UNIDO
(2020) supports this finding, as several BRICS have followed participatory processes in
the preparation of national 4IR strategies. Country responses are often built around a
‘multiple helix’ involving government, academia, private entities, and civil society; the
challenging nature of these consultative approaches cannot be underestimated. Future
research on the BRICS should consider how to organize, govern, and sustain such partici
patory processes and, more importantly, how BRICS are moving forward in the implemen
tation of those strategies at the individual and, as we discuss in section 26.5.1, the collec
tive level.
26.5.1 Collective Responses
At the end of their most recent meeting, the BRICS industry ministers acknowledged that
tightening collaboration in industrial development matters should assist them in address
ing what they termed the ‘Digital Industrial Revolution’ (BRICS Industry Ministers, 2018).
They committed to boosting industrialization through new high-tech-driven innovative
policy and regulatory frameworks (BRICS, 2017b) and the adoption of joint initiatives
around the 4IR (Business Report, 2018). While at this early stage it is difficult to docu
ment BRICS initiatives in the field, few proposals have been tabled. Based on a somewhat
common approach to industrial policy design ideas, such proposals can be characterized
along three core areas: building basic framework conditions, fostering the demand for
and adoption of 4IR technologies, and strengthening skills and research capabilities at
different levels (López-Gómez et al., 2017; UNIDO, 2020).
Page 20 of 34
(p. 767) 26.5.1.1 Building Basic Framework Conditions

Framework conditions refers to regulations and digital infrastructure. They also include
the institutional framework for strategy setting, policy formulation, and governance,
which is increasingly perceived as multi-stakeholder and involving close private‒public
collaboration. BRICS have started the dialogue. At the 2018 summit in Johannesburg, the
BRICS leaders endorsed the creation of a BRICS Partnership on New Industrial Revolu
tion (PartNIR) (BRICS, 2018), as previously agreed by the respective ministers of industry
at their third ministerial meeting (BRICS Industry Ministers, 2018). An Advisory Group to
include representatives from BRICS ministries of industry and experts from all BRICS
members will be responsible for operationalizing the PartNIR. The terms of reference and
work plan will focus on digitalization, industrialization, innovation, inclusiveness, and in
vestment (BRICS Industry Ministers, 2018). Close collaboration with the private sector is
prescribed through working groups on a wide variety of industrial and manufacturing
sectors.
The ultimate intended outcomes of the PartNIR include boosting economic growth,
strengthening sustainable industrial production capacity, and creating networks of sci
ence parks and technology business incubators; of particular importance is the inclusion
of SMEs in technology-intensive areas (BRICS, 2018: item 56). Concrete initiatives in
clude the establishment of the BRICS Networks of Science Parks, Technology Business In
cubators and Small and Medium-sized Enterprises (BRICS, 2018).
Regarding infrastructure, UNIDO (2020) documents that digitalization remains a binding

constraint for the development of advanced manufacturing. South Africa, a laggard
among BRICS, needs to address high-cost and limited bandwidth, skill shortages in ICT
and data analysis, not to mention innovation system weaknesses such as poor interaction
between firms, universities, and research centres (UNIDO, 2020). More generally, up
grading of ICT infrastructure and connectivity are among BRICS priorities; they are com
mitted to promoting internationally accepted and applicable rules for security of ICT in
frastructure, data protection, and the Internet (BRICS, 2017b). All these initiatives are
consequential on the BRICS’ intention to foster the digital economy through the BRICS
ICT Development Agenda and Action Plan (BRICS, 2017b).
Global debates around the developmental implications of big data cannot be overlooked
as capabilities to generate, process, and extract value from rapidly expanding pools of
datasets are prone to high concentration both geographically and in terms of monopolis
tic power (Cable, 2018; Marr, 2018). The BRICS can potentially play major roles in ad
vancing a global agenda around the development of big data and its associated implica
tions for security, privacy, and the capabilities required for handling growing and increas
ingly complex volumes of varied forms of data (Mahrenbach, Mayer, and Pfeffer, 2018).
China is among the most advanced developing countries in the use of data as an econom
ic driver; development strategies prescribe data analytics and big-data platforms to sup
port new industries and foster upgrading of traditional ones (Mahrenbach, Mayer, and Pf
effer, 2018). The Chinese government encourages innovation and funds big-data re
Page 21 of 34
search, while scientists are compelled to link and share research findings with industry,
and companies are expected to increase productivity and growth via incorporation of big
data and associated technologies in high-tech manufacturing and in new business models
(Mahrenbach, Mayer, and Pfeffer, 2018). All private-sector activities have to remain with
in specified regulatory and incentive frameworks.
While this performance potentially locates China among the forerunners in the develop
ment of big data at the global level, any counterbalancing or supplementary role for the
other BRICS in the consolidation of generally agreed guidelines for governing develop
ment of markets for big data is yet to emerge. Sing (2017) proposes three possible emerg
ing models. At the one end is a model led mostly by the United States, centred on a global
laissez-faire approach and dominated largely by domestic digital firms and free and un
regulated data flows. At the opposite end is the Chinese approach to state-led capitalism,
where the government enjoys strong surveillance rights and control of digital transac
tions. In between these two, Sing (2017) places mixed-economy approaches to digitaliza
tion, as emerging in India or the European Union, where governments play major roles in
fostering infrastructure and regulation to support competitive and efficient open data
market operations, without renouncing control of monopolies and concentration, or the
oversight of areas of social and economic importance.
Similarly, as part of the BRICS Action Agenda on Economic and Trade Cooperation, the
group adopted an Intellectual Property Rights Cooperation Mechanism (BRICS IPRCM)
which should enhance cooperation and coordination in this area, particularly through the
BRICS IPR Cooperation Guidelines and an Action Plan on BRICS IPR Cooperation (BRICS,
2017a). Concrete activities include fostering information exchange among the IPR offices
of BRICS countries and capacity building on IPR issues.
26.5.1.2 Fostering Demand and Adoption of 4IR Technologies

UNIDO (2020) argues that even if suitable framework conditions are put in place, coun
tries need to encourage adoption and adaptation of advanced digital production technolo
gies. Given the novelty of many of the products and services that can be derived from the
4IR technologies, economic agents find it difficult to appraise their value propositions rel
ative to competing existing technologies, while challenges to assessing the expected re
turn on investment contrast with perceived high upfront investment requirements (Stein
mueller, 2001). Addressing this situation requires concentrated policy efforts to raise
awareness of firms about the potential benefits of the new technologies. It also requires
funding of their adoption to be facilitated. Countries may adopt and jointly implement
dedicated programmes and incentives to raise awareness and rally the interest of domes
tic agents, or to foster emerging behaviours and initiatives related to the 4IR.
The BRICS expect to enhance collaboration on STI through joint proposals to mobilize
their accumulated technological and R&D capabilities (BRICS, 2018). (p. 769) BRICS ex
ploration of collaboration is already motivating joint research activities and innovation
agendas on big data, ICT, and other advanced manufacturing technologies and their ap
plications, and on ICT infrastructure and connectivity (BRICS, 2017b). Concrete propos
Page 22 of 34
als include establishing the BRICS Institute of Future Networks, which will support joint
BRICS research and innovation in ICT—including Internet of things, cloud computing, big
data, data analytics, nanotechnology, artificial intelligence, and 5G—and their applica
tions (BRICS, 2017b). The suggested approach is to encourage partnerships between mul
tiple stakeholders to devise and implement proofs-of-concept and pilot projects in these
areas.
26.5.1.3 Strengthening Skills and Research Capabilities

UNIDO (2020) documents that governments can play a major role in the creation of the
capabilities required to adopt 4IR technologies, for example, through dedicated learning
centres and new approaches to technical and vocational education and training. Expand
ing the scope and number of research institutions specifically to deal with these technolo
gies is key for their absorption and adaptation to local circumstances.
López-Gómez et al. (2017) advocate the building of human capabilities, particularly to en
able adjustments in labour markets and systems of education, training, and retraining.
Concrete activities along these lines are yet to emerge but, as agreed in the most recent
ministers of industry meeting, the BRICS intend to enhance policy coordination around
advanced technical skills and training, exchange of information, and best practices with
respect to digitization. Securing inclusive and equitable growth, while developing syner
gies in the use of financial and human resources, is likewise important (Business Report,
2018).
Interesting initiatives implemented recently at the individual country level may serve as
examples for larger, BRICS-level interventions. For instance, in Russia, the Agency for
Strategic Initiatives (ASI) and Skolkovo (Moscow School of Management) have developed
the Atlas of New Professions, an online platform intended to help understand what the
twenty-five most promising sectors will be for the next fifteen to twenty years, together
with relevant occupations and their respective skill requirements and associated organi
zational practices; likewise important is the stocktaking of more than fifty occupations
that are bound to become outdated—including both routine and high-skilled ones (ASI
and Skolkovo, 2019). In parallel, ASI initiatives such as Skill Development for Industrial
Growth are working to improve matching between industry’s skill demands and the sup
ply from educational establishments (Roland Berger GMBH, 2016).
According to Tourk and Marsh (2016), China is fostering programmes to attract foreign
talent to supplement its limited domestic innovation capabilities. For example, in 2013,
the Fozhou Bureau of Foreign Experts Affairs was established in Fozhou, capital city of
Fujian province, providing free work spaces and funding for start-up companies and
arranges matching them with potential investors (Tourk and Marsh, 2016). By the end of
2014, ten start-ups had settled in the city, and it recorded a (p. 770) significant flow of re
turning students; moreover, the model had attracted the attention of other provincial gov
ernments in China (Tourk and Marsh, 2016).
Page 23 of 34
26.6 The International Dimension

The BRICS cooperation agenda is broad, with an ever-growing number of items. From an
initial focus on global governance and diplomacy—including UN reform and enlargement
of the UN Security Council to include some BRICS as permanent members—or the reform
of current international financial architecture, the agenda has evolved to include STI top
ics and economic development. BRICS expect to influence global peace and security, en
ergy and climate change, and social and economic issues in a non-confrontational spirit
together with other global governance arrangements (BRICS, 2019).
The creation of a New Development Bank (NDB) is a major initiative articulating the
BRICS’ collective presence and global influence. The NDB mobilizes resources for infra
structure and sustainable development projects in BRICS and other emerging and devel
oping economies (New Development Bank, 2019a).3 In addition to strengthening intra-
BRICS cooperation, the NDB expects to supplement the funding of multilateral and re
gional financial institutions (BRICS, 2014). The bank became operational with the signing
of the Headquarters Agreement with the government of China and the Memorandum of
Understanding with the Shanghai Municipal People’s Government on 27 February 2016
(BRICS, 2019). The first loans, approved in 2016‒17, support projects in green and re
newable energy, transportation, water sanitation, and irrigation (BRICS, 2019). By the
end of 2019, the total loan portfolio should amount to US$16 billion (New Development
Bank, 2019c), double the US$8.1 billion at the end of 2018 (New Development Bank,
2019b).
Creation of the NDB has not been without tensions. Its operational structure prescribes
an initial authorized capital of US$100 billion, with a first subscription of US$50 billion
equally shared among founding members. Governance and leadership are split among the
BRICS, with headquarters in Shanghai and an Africa Regional Centre in South Africa
(BRICS, 2014). Kejin (2014) and Degaut and Meachan (2015) assert that although the
BRICS contribute equally to the first tranche of capitalization of the NDB, China seems
the only member standing ready and able to expand and fulfil future commitments. China
could take the lion’s share of the decision-making power, particularly as the conditions for
non-BRICS membership of the NDB remain uncertain (Kejin, 2014). (p. 771)
26.6.1 Africa in Focus
It is no surprise that Africa occupies a prominent place in the BRICS agenda. Even before
their integration, each had established close social, political, and economic ties with the
continent. The steady increase in BRIC FDI in Africa suggests attempts at breaking cus
tomary regional specialization regarding destination countries. At the same time, the
growing presence of BRIC among Africa’s major trade and investment partners is per
ceived as a counterbalance to traditional partners, the United States, Europe, or Japan
(Kimenyi and Lewis, 2011). Similarly, the BRIC were already important donors in the re
Page 24 of 34
gion through concessionary and soft loans, credit lines and grants; China in particular ac
tively supports infrastructure projects (Kimenyi and Lewis, 2011).
Coincident with South Africa’s accession to BRICS, the Sanya Declaration made explicit
the BRICS’ intention to collaborate with Africa, with the New Partnership for Africa’s De
velopment (NEPAD) identified as the relevant interlocutor (BRICS, 2011). Engagement
with South African Union countries followed at the fifth BRICS leaders’ summit in Dur
ban, under the overarching theme of ‘BRICS and Africa: Partnership for Development, In
tegration and Industrialization’ (BRICS, 2013). BRICS support to Africa’s industrialization
seeks to potentiate the continent’s development efforts through FDI, knowledge ex
changes, capacity building, and trade diversification. A priority is to assist the continent
in redressing its infrastructure deficits, required to boost regional integration and indus
trialization (BRICS, 2013). This is consistent with the long-standing BRIC presence in the
African infrastructure market (Kimenyi and Lewis, 2011) and the potential gains offered
by a continent where investments in infrastructure to provide basic services to the popu
lation compete heavily against resources to improve or expand on transport and other
economic infrastructure (Oxford Economics and Global Infrastructure Hub, 2017). The es
tablishment of a Multilateral Agreement on Infrastructure Co-Financing for Africa with
support from the BRICS’ export‒import and development banks is proposed to move col
laboration forward (BRICS, 2013).
The BRICS presence in Africa provokes some criticisms. Kimenyi and Lewis (2011) decry
the perceived appetite for Africa’s natural resources, notably oil—with principally China’s
and Brazil’s expanding presence in Angola, Nigeria, and Sudan—mining in Liberia and
Mozambique, and gas in Nigeria. Similarly, India’s presence has expanded in textiles, so
cial services—health and education—ICT, and automobiles. Mbele (2018) stresses the
asymmetrical patterns of trade and investment between the BRICS and Africa. South
Africa’s accession to BRIC provoked strong criticism; in addition to the huge economic
asymmetries, South Africa is perceived as supporting the BRIC and particularly Chinese
interest in the continent. Kimenyi and Lewis (2011) note ties between Chinese invest
ments and Chinese-sponsored infrastructure development projects over the last two
decades, while South African firms provide the platform for a cooperation model based on
African exports of raw materials and manufacturing imports (Carmody, 2017).
Africa is likely to continue to play a major role in the context of China’s Belt and
(p. 772)
Road Initiative (BRI), launched in 2013, which endeavours to transform six trade and in
vestment corridors across Asia, Europe, and Africa, with an emphasis on infrastructure
and connectivity. Underscoring the benefits of BRICS’ closer collaboration will be the pos
sibilities of channelling funding through the NDB and other funding mechanisms—for ex
ample the Asian Infrastructure Investment Bank—in which BRICS participate. However,
geopolitical tensions associated with BRI should not be ignored, particularly as India and
Russia maintain close ties and interests in regions directly located along several BRI main
economic corridors.
Page 25 of 34
26.7 Conclusions
The conformation of the BRICS block is a work in progress, at very early stages of devel
opment. The process faces significant constraints rooted in the history, structure, and re
cent economic and industrial performance of the countries involved. Several significant
shortcomings could deepen in the near future. In particular, this chapter has corroborat
ed the widening gap in performance between China and the other BRICS. This is poten
tially a major hindrance, as it exacerbates differences in interests and capabilities within
the group (Degaut and Meachan, 2015). While BRICS integration would benefit from clos
ing productive and technological capability gaps, recognition of this challenge remains
largely absent from policy documents governing the integration process.
BRICS seem to understand the value of conforming a strong block. Those who dismiss the
possibility of their achieving closer collaboration should remember that BRICS is mostly
an informal collaboration mechanism; each participant plays according to its own rules
and interests (Kejin, 2014). BRICS integration is and will continue to be gradual but sys
tematic, through dedicated strategies or action plans to govern collaboration in several
areas. Other, more mature international governance mechanisms—the G7 or even the
G20—have been built progressively; there is no reason to expect the BRICS to move any
faster or any deeper.
As areas for collaboration expand at every leaders’ or ministerial or expert meeting, sys
tematic and empirically driven evaluation of commitments and their tangible outcomes is
necessary to draw conclusions. This chapter proposes that industrial development offers
great potential for further follow-up in the future. Despite its novelty within the BRICS
collaboration agenda, industrial development is an area that could integrate and expand
individual BRICS approaches to STI, IPRs, e-commerce, infrastructure, and other related
areas.
BRICS, and particularly China, attract significant attention regarding strategic policy re
sponses to the 4IR. Unlike many other developing countries, they have generally under
gone significant structural change with strong reliance on manufacturing. They are, and
will remain, role models for other developing countries where (p. 773) industrial develop
ment is yet to take root. As the ongoing technological and productive transformations
continue to unfold in uncertain directions, BRICS are continuing to foster structural
change and the development of innovation-driven economies, moving away from com
modities and traditional industrial products and increasingly into higher-value-added sec
tors and products. At the same time, BRICS’ strides up the ranks of global industrial lead
ership increase the risks of widening gaps relative to least-developed economies (Liu et
al., 2017; Mayer, 2018).
Among the BRICS, China seems ready to tap into the current juncture to explore new de
velopment paths, steadily shifting from a catching-up mindset centred on labour-cost ad
vantages to one of capitalizing on its growing role as a leading global technology and
Page 26 of 34
manufacturing centre (Li, 2017; The Economist, 2015). Implications for the future of in
dustrial dynamics in both developed and developing countries may be significant.
Finally, yet importantly, from a historical perspective the experience of BRICS reminds us
that catching up entails active industrial policies to underpin the creation of absorptive
technological and productive capabilities. This characteristic should continue to assist the
BRICS in their individual and collective development efforts.
Acknowledgements
Earlier versions of this manuscript benefited from comments and suggestions from
Nobuya Haraguchi, Professor Fantu Cheru, Professor Li Chen, the editors of this volume,
and participants at two review workshops held in Addis Ababa in April and September
2019.
References
Abramovitz, Moses (1986) ‘Catching up, Forging ahead, and Falling behind’, The Journal
of Economic History 46(2): 385–406.
Amirapu, Amrit and Arvind Subramanian (2015) ‘Manufacturing or Services? An Indian Il
lustration of a Development Dilemma’, SSRN Scholarly Paper ID 2623158. Rochester, NY:
Social Science Research Network. Available at https://papers.ssrn.com/ab
stract=2623158.
Andreff, Wladimir (2015) ‘Outward Foreign Direct Investment from BRIC Countries: Com
paring Strategies of Brazilian, Russian, Indian and Chinese Multinational Companies’,
The European Journal of Comparative Economics 12(2): 79–131.
ASI and Skolkovo (2019) ‘Atlas of Emerging Jobs’. Available at http://atlas100.ru/en/.
BRIC (2009) ‘Joint Statement of the BRIC Countries’ Leaders’. Yekaterinburg, Russia, 16
June. BRICS Information Centre—University of Toronto. Available at http://
www.brics.utoronto.ca/docs/090616-leaders.html.
BRIC (2010) ‘2nd BRIC Summit of Heads of State and Government: Joint Statement’.
Brasília, 15 April. BRICS Information Centre—University of Toronto. Available at http://
BRICS (2011) ‘Sanya Declaration’. Sanya, Hainan, China, 14 April. BRICS Infor
(p. 774)
mation Centre—University of Toronto. Available at http://www.brics.utoronto.ca/docs/

110414-leaders.html.
BRICS (2013) ‘BRICS and Africa: Partnership for Development, Integration and Industri
alisation. EThekwini Declaration’. Durban, South Africa, 27 March. BRICS Information
Page 27 of 34
Centre—University of Toronto. Available at http://www.brics.utoronto.ca/docs/130327-

statement.html.
BRICS (2014) ‘Fortaleza Declaration’. 6th BRICS Summit, 15 July, Fortaleza, Brazil.
BRICS Information Centre—University of Toronto. Available at http://
BRICS (2015a) ‘2015 Ufa Declaration’. 7th BRICS Summit, Ufa, Russia, 9 July. BRICS In
formation Centre—University of Toronto. Available at http://www.brics.utoronto.ca/
docs/150709-ufa-declaration_en.html.
BRICS (2015b) ‘Strategy for the BRICS Economic Partnership’. BRICS. Available at file:///
C:/Users/santiagf/Downloads/partnershipstrategy_eng.pdf.
BRICS (2017a) ‘BRICS Action Agenda on Economic and Trade Cooperation’. BRICS. Avail
able at http://economy.gov.ru/wps/wcm/connect/88d87c66-ec8e-4398-ae01-
c1dfad7423a8/2017040910.pdf?MOD=AJPERES&CACHEID=88d87c66-
ec8e-4398-ae01-c1dfad7423a8.
BRICS (2017b) ‘BRICS Leaders Xiamen Declaration’. BRICS. Available at http://

www.brics.utoronto.ca/docs/170904-xiamen.html.
BRICS (2017c) ‘Progress Report on the Implementation of the Strategy for BRICS Eco
nomic Partnership’. BRICS. Available at https://www.ranepa.ru/images/media/brics/
2017/BRICS%20Strategy%20Progress%20Report_25.08.2017.pdf.
BRICS (2018) ‘BRICS in Africa: Collaboration for Inclusive Growth and Shared Prosperity
in the 4th Industrial Revolution. Johannesburg Declaration’. 10th BRICS Summit, Johan
nesburg, South Africa, 26 July. BRICS Information Centre—University of Toronto. Avail
able at http://www.brics.utoronto.ca/docs/180726-johannesburg.html.
BRICS (2019) ‘BRICS Information Portal’. Available at http://infobrics.org/.
BRICS Industry Ministers (2015) ‘Declaration of the BRICS Industry Ministers. Industrial
Cooperation of the BRICS Countries: New Opportunities for Growth’. Moscow, Russia, 20
October. BRICS. Available at http://www.brics.utoronto.ca/docs/151020-industry.pdf.
BRICS Industry Ministers (2017) ‘BRICS Heads Decide on a 7-Point Action Plan’. BRICS
Information Portal. Available at http://infobrics.org/post/25407.
BRICS Industry Ministers (2018) ‘Declaration of the Third BRICS Industry Ministers’
Meeting’. TRALAC. Available at https://www.tralac.org/news/article/13227-declara
tion-of-the-third-brics-industry-ministers-meeting.html.
Brigante Deorsola, Adriana, Marcia Cristiane Martins Ribeiro Leal, Milene Dantas Caval
cante, Ingrid Jensen Schmidt, and Edimilson Junqueira Braga (2017) ‘Intellectual Proper
Page 28 of 34
ty and Trademark Legal Framework in BRICS Countries: A Comparative Study’, World

Patent Information 49: 1–9.
Business Report (2018) ‘Declaration of the BRICS Industry Ministers’. IOL Business Re
port. Available at https://www.iol.co.za/business-report/economy/declaration-of-the-
brics-industry-ministers-15835648.
Cable, Vince (2018) ‘The Tech Titans Must Have Their Monopoly Broken—and This Is
How We Do It’, The Guardian, 20 April. (p. 775)
Carmody, Pádraig (2017) ‘The Geopolitics and Economics of BRICS’ Resource and Market
Access in Southern Africa: Aiding Development or Creating Dependency?’, Journal of
Southern African Studies 43(5): 863–77.
Cassella, Bruno (2019) ‘Looking through Conduit FDI in Search of Ultimate Investors—a
Probabilistic Approach’, Transnational Corporations: Investment and Development 26(1):
109–46. Available at https://unctad.org/en/PublicationsLibrary/
diaeia2019d1_en.pdf.
Cimoli, Mario, Giovanni Dosi, and Joseph E. Stiglitz (2009) The Political Economy of Capa
bilities Accumulation: The Past and Future of Policies for Industrial Development. Oxford:
CSIR and DST (2016) ‘A South African Additive Manufacturing Strategy’. Council for Sci
entific and Industrial Research and Department of Science and Technology. Available at
https://site.rapdasa.org/wp-content/uploads/2017/02/South-African-Additive-Man
ufacturing-Strategy.pdf.
Cui, Yu, Jie Jiao, and Hao Jiao (2016) ‘Technological Innovation in Brazil, Russia, India,
China, and South Africa (BRICS): An Organizational Ecology Perspective’, Technological
Forecasting and Social Change 107: 28–36.
Daudt, Gabriel and Luiz Daniel Willcox (2016) ‘Reflexões Críticas a Partir Das Experiên
cias Dos Estados Unidos e Da Alemanha Em Manufatura Avançada’, BNDES Setorial 44:
5–45.
Degaut, Marcos and Carl Meachan (2015) ‘Do the BRICS Still Matter?’. Center for Inter
national and Strategic Studies. Available at http://enotus.com/wp-content/uploads/
2015/07/Do-the-Brics-Still-Matter_.pdf.
Di Maio, Michele (2015) ‘Industrial Policy in the BRICS: Similarities, Differences, and Fu
ture Challenges’, in Wim Naudé and Nobuya Haraguchi (eds) Structural Change and In
dustrial Development in the BRICS. Oxford: Oxford University Press, pp. 429–54.
Dominguez Lacasa, Iciar, Björn Jindra, Slavo Radosevic, and Mahmood Shubbak (2019)
‘Paths of Technology Upgrading in the BRICS Economies’, Research Policy 48(1): 262–80.
Page 29 of 34
Franco, Camila and Renan Henrique de Oliveira (2017) ‘Inputs and Outputs of Innova
tion: Analysis of the BRICS: Theme 6—Innovation Technology and Competitiveness’, RAI
Revista de Administração e Inovação 14(1): 79–89.
Government of the Russian Federation (2017) ‘Digital Economy Program of the Russian
Federation’. Moscow: Government of the Russian Federation.
Gray, Kevin and Barry K. Gills (2017) Rising Powers and South–South Cooperation. Lon
don: Routledge.
Gray, Kevin and Craig N. Murphy (2013) ‘Introduction: Rising Powers and the Future of
Global Governance’, Third World Quarterly 34(2): 183–93.
Hallward-Driemeier, Mary and Gaurav Nayyar (2017) Trouble in the Making? The Future
of Manufacturing-led Development. Conference edition. Washington, DC: World Bank.
Haraguchi, Nobuya, Charles Fang Chin Cheng, and Eveline Smeets (2017) ‘The Impor
tance of Manufacturing in Economic Development: Has This Changed?’, World Develop
ment 93: 293–315.
Hobday, Michael, Howard Rush, and John Bessant (2004) ‘Approaching the Innovation
Frontier in Korea: The Transition Phase to Leadership’, Research Policy 33(10): 1433–57.
Kahn, Michael, Luiz Martins de Melo, and Marcelo Pessoa de Matos (2014) Financing In
novation: BRICS National Systems of Innovation. New Delhi: Routledge and IDRC.
Kejin, Zhao (2014) ‘The Limits of Cooperation among BRICS Countries’. Carnegie-Ts
inghua Center (blog). Available at https://carnegietsinghua.org/2014/12/01/limits-of-
cooperation-among-brics-countries-pub-57341. (p. 776)
Kimenyi, Mwangi and Zenia Lewis (2011) ‘The BRICs and the New Scramble for Africa’,
in Foresight Africa: The Continent’s Greatest Challenges and Opportunities for 2011.
Washington, DC: Brookings Institution Press, pp. 19–21. Available at https://
www.brookings.edu/multi-chapter-report/foresight-africa-the-continents-great
est-challenges-and-opportunities-for-2011/.
Lee, Keun (2019a) ‘Economics of Technological Leapfrogging’. Background paper for the
UNIDO Industrial Development Report 2019. Inclusive and Sustainable Development
Working Paper Series. Vienna: United Nations Industrial Development Organization.
Lee, Keun (2019b) The Art of Economic Catch-up: Barriers, Detours and Leapfrogging in
Lee, Keun and Franco Malerba (2017) ‘Theory and Empirical Evidence of Catch-up Cycles
and Changes in Industrial Leadership’, Research Policy 46(2): 337.
Lee, Keun, Chan-Yuan Wong, Patarapong Intarakumnerd, and Chaiyatorn Limapornvanich

(2019) ‘Is the Fourth Industrial Revolution a Window of Opportunity for Upgrading or Re
Page 30 of 34
inforcing the Middle-income Trap? Asian Model of Development in Southeast Asia’, Jour
nal of Economic Policy Reform. Available at https://doi.org/
10.1080/17487870.2019.1565411.
Li, Ling (2017) ‘China’s Manufacturing Locus in 2025: With a Comparison of “Made-in-
China 2025” and “Industry 4.0”’, Technological Forecasting and Social Change 135©: 66–
74.
Liu, Xielin, Sylvia Schwaag Serger, Ulrike Tagscherer, and Amber Y. Chang (2017) ‘Be
yond Catch-up—Can a New Innovation Policy Help China Overcome the Middle-income
Trap?’, Science and Public Policy 44(5): 656–69.
López-Gómez, Carlos, David Leal-Ayala, Michele Palladino, and Eoin O’Sullivan (2017)
‘Emerging Trends in Global Advanced Manufacturing: Challenges, Opportunities and Poli
cy Responses’. United Nations Industrial Development Organization and University of
Cambridge’s Institute for Manufacturing.
Mahrenbach, Laura, Katja Mayer, and Jürgen Pfeffer (2018) ‘Policy Visions of Big Data:
Views from the Global South’, Third World Quarterly 39(10): 1861–82.
Marr, Bernard (2018) ‘Here’s Why Data is Not the New Oil’, Forbes (blog). Available at
https://www.forbes.com/sites/bernardmarr/2018/03/05/heres-why-data-is-not-the-
new-oil/.
Mayer, Jörg (2018) ‘Digitalization and Industrialization: Friends or Foes?’. Research Pa
per No. 25. Geneva: UNCTAD. Available at https://unctad.org/en/pages/
PublicationWebflyer.aspx?publicationid=2263.
Mbele, Sizwe (2018) ‘Opportunity Africa: Will the BRICS Countries Manage Mutual Coop
eration?’. BRICS. Available at http://infobrics.org.
Mittelman, James H. (2016) ‘Repositioning in Global Governance: Horizontal and Vertical

Shifts amid Pliable Neoliberalism’, Third World Quarterly 37(4): 665–81.
Nelson, Richard R. (1994) ‘The Co-Evolution of Technology, Industrial Structure, and Sup
porting Institutions’, Industrial and Corporate Change 3(1): 47–63.
New Development Bank (2019a) ‘History’. New Development Bank (blog). Available at
https://www.ndb.int/about-us/essence/history/.
New Development Bank (2019b) ‘NDB Board of Governors and Board of Directors Meet
ings Held in Cape Town, South Africa’. New Development Bank (blog). Available at
https://www.ndb.int/press_release/ndb-board-governors-board-directors-meet
ings-held-cape-town-south-africa/.
New Development Bank (2019c) ‘New Development Bank (NDB) Outlines Ambitious Plans
to Boost Loans, Increase Impact of Investment’. New Development Bank (blog). Available
Page 31 of 34
at https://www.ndb.int/press_release/new-development-bank-ndb-outlines-
(p. 777)
ambitious-plans-boost-loans-increase-impact-investment/.
OECD (2017) ‘China and the Next Production Revolution’, in The Next Production Revolu
tion: Implications for Governments and Business. Paris: Organisation for Economic Coop
eration and Development, pp. 397–438.
O’Neill, Jim (2001) ‘Building Better Global Economic BRICs’. Global Economics Paper,
Goldman Sachs.
O’Neill, Jim (2018) ‘Jim O’Neill: Revisiting the Next BRIC Economies’. Available at
https://www.fnlondon.com/articles/jim-oneill-revisiting-the-next-bric-
economies-20180420.
O’Neill, Jim, Dominic Wilson, and Roopa Purushothaman (2005) ‘How Solid are the
BRICs?’. Global Economics Paper, Goldman Sachs.
Oxford Economics and Global Infrastructure Hub (2017) ‘Global Infrastructure Outlook:
Infrastructure Investment Needs 50 Countries, 7 Sectors to 2040’. Oxford Economics.
Available at https://outlook.gihub.org/.
Rodriguez-Arango, Liliana, and Maria Alejandra Gonzalez-Perez (2016) ‘Giants from

Emerging Markets: The Internationalization of BRIC Multinationals’, in Rob Van Tulder,
Alain Verbeke, Jorge Carneiro, and Maria Alejandra Gonzalez-Perez (eds) The Challenge
of BRIC Multinationals. Progress in International Business Research 11. Bingley: Emerald
Group Publishing, pp. 195–226.
Roland Berger GMBH (2016) ‘Skill Development for Industry 4.0: White Paper Summary’.
BRICS Business Council, India Section. Available at http://www.ficci.in/spdocument/
20786/Whitepaper-brics-skill.pdf.
Santiago, Fernando (2015) ‘Innovation Policy and Industrial Policy at the Crossroads: A
Review of Recent Experiences in Advanced Developing Countries’. Inclusive and Sustain
able Development Working Paper No. 09/2015. Vienna: UNIDO, Research, Statistics and
Industrial Policy Branch.
Santiago, Fernando (2018) ‘You Say You Want A Revolution: Strategic Approaches to In
dustry 4.0 in Middle-income Countries’. Inclusive and Sustainable Development Working
Paper No. 19. Vienna: UNIDO, Research, Statistics and Industrial Policy Branch.
Scerri, Mario and Helena Maria Martins Lastres (eds) (2013) The Role of the State:
BRICS National Systems of Innovation. New Delhi and Abingdon: IDRC and Routledge.
Schwab, Klaus (2016) The Fourth Industrial Revolution. New York: Crown Business.
Simachev, Yuri, Mikhail Kuzyk, Boris Kuznetsov, and Evgeniy Pogrebnyak (2014) ‘Russia
on the Path towards a New Technology Industrial Policy: Exciting Prospects and Fatal
Traps’, Foresight-Russia 8(4): 6–23.
Page 32 of 34
Sing, Parminder Jeet (2017) ‘Digital Industrialisation in Developing Countries—A Review

of the Business and Policy Landscape’. Bangalore: IT for Change. Available at http://
itforchange.net/sites/default/files/1468/Digital-industrialisation-May-2018.pdf.
Steinmueller, Edward (2001) ‘ICTs and the Possibilities for Leapfrogging by Developing
Countries’, International Labour Review 140(2): 193–210.
Stiglitz, Joseph E. (2017) ‘Industrial Policy, Learning, and Development’, in John Page and
Finn Tarp (eds) The Practice of Industrial Policy: Government–Business Coordination in
Africa and East Asia. WIDER Studies in Development Economics. Oxford: Oxford Univer
The Economist (2015) ‘Made in China?’, The Economist, 12 May. Available at http://
stage.economist.com/news/leaders/21646204-asias-dominance-manufacturing-
will-endure-will-make-development-harder-others-made. (p. 778)
Tourk, Khairy and Peter Marsh (2016) ‘The New Industrial Revolution and Industrial Up
grading in China: Achievements and Challenges’, Economic and Political Studies 4(2):
187–209.
UIS (2019) ‘How Much does your Country Invest in R&D?’. Available at http://
www.uis.unesco.org/_LAYOUTS/UNESCO/research-and-development-spending/in
dex-en.html.
UNCTAD (2013) ‘Global Investment Trends Monitor’. Global Investment Trends Monitor
No. 12. Geneva: United Nations Conference on Trade and Development. Available at
https://unctad.org/en/PublicationsLibrary/webdiaeia2013d6_en.pdf.
UNCTAD (2019) ‘World Investment Report 2019’. Geneva: United Nations Conference on
Trade and Development. Available at https://unctad.org/en/pages/
PublicationWebflyer.aspx?publicationid=2460.
UNIDO (2002) ‘Industrial Development Report 2002/3: Competing through Innovation

and Learning’. Vienna: United Nations Industrial Development Organization.
UNIDO (2005) ‘Industrial Development Report 2005: Capability Building for Catching-up.
Historical, Empirical and Policy Dimensions’. Vienna: United Nations Industrial Develop
ment Organization.
UNIDO (2019) ‘Competitive Industrial Performance Report 2018’. Biennial CIP Report.
Vienna, Austria: United Nations Industrial Development Organization. Available at
https://www.unido.org/sites/default/files/files/2019–05/CIP.pdf.
UNIDO (2020) ‘Industrial Development Report 2020. Industrializing in the Digital Age’.
Vienna: United Nations Industrial Development Organization.
Vértesy, Dániel (2011) ‘Interrupted Innovation: Emerging Economies in the Structure of

the Global Aerospace Industry’, PhD thesis, Maastricht University.
Page 33 of 34
Wilson, Dominic and Roopa Purushothaman (2003) ‘Dreaming with BRICs: The Path to
2050’. Global Economics Paper, Goldman Sachs.
Notes:
(1) Emphasis in the original by the author.
(2) The classification further includes a group of latecomers, economies with marginal
patent or trade activity in this field, but that have already engaged with 4IR technologies.
Finally, a larger group of laggards includes economies with no or very low engagement
with these technologies (UNIDO, 2020).
(3) The BRICS have also enacted a Contingency Reserve Arrangement to support one an
other in situations of balance-of-payments instability (BRICS, 2014).
Fernando Santiago
Fernando Santiago is an industrial policy officer at the United Nations Industrial De
velopment Organization (UNIDO) where he specializes in the design, monitoring, and
evaluation of industrial policies. He is also a member of the scientific board of the
Latin American Network for the Study of Learning Systems, Innovation and Compe
tency Building (LALICS). He has published on topics related to innovation and indus
trial policy. He is an economist from the National Autonomous University of Mexico
and holds a master’s degree from SPRU-University of Sussex and a PhD in economics
and policy studies of technical change from UNU-MERIT/University of Maastricht.
Page 34 of 34
Successes and Failures of Industrial Policy in Transition Economies of Eu
rope and Asia
Successes and Failures of Industrial Policy in Transi

tion Economies of Europe and Asia
Vladimir Popov
This chapter discusses the achievements and failures of various models of industrial poli
cy in post-communist countries. If industrial policy is needed, how are industries that
have to be supported to be selected? What are the appropriate tools/instruments to sup
port particular industries? Eastern European countries in general did not have any explic
it industrial policy, either via tax concessions and/or subsidies, or via under/overpricing
the exchange rate. Many countries of the former Soviet Union carried out large import-
substitution programmes through the regulation of domestic fuel and energy prices (di
rectly and via export tax) that subsidized all energy consumers. They also provided subsi
dies to agricultural enterprises. China and Vietnam (and to some extent Uzbekistan) car
ried out export-oriented industrial policy mostly by underpricing the exchange rate. It is
argued that export-oriented industrial policy via undervaluation of the exchange rate is
the best possible option to promote export-oriented catch-up development based on the
export of manufactured goods. It is especially needed for resource-rich countries (Azer
baijan, Kazakhstan, Russia, Turkmenistan, Uzbekistan) that are prone to ‘Dutch disease’.
Assistance to domestic producers by keeping domestic prices for fuel and energy low also
helps to stimulate growth, but at the cost of very high energy intensity.
Keywords: economic growth, economic diversification, industrial policy, exchange rate policy, industrial structure,
resource curse
27.1 Introduction: Industrial Policy and Eco

nomic Diversification
THE increase in the share of the service sector, especially of trade and finance, at the ex
pense of industry (de-industrialization) occurred in all post-communist economies. Previ
ously in the centrally planned economies, the service sector, in particular trade and fi
nance, was underdeveloped. In Russia, for instance, the share of industry in GDP fell from
about half in 1990 to about one-third by the mid-1990s, whereas in industry itself the
Page 1 of 34
rope and Asia
share of the primary sector (fuel, energy, steel, and non-ferrous metals) in total industrial
output increased from 25 per cent to over 50 per cent, and the share of secondary manu
facturing declined accordingly (Popov, 2011). Part of the decline was due to the change in
relative prices (domestic prices for resources and services increased more than for manu
factured and agricultural goods), but relative changes in physical output also played a
role—real decline in secondary manufacturing and agriculture was more pronounced
than in services and resource industries.
The structure of exports in most post-communist states has also become more primitive in
the past two decades—the share of manufactured goods in total exports either declined
or did not show any clear tendency towards increase. This was partly (p. 780) due to the
increase in resource prices and the resource boom—the expansion of fuel production and
exports in Azerbaijan, Kazakhstan, Russia, Turkmenistan, and Uzbekistan. In Russia, the
share of fuel, minerals, metals, and diamonds in total exports grew from 52 per cent in
1990 (USSR) to 67 per cent in 1995 and to 81 per cent in 2012, whereas the share of ma
chinery and equipment fell from 18 per cent in 1990 (USSR) to 10 per cent in 1995 and to
4.5 per cent in 2012. But the real shifts (in constant prices) were also in favour of re
sources at the expense of ready-made goods.
International comparison of the structure of the economy at different stages of develop

ment may provide some insights into what kind of structural shifts occur during the tran
sition from the agricultural to the industrial stage and to the post-industrial stage. The
Chenery hypothesis (Chenery, 1960) predicts changes in the relative shares of three ma
jor sectors of the economy (agriculture, industry, services) in the process of economic de
velopment. The data suggest that the share of agriculture constantly decreases, the share
of services constantly increases, whereas the share of manufacturing increases before
countries reach a level of per capita income of about 6,000 international dollars of 1990
(ln 6000 = 8.7), that is, about 12,000 in today’s prices, and then declines. To put it differ
ently, at a stage of development of below $12,000 GDP per capita in today’s prices, re
sources were transferred from agriculture to industry and services, whereas after the lev
el of $12,000 per capita GDP, resources were transferred from both agriculture and in
dustry to services (i.e. the service sector was growing at the expense of not only agricul
ture, but also industry) (Herrendorf et al., 2013: figure 1).
The level of development, of course, is not the only determinant of the structure of the
economy. The other determinants are the size of the economy and its resource endow
ment, as was pointed out in the early literature on the issue. Syrquin and Chenery (1989)
concluded that higher income growth and more marked transformation are found among
countries with large populations, a predominance of manufactures in exports, and a larg
er role of exports.
Changes in industrial structure are not the result of the ‘invisible hand of the market’
alone. As Greenwald and Stiglitz (1986, 2013) state, market failures are pervasive; pri
vate rewards and social rewards virtually always differ, so governments are inevitably in
volved in shaping the industrial structure of the economy, both by what they do and by
Page 2 of 34
rope and Asia
what they do not do. As many authors point out, the secret of ‘good’ industrial policy in
East Asia, as opposed to ‘bad’ industrial policy in the former Soviet Union, Latin America,
and Africa may be associated with the ability to reap the benefits of export externalities
(Khan, 2007; Gibbs, 2007). Exporting to world markets, especially to developed countries,
enables the upgrading of quality and technology standards and yields social returns in ex
cess of returns to particular exporters. The greatest increases in productivity are regis
tered at companies that export to advanced (Western) markets and that export high-tech
goods (Harris and Li, 2007; Shevtsova, 2012). It has been also shown that the gap be
tween the actual level of development (per capita income) and the hypothetical level that
corresponds to the degree of sophistication of a country’s exports is strongly correlated
with productivity growth rates (Hausmann et al., 2005). (p. 781) To put it differently, it
pays to promote exports of sophisticated and high-tech goods. Not all the countries that
try to promote such exports succeed, but those that do not try, virtually never engineer
growth miracles.1
It should be noted though that there is an opposite view—for instance, the recent World
Bank paper (Gill et al., 2014). It concludes that it is not clear whether diversifying exports
and production is necessary for development and that governments need to worry less
about the composition of exports and the profile of production and more about national
asset portfolios—the blend of natural resources, built capital, and economic institutions.
27.2 Overview of Changes in the Structure of

Economies in Post-communist Countries
After the fall of the Berlin Wall (1989), the collapse of the Union of Soviet Socialist Re
publics (USSR, 1991), and market-oriented reforms in successor states, the performance
in the post-Soviet space varied greatly (Figure 27.1). In retrospect, it is obvious that rapid
economic liberalization did not pay off: many gradual reformers from the former Soviet
Union in this region performed better than the champions of ‘big bang’ liberalization—the
Baltic states and Central Europe. In Turkmenistan and Uzbekistan, for instance, privatiza
tion was rather slow—over 50 per cent of their GDP is still produced at state enterprises,
but their performance is superior to that of more liberalized economies. Resource abun
dance has definitely helped resource exporters, such as Azerbaijan, Kazakhstan, and
Turkmenistan, to maintain higher incomes recently, when resource prices were high, but
was not a sine qua non for growth—resource-poor Tajikistan, as well as Uzbekistan, self-
sufficient in fuel and energy, did much better than resource-rich Russia (Figure 27.1).
As past research shows, the crucial factor in economic performance was the ability to
preserve the institutional capacity of the state (Popov, 2000, 2007, 2011 for a survey). The
story of transition was very much a government failure, not a market failure story.
(p. 782) In all former Soviet republics and in Eastern European countries, government
spending fell during transition and the provision of traditional public goods, from law and
Page 3 of 34
rope and Asia
order to health care and infrastructure, worsened. This led to an increase in crime, the
shadow economy, income inequalities, corruption, and mortality.
Figure 27.1 GDP change in economies of the former

Soviet Union, 1989 = 100%
Note: Central Europe is the unweighted average for

Czech Republic, Hungary, Poland, Slovakia, and
Slovenia.
Source: EBRD Transition Reports for various years.
But in countries with the smallest decline in government spending (countries very differ
ent in other respects—from Central European states to China and Vietnam and to Uzbek
istan), these effects were less pronounced, and the dynamics of output was better (Popov,
2011). China carried out gradual reforms from 1979, Vietnam, gradual reforms in 1986‒
89 and radical reforms in 1989, but both countries managed to avoid the (p. 783) transfor
mational recession2 due to their ability to preserve state institutional capacity (Popov,
2000).
All former communist countries had centrally planned economies and all of them inherit
ed distortions in industrial structure and interregional trade patterns created by central
planning. Central planners gave high priority to industry, especially heavy and high-tech
industries (at the expense of resources industries and agriculture) and to self-sufficiency
achieved through import-substitution policy. As a result, secondary manufacturing indus
tries took a high share of the national economy (as compared to other countries with simi
lar per capita incomes), but they were less efficient and competitive.
After the deregulation of prices and the opening of the economy, market forces were al
lowed to influence the allocation of resources and most secondary manufacturing and
high-tech industries proved to be uncompetitive, so their output shrank. All former com
munist countries (with the exception of China and Vietnam) experienced transformational
Page 4 of 34
rope and Asia
recession, de-industrialization and a decline in R&D potential in the 1990s after the tran
sition to the market economy. There was also rapid ‘resource-ialization’ after the transi
tion to the market in 1992. An increase in the share of the service sector, especially trade
and finance, at the expense of industry (de-industrialization) occurred in all post-commu
nist economies.
It seems, however, that in many of these economies de-industrialization went too far. In
Eastern European countries, especially in the countries of Central Europe (Czech Repub
lic, Hungary, Poland, Slovenia, Slovakia) the decline was less pronounced than in many
former Soviet Union (FSU) countries. In Tajikistan, for example, the share of services in
GDP nearly doubled, increasing from about 30 per cent in the early 1990s to 57 per cent
in 2010, whereas the share of manufacturing in GDP fell from 25 per cent in 1990 to 10
per cent in 2010 (Figures 27.2, 27.3, and 27.4). In Armenia, the share of industry fell from
47 per cent in 1991 to 25 per cent in 2017 and the share of manufacturing, from about 30
per cent to 10 per cent, whereas the share of services increased from about 20 to 30 per
cent in the late 1980s/early 1990s, and to over 50 per cent in 2017.
Figure 27.2 Share of manufacturing value added in

GDP in post-communist transition countries between
1985 and 2017, %
Source: World Development Indicators (there is no

data for Uzbekistan in WDI).
Page 5 of 34
rope and Asia
Figure 27.3 Share of industry value added in GDP of

post-communist countries, 1985–2017, %
Source: World Development Indicators.
Figure 27.4 Share of service value added in GDP in

selected post-communist countries, 1985–2017, %
In resource-rich FSU countries, the decline of the share of manufacturing was most dra
matic—it was driven not only by the expansion of the service sector, but also by the ex
pansion of the resource industries that became most competitive with the new price ra
tios. The contrast with China and Vietnam—two countries with active industrial policies
to support export-oriented manufacturing—was extremely pronounced. In Azerbaijan,
Page 6 of 34
rope and Asia
Kazakhstan, and Russia the share of manufacturing fell below the level of much poorer
Vietnam (15 per cent), not to mention China (29 per cent).
A comparison of Figures 27.2 and 27.3 is quite instructive. The share of manufacturing in
total value added (GDP) has declined in almost all post-communist countries (p. 784) and
is now at a very low level of 4 to 15 per cent (Figure 27.2), whereas in the late 1980s it
was at a level of about 20 to 30 per cent. And of course this share is way lower than in
China, where it has remained at a level of about 30 per cent in recent years, considerably
higher than in other developing countries.
The share of industry as a whole (which includes not only manufacturing, but also
(p. 785)
mining and utilities—electricity and gas distribution; Figure 27.3),3 as a rule of thumb, de
spite sharp fluctuations, did not decline much in resource-rich countries (Azerbaijan,
Kazakhstan, Russia, Turkmenistan), but declined in non-resource-rich countries. The only
exception is Uzbekistan, which has a medium resource abundance (self-sufficient in ener
gy), but managed to increase the share of industry to over 30 per cent after it fell from 35
to below 20 per cent in 1987‒2002 (Figure 27.3).
The structure of exports in most post-Soviet states has also become more primitive in the
last two decades; the share of manufactured goods in total exports either declined
or did not show any clear tendency towards increase. Perhaps surprisingly, Kyr
(p. 786)
gyzstan was the only country where the share of manufacturing exports in total exports
increased (Figure 27.5).
27.3 Structural Economic Transformation and

Industrial Policy
The crucial question, of course, is whether the government should participate in the
process of structural shifts (and if yes, then how), or whether the structure of the econo
my should be totally determined by market forces. Government preferences (tax, trade,
credit, other) for particular industries are called industrial policy and the huge literature
on industrial policy (see Popov, 2011, for a summary) focuses on three key issues:
(p. 787)
1. Is industrial policy necessary for successful development or does the market know
better how to allocate resources?
2. If industrial policy is needed, how should industries that need to be supported be
selected?
3. What are the appropriate tools/instruments to support particular industries?
Page 7 of 34
rope and Asia
Figure 27.5 Manufacturing exports, % of merchan

dise export, in some post-communist countries
To give one example of what industrial policy can do, consider a resource-rich country
that previously (under central planning) favoured secondary manufacturing and now (af
ter transition to the market and the change in relative prices of resources and ready-
made goods) is experiencing, under the pressure of market forces, the reallocation of cap
ital and labour from manufacturing to mining and primary manufacturing (oil, gas, elec
tric energy, diamonds, steel, non-ferrous metals). Available calculations of total factor
productivity show that in resource-rich countries, such as Azerbaijan, Kazakhstan, Russia,
Turkmenistan, Uzbekistan, it is much higher in mining than in other industries, with the
possible exception of trade and finance (Popov, 2000, 2014; Gharleghi and Popov, 2018b).
No wonder that capital and labour are being reallocated from non-resource to resource
industries. Should the government just observe the process without interfering, should it
oppose the market forces, or should it try to promote structural shifts already under way
to ensure that they happen faster? And what tools should be used to promote desirable
shifts: subsidies, cheap credits, tax concessions, import tariffs, or exchange rate manage
ment?
Page (2011) argues that Africa should industrialize because without structural
(p. 788)
change it cannot sustain recent growth. Economies with more diverse and sophisticated
industrial sectors tend to grow faster. But since 1980 Africa has de-industrialized. Page’s
paper shows that between 1975 and 2005 the size, diversity, and sophistication of indus
try in Africa have all declined. He argues that an industrialization strategy containing two
elements is needed:
▪ The first is straightforward: refocusing current investment climate reforms on infra

structure, skills, and regional integration. These actions alone will not be sufficient,
however.
▪ Africa must also learn to compete through strategies to create an export push, devel
op industrial clusters, and attract task-based production.
Page 8 of 34
rope and Asia
A recent IMF paper (Cherif and Hasanov, 2019) highlights the key role of industrial policy
in the Asian miracles, for the United States after the Civil War, for Germany under Bis
marck, and for Japan after the Meiji Restoration. Jomo (2019) argues that East Asian in
dustrial policies contributed to the economic miracle stories and, despite the current con
ventional wisdom, can be successfully emulated by other developing countries.
Stiglitz (2018), however, states that export-led growth was the model behind the twenti
eth-century growth miracles. There was unprecedented growth in East Asia—closing the
gap in income per capita and standards of living with the advanced countries. That mod
el, he claims, will not work in the future in the way and to the extent that it did in the
past. He suggests a multi-pronged strategy for developing countries, entailing a combina
tion of manufacturing, agriculture, services, and natural resources.
It is always difficult, of course, to predict the efficiency of the growth model. The famous
example is that of Gunnar Myrdal (1968) who in his Asian Drama was extremely pes
simistic about the prospects of Asian future growth exactly at a time when Asian dragons
and tigers were taking off to propel themselves out of poverty and become economic mir
acles.
27.4 Industrial Policy: Which Industries to Sup

port?
As the UN ESCAP (2014) report on diversification states:
Implementing strategic diversification requires an industrial policy—the selective

promotion of particular economic activities over others. Here, new economic activ
ities should be promoted that would encourage greater levels of product (p. 789)
complexity and allow for further diversification in the future. Active public inter
vention is required that is aimed at supporting infant industries and creating the
necessary complementary productive infrastructure, including industrial estates
and economic zones. Intervention would also be aimed at encouraging marketing
and export market development, together with other promotional measures under
industrial policy.
(UN ESCAP, 2014: viii)
Not only is industrial structure shaped by the development process; it also has important
implications for economic development. The Chenery hypothesis (Chenery, 1960) states
that countries at similar levels of economic development should have similar patterns of
resource allocation between sectors. But in theoretical models it is often assumed that
there are externalities from industrialization and industrial export (Murphy et al., 1989;
Polterovich and Popov, 2004, 2005). There is growing evidence that countries which are
more industrialized and countries with more technologically sophisticated industrial ex
ports are growing faster than others (Hausmann et al., 2005; Rodrik, 2006).
Page 9 of 34
rope and Asia
The important stylized fact is that no economic miracle in the developing world was
based either on agricultural or service industries. A Canadian economist, Harrold Innis,
was the author of the staple theory of economic development (Innis, 1956). He claimed
that Canadian economic (and not only economic, but also social and cultural) develop
ment was determined by exports of staple goods, in chronological order since the seven
teenth century: furs, fish, lumber, wheat, mined metals, and coal. It could be also claimed
that some countries that are now members of the ‘rich-country club’ made their fortunes
on resource and agricultural exports; examples would be (in addition to Canada) Aus
tralia, New Zealand, and the United States. But by the twentieth century, after the in
come gap between developed and developing countries widened, cases of successful
catch-up development (Popov and Jomo, 2017) were associated with manufacturing ex
ports, not with exports of agricultural or resource commodities. Oil-rich countries, like
Persian Gulf states or Equatorial Guinea, which have now achieved developed-country
levels of per capita income, did so due to terms-of-trade improvement (increase in fuel
prices), not due to exceptionally high rates of growth of output (Oman is the exception,
see footnote 1).
The reduction in the share of industry and manufacturing in GDP and the increase in the
share of services is an objective process; but in fast-growing countries (e.g. China), this
decline was slower than in others with a similar level of development (Figures 27.2, 27.3,
and 27.4). The increase in the share of machinery and equipment in manufacturing out
put, as in China, usually accompanies rapid growth or even becomes the engine of
growth. We are not aware of cases of rapid growth (‘economic miracles’) that are based
on accelerated growth of the service sector.
The question ‘What are the particular manufacturing industries which could become the
engine of growth?’ is a difficult one (Popov and Chowdhury, 2016). Unfortunately, eco
nomic theory does not suggest any definite clues, with the exception of the idea that
these industries should have the highest externalities, that is, their social returns should
be higher than private returns. Yet, it is not easy to measure these externalities.
(p. 790)
Nevertheless, upon examination of the literature and the experience of countries with in
dustrial policy, it is possible to isolate methods which can aid in identification of indus
tries that should be supported.
For example, one can support several industries which seem promising on condition that
assistance ends if the increase in exports is not achieved within, say, five years. This is
called ‘EPconEP’—effective protection conditional on export promotion (Jomo, 2013). Eco
nomic policymakers in this case are similar to the military commander who begins an of
fensive on several fronts, but throws reserves at where there has been a breakthrough.
Alternatively, one can attempt to determine specific industries where limited investment
can produce the greatest effect, leading to the creation of globally competitive produc
tion. These are most likely to be industries that lag behind the most advanced countries
in total factor productivity, but not so much as other industries.
Page 10 of 34
rope and Asia
It is also possible to choose at random. In this case, it is important to be consistent by em
barking on the path of support for a particular industry and not withdrawing even if there
is no immediate success or breakthrough in world markets. After all, the modern theory
of international trade explains country specialization not by comparative advantages, but
rather by ‘learning by doing’.
If a country does not have any comparative advantage, like post-war Japan for example, it
is necessary to create them (‘dynamic comparative advantages’) by mastering the produc
tion of goods that have not been produced before. Supporting such production and con
sistently encouraging exports for some time without giving up is likely to have the learn
ing-by-doing effect, allowing the country to gradually become competitive.
There are two opposing views on how technologically advanced industries supported in
the framework of industrial policy ought to be. Justin Lin, former chief economist of the
World Bank, developed the idea of comparative-advantages-following (CAF) and compara
tive-advantages-defying (CAD) industrial strategy. The best result, according to his argu
ment, can be achieved if countries develop industries that are consistent with their com
parative advantages, as determined by their endowment structure, and do not try to by
pass necessary stages aiming at exporting the goods which are exported by very ad
vanced countries (Lin, 2011). Middle-income oil-rich countries, like Kazakhstan and Azer
baijan, for instance, according to this logic, should aim at developing heavy chemical, not
computer industries.
This view is consistent with the ‘flying geese’ paradigm: as more competitive countries
move to more advanced types of exported products, the vacated niches are occupied by
less developed countries. It is known that relatively poor countries began to export tex
tiles and shoes, then moved on to the export of steel products and heavy chemicals, then
to the export of cars and electrical consumer products such as washing machines and re
frigerators, then to consumer electronics and computers. Newcomers can benefit from
the experience of other countries by trying to replicate their success.
The transition from one exported good to another could be dictated by the cycle
(p. 791)
of innovations. As Lee (2013) suggests, the cycle is short for electronics and long for
pharmaceuticals and chemicals. This may explain why East Asian countries which fo
cused mostly on industries with short cycles managed to avoid growth slowdowns while
moving from one export niche to another.
The debate between Justin Lin and Ha-Joon Chang (Lin and Chang, 2009) is telling in this
respect. The latter was defending the idea of CAD industrial policy which favours indus
tries that defy the country’s comparative advantages. Such industries take time to devel
op, yet they can be worthwhile. For example, ‘Japan had to protect its car industry with
high tariffs for nearly four decades, provide a lot of direct and indirect subsidies, and vir
tually ban foreign direct investment in the industry before it could become competitive in
the world market. It is for the same reason that the electronics subsidiary of the Nokia
group had to be cross subsidized by its sister companies for 17 years before it made any
Page 11 of 34
rope and Asia
profit. History is full of examples of this kind, from eighteenth-century Britain to late
twentieth-century Korea’ (Lin and Chang, 2009).
Hausman, Hwang, and Rodrik’s position (Hausmann et al., 2005; Rodrik, 2006) may be
rather different from Chang’s CAD strategy and the policy of promoting high-tech indus
tries and R&D in relatively poor countries. The CAD strategy does not necessarily imply a
transition to more technologically sophisticated industries, but rather, to industries that
are not linked to the comparative advantages of a particular country. Theoretically, it
could be a transition from chemicals to machine building with the same, or even a lower,
level of R&D intensity and technological sophistication. Hausman, Hwang, and Rodrik’s
idea is that externality benefits from the production and export of new products are pro
portional to the degree of their technological sophistication, which is measured by the
comparison of the export structures of rich versus poor countries. High-income countries
export on average more high-tech products. Developing high-tech production in poor
countries may be costly, yet the returns from such a policy can be greater. It may well pay
a relatively poor country to make ‘a big leap forward’ by investing heavily in the educa
tion of the labour force and high-tech industries, bypassing the intermediate stages of
producing goods with medium research intensity.
Uzbekistan, for instance (a low-income country, not even a middle-income country), start
ed to invest massively in the development of the auto and heavy chemical industries at
the stage of development at which other countries export mostly resource goods, textiles,
and leather products. It remains to be seen whether these investments will be justified
(see Box 27.1).
Box 27.1 What Uzbekistan knows about industrial policy that other countries do not
know
In the last two decades the economy of Uzbekistan has been extremely successful—high
growth (8 per cent), low unemployment and reasonable macroeconomic stability, low do
mestic and international debt, relatively low inequality. Even more impressive are the
structural shifts that have happened in less than thirty years since Uzbekistan became an
independent country: (1) decrease in production and export of cotton (which was previ
ously a mono culture), increase in food production and achievement of self-sufficiency in
food; (2) achievement of self-sufficiency in energy and becoming a net fuel exporter; (3)
increase in the share of industry in GDP and the share of machinery and equipment in in
dustrial output and export (a competitive export-oriented auto industry was created from
scratch) (Popov, 2013, 2014).
In recent years Uzbekistan has promoted heavy chemical industries (production of syn
thetic fuel and polypropylene goods from natural gas). This is the next stage of industrial
policy after reaching food and energy self-sufficiency and successful auto-industry devel
opment. In 2011 it became the fifteenth country in the world to launch a high-speed train
line, between Tashkent and Samarkand (it was continued to Bukhara and Karshi in 2015
Page 12 of 34
rope and Asia
and 2016). The train is made by Spanish company Talgo and travels the 600 km between
Tashkent and Bukhara in 3 hours 20 minutes.
Uzbekistan’s development achievements, even though not as spectacular as those of Chi

na, have been due to deliberate government policies. True, Uzbekistan enjoyed a
favourable external environment, but its rapid growth is due to reasonable macroeconom
ic stability and industrial policies rather than simply to market reforms that triggered
growth conforming to its factor endowment and/or natural comparative advantages.
What makes Uzbekistan different and even unique is its low exchange rate policy. It pro
motes export-oriented development, similar to Japan in the 1950s to 1970s, South Korea
in the 1960s to 1980s and China and the ASEAN countries since the 1990s (Dollar, 1992;
Easterly, 2001; Polterovich and Popov, 2004; Rodrik, 2008; Bhalla, 2012). Undervaluation
of the exchange rate via accumulation of foreign exchange reserves becomes a powerful
tool of industrial policy creating stimuli for tradable goods at the expense of non-trad
ables (Greenwald and Stiglitz, 2013). Other former communist countries of Eastern Eu
rope and the USSR did not implement such a policy; on the contrary, their exchange rates
were and often are overvalued, especially in countries that export resources (they suffer
from the ‘Dutch disease’).
Since 2000 Uzbekistan has probably been the only country in the post-Soviet space that
has carried out gradual, predictable nominal devaluation of the currency to a slightly
greater extent than is needed to counter the differences in inflation rates between Uzbek
istan and its major trading partners, so that the real effective exchange rate is depreciat
ing slowly. The real exchange rate of the som versus the US dollar has appreciated a bit,
though not as much as currencies of other countries. However, the real effective
exchange rate (i.e. with respect to currencies of all major trading partners) decreased by
over 50 per cent in 2000‒07—a sharp contrast with other countries of the region on
which data are available.
The relatively successful economic performance of the country is even more impressive
given that Uzbekistan is not a major oil and gas exporter and is one of two doubly land
locked countries in the world—that is, a country completely surrounded by other land
locked countries (the other being Liechtenstein).
Uzbekistan remains a poor country, with PPP GDP per capita of below US$6,000 in 2014
against over $20,000 in Russia and Kazakhstan, $17,000 in Azerbaijan, and over $14,000
in Turkmenistan; and many Uzbeks are migrating to find work in Russia, not vice versa. It
is important, however, to distinguish between growth rates and the level of per capita in
come. It is necessary to separate the effects associated with the dynamics of output from
the effects of the terms of trade and financial flows. At the end of the Soviet era, in the
1980s, real incomes in Uzbekistan were about half those in Russia. After the collapse of
the USSR real incomes in non-resource republics fell dramatically due to the change in
relative prices—oil, gas, and other resources became several times more expensive rela
tive to ready-made goods (Uzbekistan was a large importer of oil and its trade with all
countries, including other Soviet republics, if recalculated in world prices, yielded a
Page 13 of 34
rope and Asia
deficit of 9 per cent of GDP; IMF, 1991). To add insult to injury, with the collapse of the
Soviet Union financial flows from Moscow dried up (in 1990 inter-budgetary transfers—
from the Union budget—amounted to 31 per cent of the revenues of the republican bud
get; IMF, 1991).
Diversification in industry and expansion of manufacturing exports was mostly the result
of protectionism and the government/central bank’s low exchange rate policy. Like China,
Uzbekistan maintained a low (undervalued) exchange rate due to rapid accumulation of
foreign exchange reserves. In addition, there were non-negligible tax measures to stimu
late exports of processed goods (50 per cent lower tax rate for manufacturing companies
that export 30 per cent or more of their output). Although comparable statistics from WDI
for Uzbekistan are lacking, national statistics suggest that the share of non-resource
goods in exports increased to over 70 per cent against less than 30 per cent in 1990 be
fore independence (Popov, 2013, 2014).
Rodrik et al. (2016) consider two sources of productivity growth: within the industry; and
due to structural shifts, that is, reallocation of resources to more productive industries.
Which is more efficient: relying on productivity growth within existing industries or pro
moting structural shifts from less productive to more productive industries?
In a similar vein, Rodrik (2012) describes two approaches to development: bottom up and
top down. The former focuses directly on the poor, and on delivering services, such as ed
ucation, health care, and microcredit, to their communities. As Reddy (2013) (p. 792)
(p. 793) explains, this tradition’s motto could be: ‘Development is accomplished one
project at a time.’ The other approach takes an economy-wide perspective. It emphasizes

broad reforms that affect the overall economic environment, and thus focuses on areas
such as international trade, finance, macroeconomics, and governance.
The first approach uses widely randomized controlled trials as an instrument that could
allow good policies to be formulated: vaccinations and microcredit, additional teachers in
schools, and mosquito nets dipped in insecticide being examples of small projects that are
leading to big breakthroughs. But without reforms at the macro level it is often impossi
ble to ensure the efficiency of micro projects (Reddy, 2013). If assistance provided for par
ticular investment projects merely crowds out government or private investment in other
areas, the macro impact of the assistance will be zero.
As Rodrik (2012) writes, ‘poverty is often best addressed not by helping the poor be bet
ter at what they already do, but by getting them to do something different’. This latter ap
proach is exactly the one defended here: the global South can gain much more from econ
omy-wide reforms aimed at promoting export-oriented growth based on domestic savings
than from meagre official foreign assistance or even from all foreign financing. A benevo
lent Western attitude to macro-structural reforms and industrial (p. 794) policy would be
more beneficial to catch-up development than a thousand minor specific development
projects with noble goals.
Page 14 of 34
rope and Asia
A UN ESCAP study (2014) suggests that there is a strong link between the share of manu
facturing in GDP and the share of the poor in total population. It predicts that industry-
oriented structural transformation that enhances agricultural productivity through sus
tainable agriculture and overall efficiency improvements through innovations has the po
tential to lift an additional 71 million people out of poverty, create 56 million additional
jobs in South Asia and boost GDP by 15 to 30 per cent by 2030 over and above the busi
ness-as-usual scenario.
27.5 Industrial Policy: What Tools to Use?

Industrial policy tools can be divided into two broad categories: selective and non-selec
tive. Selective tools are those that apply to specific industries, regions, and companies.
Examples are import duties, subsidies, tax concessions. Non-selective tools are govern
ment investment in infrastructure, education, health care, law and order, and so on, that
helps to create a better business climate for all businesses. Management of the exchange
rate is another important tool of non-selective industrial policy—the maintenance of the
undervalued real exchange rate via accumulation of foreign exchange reserves (above the
normal amount needed to ensure smooth trade and capital account transactions) is an im
portant instrument for promoting economic growth based on export of tradable goods, al
though at the expense of non-tradables (Polterovich and Popov, 2004).
There are important differences between import duties and devaluation of the exchange
rate. As Larry Summers once observed: ‘A ten percent decline in the dollar exchange rate
is equivalent to a ten percent tariff on all imported goods and a ten percent subsidy for all
exported goods’ (The New Republic, 25 January, 1988, p. 14). Import duties raise the real
exchange rate (level of prices in the country as compared to the world), whereas real de
valuation lowers the real exchange rate. Besides, ‘exchange rate protectionism’ is a more
efficient policy to stimulate growth because decisions on import duties and government
taxes/spending are affected by poor quality of institutions (corruption and inefficiency of
implementation), whereas low exchange rate policy is indiscriminate and non-selective by
nature: it cannot be captured and ‘privatized’ by particular interest groups, which makes
it an especially efficient growth-promoting instrument in poor and middle-income coun
tries that generally suffer from corruption (Polterovich and Popov, 2004).
As the UN flagship report states (UN WESP, 2016), ‘reserve accumulation can have posi
tive externalities on the production and export of tradables and industrial development
and can thus be a feature of the country’s development model. Undervaluation (p. 795) of
the exchange rate can increase the competitiveness of exports, without the need for sec
tor- or firm-specific subsidies or interventions.’
As Griffith-Jones and Ocampo (2010) observe, the rationale for the accumulation of for
eign exchange reserves ‘is usually found in either one of two explanations: the
“competitiveness” (or, in more pejorative terms, “mercantilist”) and the “self-insurance”
motives’. This mercantilist view that undervaluation of the exchange rate via accumula
tion of foreign exchange reserves is in fact an industrial policy, aimed at promoting ex
Page 15 of 34
rope and Asia
port-oriented growth by benefiting the producers of tradables and exporters at the ex
pense of the producers of non-tradables and importers, which is gaining support in the lit
erature (Dollar, 1992; Easterly, 2001; Rodrik, 2008; Bhalla, 2012; Greenwald and Stiglitz,
2013). If there are externalities from the export and production of tradables (industrial
ization, development of high-tech sectors), undervaluation of the exchange rate resulting
from the accumulation of reserves is an efficient way to provide a subsidy for these activi
ties and this subsidy is automatic, that is, it does not require a bureaucrat to select possi
ble beneficiaries.
In short, this is a non-selective industrial policy promoting export and production of trad
ables that seems to be quite efficient, especially in countries with high corruption and
poor-quality institutions. Thus, accumulation of reserves and undervaluation of the ex
change rate may be good for long-term growth. The formal model demonstrating how the
accumulation of reserves can spur growth, together with empirical evidence, is presented
in Polterovich and Popov (2004), where it is also shown that, in practice, accumulation of
reserves leads to a disequilibrium exchange rate, which in turn causes an increase in ex
port/GDP and trade/GDP ratios, which stimulates growth. There is strong evidence that
accumulation of reserves can spur long-term growth in developing countries, although
not in rich countries (Polterovich and Popov, 2004).
In practical terms, there are no formal limits to the accumulation of reserves by develop
ing countries, but ‘exchange rate protectionism’ can result in ‘beggar-thy-neighbour poli
cies’: obviously all countries cannot exercise these policies at the same time to achieve
undervaluation of their exchange rates. If all countries use these policies, they will all
lose, and moreover, it does not work for developed countries. But for developing coun
tries it does work, and there are good reasons why these countries should have sufficient
policy space to use this tool to promote catch-up development.
Trade surpluses resulting from undervaluation of the exchange rate due to reserve accu
mulation may lead to what are now called ‘global imbalances’, driving the other countries
into debt. However, there is some room for such a trend, which will reverse the opposite
trend of the nineteenth and twentieth centuries (the United States enjoyed a trade sur
plus for nearly a hundred years after the Civil War of the 1860s, driving many developing
countries into debt; see Popov, 2010a, 2010b, 2011, for details).
The policy of reserve accumulation is often considered to be self-defeating because in or

der to avoid inflation (which would eat up the impact of devaluation on the real exchange
rate) it is necessary for the monetary authorities to carry out a sterilization (p. 796) policy,
that is, to sell government bonds in order to neutralize the impact on the money supply of
the purchase of foreign currency. But sales of government bonds lead to higher interest
rates, which in turn attract capital from abroad that contributes to an increase in foreign
exchange reserves, which should be sterilized again, creating a vicious circle. That is why
economists talk about an ‘impossible trinity’: a country cannot maintain at one and the
same time an open capital account, a managed exchange rate, and an independent mone
tary policy.
Page 16 of 34
rope and Asia
But many developing countries exercise control over capital flows (China and India are
prime examples) and even without such control, capital mobility—especially for large
economies—cannot be considered perfect. In practice, as the statistics show, the accumu
lation of foreign exchange is financed through government budget surplus and debt accu
mulation, but not through printing money. That is to say, most countries that have accu
mulated reserves rapidly exhibit low inflation and low budget deficit (or budget surplus),
but increased holdings of government bonds by the public (see Polterovich and Popov,
2004).
27.6 Industrial Policy in Resource-rich Coun

tries
Resource abundance should, logically, be a plus for economic development, but very often
it becomes a constraint—a ‘resource curse’ rather than a ‘resource blessing’ (Sachs and
Warner, 2001; Sala-i-Martin and Subramanian, 2003; Stiglitz, 2005; Polterovich et al.,
2007, 2008, 2010). Whereas resource-rich countries have generally overvalued their ex
change rate (‘Dutch disease’), they also maintain a relatively low level of domestic fuel
prices (Polterovich et al., 2007, 2008, 2010). This is another important industrial policy in
strument that has at least two implications: first, like the undervaluation of the RER, low
domestic prices for fuel provide competitive advantages to domestic producers and stimu
late exports and production (especially of energy-intensive products); second, low fuel
prices lead to energy waste due to higher energy intensity, and hence imply higher costs.
Today in all resource-rich post-communist countries, Russia being the prime example, do
mestic prices for fuel are kept below the world market level through taxes on exports of
fuel and direct restrictions on exports (such as access to the pipeline). As a result, domes
tic prices for oil and gas are considerably lower than in the rest of the world, and this al
lows energy-intensive industries to flourish. In Russia, for instance, the production of en
ergy-intensive aluminium is very competitive due to low energy prices—aluminium is one
of Russia’s top export commodities, even though half of it is produced from imported
bauxite.
The argument developed in Polterovich et al. (2007, 2008, 2010) is that the undervalua
tion of the exchange rate is a preferred industrial policy tool for resource-rich (p. 797)
countries, compared to keeping domestic fuel and energy prices below the world market
level. It allows export-oriented development to be stimulated without high energy intensi
ty. In order to make a transition to a new policy, a delicate policy manoeuvre is needed.
Theoretically it is possible, but it requires bureaucratic competence to achieve the follow
ing:
• Gradual increase in domestic fuel and energy prices (by phasing out export tax and
lifting restrictions on access to pipelines) up to world levels;
• Higher taxes on fuel companies to capture windfall profits from increasing domestic
fuel prices;
Page 17 of 34
rope and Asia
• Spending of increased budget revenues on infrastructure and non-tradables;

• Lower real exchange rate (via accumulation of foreign exchange reserves and import
subsidies) to compensate losses of non-fuel industries from higher domestic fuel
prices.
27.7 Innovation, Research, and Development

Should countries which are behind the technological frontier invest in the adoption of ex
isting technologies or should they aim to develop completely new technologies and prod
ucts? The notion of the ‘advantages of backwardness’ introduced by Alexander Ger
schenkron (1962) implies, among other things, that relatively backward economies can
grow rapidly by investing in and adopting already existing technologies.
Justin Lin believes that countries should not leapfrog over the consecutive stages by go
ing from processing agricultural goods directly to high-tech industries (see section 27.6).
He suggests, for instance, that Uzbekistan could gain greater benefits by developing less
sophisticated industries such as food, textiles, and leather goods.4 The arguments
against, however, are supported by the examples of Israel and South Korea, which at the
end of the twentieth century had mastered the production of high-tech goods (electron
ics) and are now leading the world in R&D expenditure as a share of GDP (Figure 27.6).
Ricardo Hausmann, Jason Hwang, and Dani Rodrik (Hausmann and Rodrik, 2006; Haus
mann et al., 2005; Rodrik, 2006) hypothesize that the more technologically sophisticated
the export structure of a country is, the greater the stimuli for economic growth. China in
1992 and 2003, for example, had the greatest gap between hypothetical per capita in
come (computed based on the technological sophistication of its export structure) and the
actual per capita income. That is to say, the structure of (p. 798) Chinese exports was simi
lar to that of countries with several-fold higher levels of economic development, and this
gap stimulated economic growth.
Page 18 of 34
rope and Asia
Figure 27.6 R&D (research and development) Ex

penditure in selected countries, % of GDP
In another article (Hausmann and Rodrik, 2006), the process of transition from the pro
duction and export of one group of goods to another is compared to the movement of
monkeys in a forest from closer to more distant trees in search of food. The trees rich
with fruits are far away, whereas closer trees do not have as much. Thus, the monkeys
must compare the costs of movement with the benefits of reaching the more fruit-abun
dant trees. Like the monkeys, firms and society as a whole must compare the cost of mas
tering the new output and export (low for ‘nearby’ industries which are close to existing
technologies and high for ‘far-away’ industries with totally new technological processes)
with the benefits (externalities) associated with developing particular industries (theoreti
cally, the more sophisticated these industries are, the higher the benefits).
Today the share of R&D expenditure in GDP in most post-communist economies is lower
than 1 per cent (except for several Eastern European countries, Russia and China; Table
27.1), no higher than in countries with a similar level of economic development, whereas
in the communist past the opposite was normally true. Determining the optimal level of
R&D spending at different stages of development is an important research question. The
answer is probably specific to every country: it depends on the size of the country, the
structure of the national economy and the share of industries that are close to the techno
logical frontier.
Page 19 of 34
rope and Asia
Table 27.1 R&D expenditure in post-communist countries as a per

centage of GDP, 2015
Country R&D as a % of GDP
Slovenia 2.2
China 2.1
Czech Republic 1.9
Estonia 1.5
Hungary 1.4
Slovak Republic 1.2
Russian Federation 1.1
Lithuania 1.0
Poland 1.0
Bulgaria 1.0
Serbia 0.9
Croatia 0.9
Latvia 0.6
Ukraine 0.6
Belarus 0.5
Romania 0.5
Macedonia, FYR 0.4
Vietnam 0.4
Moldova 0.4
Page 20 of 34
rope and Asia
Georgia 0.3
Armenia 0.3
Azerbaijan 0.2
Bosnia and Herzegovina 0.2
Uzbekistan 0.2
Albania 0.2
Kazakhstan 0.2
Mongolia 0.2
Kyrgyz Republic 0.1
Tajikistan 0.1
Source: World Development Indicators, World Bank.
27.8 State versus Private Sector, Government,

Private, and Foreign Investment
There is a debate as to what is the crucial factor behind economic fiascos: market failure
or state failure? The dominant story in the profession is that economic breakthroughs are
achieved only due to a vital and vivid private sector that is dynamic and entrepreneurial,
oriented towards innovation and not afraid of risk-taking, whereas the state is clumsy, in
efficient, and even reactionary, and restricts private initiatives. It is said that the private
(p. 800) sector contributes to economic growth and poverty eradication through the build
ing of productive capacity, creation of decent jobs, promotion of innovation, economic di
versification, and competition. In landlocked developing countries, the private sector is
actively involved in activities related to transit and trade facilitation, including as traders,
freight forwarders, insurance providers, and transporters, and the sector is a source of
tax revenue and domestic investment and is a partner for foreign direct investment. Pub
lic‒private partnerships can play an important role in infrastructure development (Almaty
Programme of Action for Landlocked Developing Countries, 2013).
Another story, however, is that of the entrepreneurial state: Mazzucato (2013) provides
ample evidence that technological breakthroughs are due to public and state-funded in
Page 21 of 34
rope and Asia
vestments in innovation and technology, and that the private sector only finds the courage
to invest after an entrepreneurial state has made the high-risk investments.
Rodrik and Subramanian (2004) describe the attitudinal shift on the part of national gov
ernments towards a pro-business (as opposed to pro-liberalization) approach, and at
tribute the acceleration of Indian economic growth to this factor: they show that the ac
celeration of growth has actually occurred since 1980 and not since 1991, when liberal
ization reforms were carried out.
The problem in many developing countries in general and in post-communist countries in

particular is that the private sector often fails to take the initiative in promoting develop
ment due to an actual or alleged ‘poor investment climate’. As can be seen from Figure
27.7 (and comparing it to Figure 27.1), there is no correlation between the share of the
private sector in the economy and the GDP dynamics. Or, if there is a correlation, it is
rather negative than positive: more privatized economies are doing worse than less priva
tized ones.
Some studies suggest that government investments do not crowd private investment, but
have a ‘crowding in’ effect. As can be seen from Figure 27.8, it is not only private but also
public investments that contribute to the increase of the share of investment in GDP. If for
some reason private investments are in limbo, the state can achieve an increase in total
investments through the expansion of its own public investment projects financed
through taxation and/or borrowing. Government savings (financing public investment
through the government budget and/or budget surplus), as studies show, do not crowd
out private savings in a proportion of 1:1, but only in a proportion of 25‒50 cents for
every dollar (Schmidt-Hebbel et al., 1996). In low-income countries, as recent research
shows, one extra dollar of government investment does not crowd out, but crowds in pri
vate investment, raising it by roughly two dollars and output by 1.5 dollars (Eden and
Kraay, 2014).
If the private sector is not doing the job, the solution may be that government investment
and government entrepreneurship helps to resolve the bottlenecks.
It should be also noted that the impact of foreign direct investment (FDI) on development
is not always positive. Some countries created growth miracles without relying on FDI
(Japan, South Korea, China in 1979‒90), others relied on FDI extensively (Taiwan, Singa
pore, Hong Kong, China after 1990) (Polterovich and Popov, 2005).
Page 22 of 34
rope and Asia
Figure 27.7 The share of private sector in GDP in

some former Soviet republics, 1989–2009, %
Source: Transition Report, EBRD (data for years af

ter 2010 unavailable).
Figure 27.8 Public, private, and gross investment in

developing countries as a % of GDP, 2012
It may be hypothesized that FDI inflows into countries with a poor investment climate ac
tually do actually more harm than good. First, investors are self-selecting: if the invest
ment climate is bad, foreign investors come mostly for short-term profit and/or resource
projects, where the transfer of technology, the main benefit of FDI, is at best (p. 801) limit
ed. Second, foreign investors do not reinvest profits in countries with a poor investment
climate, so the outflow of profits with time outweighs the inflow of FDI. Third, purchases
Page 23 of 34
rope and Asia
of companies in countries with bad investment conditions do not necessarily lead to an in
crease in total investment because the inflow of FDI is often completely absorbed by an
outflow of short-term capital.5 (p. 802) (p. 803)
27.9 International Trade and Trade Policy

Trade specialization (the predominance of particular goods in exports and imports) is the
other side of the coin of the structure of a national economy: if the country exports manu
factured goods and imports food products, the share of manufacturing in its total output
and employment would be high and the share of agriculture low. That is why the struc
ture of the national economy depends to a large extent on trade developments—openness
to trade in particular sectors and industrial policies to promote exports of some goods
and to provide (or not to provide) protection to particular industries.
27.9.1 Trade Liberalization
What kind of trade policy is best for promoting sustainable development? The debates on
whether free trade or protectionism are more conducive to growth are as old as economic
research itself. The advocacy of free trade became commonplace in economics research
and a number of studies demonstrate its benefits (Sachs and Warner, 1995; IMF, WB, and
WTO, 2017; OECD, 2017 to name a few). However, it was noticed that free trade does not
always lead to an increase in the share of exports and imports in GDP (Rodriguez and Ro
drik, 2001; Polterovich and Popov, 2005). The best example is China after the Opium Wars
that forced the country to open up: for a hundred years afterwards, there was no increase
in the external trade-to-GDP ratio and China continued to fall behind Western countries in
per capita income. Other more recent examples are Japan, South Korea, and Taiwan in
the 1950s‒1960s and China in the 1980s‒1990s. In the 1980s Chinese import duties were
at the 40-50 per cent level and it was only in the 1990s and 2000s they were gradually re
duced to 10 per cent (Prasad, 2004). China became a member of the WTO only in 2001,
so GDP growth rates of about 10 per cent a year in the previous two decades (and of the
export-to-GDP ratio from 5 per cent in 1979 to 20 per cent in 2000) cannot be explained
by the benefits of trade liberalization.
Studies also show that the benefits of free trade depend on domestic adjustment policies,
without which it is impossible to reap the potential benefits of trade liberalization (IMF,
WB, and WTO, 2017; OECD, 2017). Stiglitz and Charlton (2005) question the conclusion
that broad trade liberalization always makes countries better off. They claim that this
conclusion is based on invalid assumptions about full employment and perfect competi
tion that are lacking in most developing countries.
It is well documented that fast-growing countries are usually more involved in interna
tional trade, and have higher and faster-growing trade/GDP ratios. In addition, there is a
correlation between the share of investment in GDP and the share of export in GDP—
countries which export more invest more as well (Polterovich and Popov, 2006). (p. 804)
However, fast-growing and more intensively trading nations are not always and were not
Page 24 of 34
rope and Asia
always more open to trade (low tariff and non-tariff barriers) than their less globalized
competitors.
Empirical studies (Rodriguez and Rodrik, 2001; O’Rourke and Williamson, 2002; see, for a
survey, Williamson, 2002) have found that there is no conclusive evidence that free trade
is always good for growth: whereas protectionist countries grew more rapidly before the
First World War, they exhibited lower than average growth after the Second World War.
Rose (2002) estimated the effect on international trade of multilateral trade agreements
such as the WTO, its predecessor the General Agreement on Tariffs and Trade (GATT),
and the Generalized System of Preferences (GSP) extended from rich countries to devel
oping countries, using the standard ‘gravity’ model of bilateral merchandise trade. He
found little evidence that countries joining or belonging to the GATT/WTO have different
trade patterns than outsiders, whereas the GSP, which gives poor countries better access
to markets in developed countries, had a very strong effect on the trade of developing
countries (an approximate doubling of trade).
27.9.2 Import Substitution versus Export Orientation
Import substitution is usually associated with protectionism: the idea is to protect domes
tic non-competitive industries with trade barriers, so that they can withstand competition
from imported goods and eventually increase their competitiveness and efficiency. The
problem with this kind of policy is that the efficiency of protected industries does not in
crease automatically, and protection that is usually designed as a temporary measure
very often becomes a permanent shield preserving industrial ‘dinosaurs’, as inefficient
and non-competitive enterprises are often called.
Export orientation is usually understood as a policy of support for exporters (via subsi
dies, credits, tax concessions, and other stimuli), but it is important to realize that export
promotion can in fact go hand in hand with protectionist measures. In fact, there is no
contradiction in imposing high import duties on a particular product and providing export
subsidies for it—higher prices in the domestic than the world market provide finances for
restructuring, whereas export orientation in this case is supported by export subsidies.
This was exactly the policy of Japan, South Korea, and Taiwan (even though not that of
Singapore and Hong Kong) in the 1950s‒1970s.
The difference between import substitution and export orientation is that the first policy
only protects domestic industries (without promoting export orientation), whereas the
second policy not only protects domestic industries, but also stimulates them to export
their output. The criteria for distinguishing between the two types of policy are the dy
namics of the share of export in GDP. If it grows fast, this should be classified as export-
oriented development, if it stagnates it is import substitution.
The authors of The East Asian Miracle (World Bank, 1993) found that government
(p. 805)
efforts to promote specific industries (without promoting exports) generally did not in
crease economy-wide productivity. But government support for exports was a highly ef
Page 25 of 34
rope and Asia
fective way of enhancing the absorption of international best-practice technologies, thus
boosting productivity and output growth.
In the former Soviet Union region, much trade liberalization occurred in the 1990s and
beyond as former Soviet republics made the transition to the market and deregulated
their export/import operations. Kyrgyzstan was the first post-Soviet country to become a
member of the WTO in 1998, followed by the Baltic states (1999‒2001), Georgia (2000),
Moldova (2001), Armenia (2003), Ukraine (2008), Russia (2012), Tajikistan (2013), and
Kazakhstan (2015). But their development looked anything but export oriented. The share
of export in GDP in Central Asian countries, for instance, fell dramatically after the disso
lution of the Soviet Union, and has not increased in the 1990s‒2010s. A yardstick for
comparison can be Turkey, a country at a similar level of development: it managed to in
crease the share of exports in GDP from 6 to 9.5 per cent in 1992‒2016, whereas in Cen
tral Asian countries this indicator either increased only marginally (Kyrgyzstan, Turk
menistan) or decreased (Kazakhstan, Tajikistan, Uzbekistan) (Gharleghi and Popov,
2018a). It is noteworthy that four former Soviet republics that are still not members of
the WTO as of 2020 (Azerbaijan, Belarus, Turkmenistan, Uzbekistan) had higher growth
rates over the last thirty years than the others (Figure 27.2).
27.10 Conclusion
In the 1990s, Eastern European and former Soviet Union countries carried out market re
forms and experienced regressive developments in their industrial structure—de-industri
alization, ‘resource-ialization’ and ‘primitivization’ of the structure of their exports. Exten
sive liberalization in the 1990s in the former Soviet Union countries led to premature de-
industrialization in the region, which inhibited economic growth.
In contrast, China and Vietnam also carried out market-type reforms (China, gradual re
forms since 1979; Vietnam, gradual reforms in 1986‒89 and radical reforms in 1989) and
achieved an acceleration of their economic growth and an impressive increase in manu
facturing output and exports.
The varying performance of post-communist countries can be attributed to a large extent

to different models of industrial policy: whereas China and Vietnam relied on active pro
motion of exports (mostly via an undervalued exchange rate), Eastern European countries
did not introduce any sizeable stimulus to particular industries, and former Soviet re
publics for a long time assisted domestic producers in a very inefficient way—by main
taining low prices for energy and fuel in the domestic market.
This is one of the most important lessons of the post-communist transition: market-orient
ed reforms alone are not enough, the state needs to maintain strong institutional capacity
and to stimulate progressive structural shifts. A successful industrial policy is (p. 806)
needed for structural economic transformation to achieve inclusive and sustainable devel
opment. Industrial policy could use protectionist instruments, but should stimulate ex
Page 26 of 34
rope and Asia
ports. Protectionism alone is not enough for upgrading the industrial structure and
speeding up economic and social development.
There are many ways to promote manufacturing exports, but the most promising tool is
the undervaluation of the exchange rate and public investment, especially in infrastruc
ture and education. These policies were widely used by virtually all ‘economic miracle’
countries and contributed to rapid growth and successful catch-up development in East
Asia and elsewhere. Undervaluation of the exchange rate is a non-selective industrial pol
icy that creates stimuli for the tradable goods sector and exports of tradables, whereas
public investments contribute to the acceleration of economic growth, not only because
they lead to the rise of the share of investment in GDP (no crowding-out effect), but also
because they provide public goods (education, infrastructure) with strong externalities
and can eliminate bottlenecks (if private investors ignore particular areas).
For resource-rich countries, the need for a special policy to lower the real exchange rate
is especially pressing (to avoid ‘Dutch disease’). Today in all resource-rich post-commu
nist countries, domestic prices for fuel are kept below world market level through taxes
on exports of fuel and direct restrictions on exports (export quotas, access to the
pipeline). This provides subsidies to all producers using fuel and energy, stimulating eco
nomic growth, but at the same time leads to high energy intensity. A more efficient way to
stimulate export-oriented growth is to gradually eliminate export taxes for fuel and ener
gy, to tax the extra profits of fuel companies and use the revenues for infrastructure in
vestment, and to stimulate producers of tradables, not via price subsidies for fuel, but via
an underpriced exchange rate.
Since the mid-1990s and especially in the 2000s and 2010s, many FSU countries have
made substantial economic advances. Six countries in the region (Turkmenistan, Uzbek
istan, Azerbaijan, Kazakhstan, Tajikistan, and Belarus) increased their output no less than
Central Europe (1.7 times and more as compared to 1989; Figure 27.1). Uzbekistan’s and
Tajikistan’s achievements are especially impressive because they are only partly based on
resource exports. Such positive dynamics are due, to a large extent, to the efficient indus
trial policy that resisted de-industrialization and supported manufacturing exports
through undervalued exchange rates and tax policies.
References
Almaty Programme of Action for Landlocked Developing Countries (2013) ‘ITT for LLDCs.
Priority 5: Structural economic transformation (2019)’. Available at http://land-
locked.org/priority-5-structural-economic-transformation/.
Bhalla, Surjit S. (2012) Devaluing to Prosperity: Misaligned Currencies and their Growth
Consequences. Washington, DC: Peterson Institute for International Economics.
Chenery, Hollis B. (1960) ‘Patterns of Industrial Growth’, The American Economic Review
50(4): 624–54. (p. 807)
Page 27 of 34
rope and Asia
Named: Principles of Industrial Policy’. IMF Working Paper No. 19/74. Washington, DC:
IMF.
Dollar, David (1992) ‘Outward-oriented Developing Economies Really Do Grow More

Rapidly: Evidence from 95 LDCs, 1976–1985’, Economic Development and Cultural
Change 40(3): 523–44.
Easterly, William (2001) The Lost Decades: Explaining Developing Countries Stagnation
1980–1998. Washington, DC: The World Bank
Eden, Maya and Aart Kraay (2014) ‘“Crowding in” and the Returns to Government Invest
ment in Low-income Countries’. Policy Research Working Paper No. 6781. Washington,
DC: World Bank.
Gerschenkron, Alexander (1962) Economic Backwardness in Historical Perspective: A

Book of Essays. Cambridge, MA: Belknap Press of Harvard University Press.
Gharleghi, Behrooz and Vladimir Popov (2018a) Changes in the Geographical Structure of
Trade in Central Asia: Real Flows in 1989–2016 versus Gravity Model Predictions. Berlin:
Dialogue of Civilizations Research Institute.
Gharleghi, Behrooz and Vladimir Popov (2018b) Farewell to Agriculture? Productivity

Trends in and Competitiveness of Agriculture in Central Asia. Berlin: Dialogue of Civiliza
tions Research Institute.
Gibbs, Murray (2007) ‘Trade Policy’. UN DESA. Available at https://esa.un.org/tech

coop/documents/PN_TradePolicyNote.pdf.
Gill, Indermit S., et al. (2014) Diversified Development: Making the Most of Natural Re
sources of Eurasia. Washington, DC: World Bank.
Greenwald, Bruce and Joseph E. Stiglitz (1986) ‘Externalities in Economies with Imper
fect Information and Incomplete Markets’, The Quarterly Journal of Economics 101(2):
229–64.
Greenwald, Bruce and Joseph E. Stiglitz (2013) ‘Learning and Industrial Policy: Implica
tions for Africa’, in Joseph E. Stiglitz, Justin Yifu Lin, and Ebrahim Patel (eds) The Indus
trial Policy Revolution II: Africa in the XXI Century. International Economic Association
Series. London: Palgrave MacMillan, pp. 25–49.
Griffith-Jones, Stephany and José Antonio Ocampo (2010) ‘Sovereign Wealth Funds: A De
veloping Country Perspective’. Paper prepared for the workshop on Sovereign Wealth
Funds organized by the Andean Development Corporation. London, 18 February 2008.
Available at http://www.stephanygj.net/papers/
SovereignWealthFundsDevelopingCountryPerspective2010.pdf.
Page 28 of 34
rope and Asia
Harris, Richard I. and Qian Cher Li (2007) ‘Learning-by-Exporting? Firm-level Evidence
for UK Manufacturing and Services Sectors’. Glasgow Business School—Economics, Uni
versity of Glasgow. Available at http://www.gla.ac.uk/media/media_36279_en.pdf.
Hausmann Ricardo and Dani Rodrik (2006) ‘Doomed to Choose: Industrial Policy as
Predicament’. First Draft. Paper prepared for the first Blue Sky seminar organized by the
Center for International Development at Harvard University, 9 September 2006. Cam
Hausmann, Ricardo, Jason Hwang, and Dani Rodrik (2005) ‘What You Export Matters’.
NBER Working Paper No. 11905. Cambridge, MA: National Bureau of Economic Re
search. Available at http://www.nber.org/papers/w11905.pdf.
Herrendorf, Berthold, Richard Rogerson, and Ákos Valentinyi (2013) ‘Growth and Struc
tural Transformation’. Paper prepared for the Handbook of Economic Growth. Washing
ton, DC: IMF. Available at https://www.imf.org/external/np/seminars/eng/2013/SPR/
pdf/rrog2.pdf.
IMF (1991) Soviet Economy: A Study of the Soviet Economy. 3 vols. Washington, DC: IMF.
IMF, WB, and WTO (2017) Making Trade an Engine of Growth for All: The Case
(p. 808)
for Trade and for Policies to Facilitate Adjustment. Washington, DC, Geneva: International
Monetary Fund, World Bank, and World Trade Organization. Available at https://
www.wto.org/english/news_e/news17_e/wto_imf_report_07042017.pdf.
Innis, Harold Adams (1956) Essays in Canadian Economic History, edited by Mary Q. In
nis. Toronto: University of Toronto Press.
Jomo, Kwame Sundaram (2013) ‘The Best Approach to Economic Development is Pragma
tism’, in Piotr Dutkiewicz and Richard Sakwa (eds) 22 Ideas to Fix the World: Conversa
tions with the World’s Foremost Thinkers. New York: New York University Press, pp. 355–
76.
Jomo, Kwame Sundaram (2019) ‘Industrial Policy Finally Legitimate?’, Inter Press Ser
vice, July. Available at http://www.ipsnews.net/2019/07/industrial-policy-finally-le
gitimate/.
Khan, Mushtaq H. (2007) ‘Governance, Economic Growth and Development since the
1960s’. DESA Working Paper No. 54, August. Available at http://www.un.org/esa/desa/
papers/2007/wp54_2007.pdf.
Lee, Keun (2013) Schumpeterian Analysis of Economic Catch-up Knowledge, Path-Cre

ation, and the Middle-income Trap. Cambridge: Cambridge University Press.
Lin, Justin Yifu (2011) ‘From Flying Geese to Leading Dragons: New Opportunities and
Strategies for Structural Transformation in Developing Countries’. Policy Research Work
Page 29 of 34
rope and Asia
ing Paper No. 5702. Washington, DC: World Bank. Available at https://
elibrary.worldbank.org/doi/abs/10.1596/1813-9450-5702.
tries Conform to Comparative Advantage or Defy it? A Debate between Justin Lin and Ha-
Joon Chang’, Development Policy Review 27(5): 483–502.
Murphy, Kevin M., Andrei Shleifer, and Robert W. Vishny (1989) ‘Industrialization and the
Big Push’, The Journal of Political Economy 97(5): 1003–26.
Myrdal, Gunnar (1968) Asian Drama: An Inquiry into the Poverty of Nations. 3 vols. New
York: Pantheon Books.
OECD (2017) Making Trade Work For All. Paris: OECD. Available at https://
www.oecd.org/trade/making-trade-work-for-all.pdf.
O’Rourke, Kevin H. and Jeffrey G. Williamson (2002) ‘From Malthus to Ohlin: Trade,
Growth and Distribution since 1500’. NBER Working Paper No. 8955. Available at http://
www.nber.org/papers/w8955.pdf.
Page, John (2011) ‘Should Africa Industrialize?’. WIDER Working Paper No. 2011/47, Au
gust. Helsinki: UNU-WIDER.
Polterovich, Victor and Vladimir Popov (2004) ‘Accumulation of Foreign Exchange Re
serves and Long-term Economic Growth’, in Shinichiro Tabata and Akihiro Iwashita (eds)
Slavic Eurasia’s Integration into the World Economy. Sapporo: Slavic Research Center,
Hokkaido University. Available at http://www.nes.ru/%7Evpopov/documents/
EXCHANGE%20RATE-GrowthDEC2002withcharts.pdf.
Polterovich, Victor and Vladimir Popov (2005) ‘Appropriate Economic Policies at Different
Stages of Development’. NES. Available at https://papers.ssrn.com/sol3/papers.cfm?
abstract_id=1753530.
Polterovich, Victor and Vladimir Popov (2006) ‘Stages of Development, Economic Policies
and New World Economic Order’. Paper presented at the Seventh Annual Global Develop
ment Conference in St. Petersburg, Russia, January. Available at http://www.http-
server.carleton.ca/~vpopov/documents/NewWorldEconomicOrder.pdf. (p. 809)
Polterovich, Victor, Vladimir Popov, and Alexander Tonis (2007) ‘Экономическая

политика, качество институтов и механизмы”ресурсного проклятия”’. Moscow: The
Higher School of Economics Publishing House. Available at http://pages.nes.ru/vpopov/
documents/Resources-2007-HSE-Polterovich-Popov-Tonis.pdf.
Page 30 of 34
rope and Asia
Polterovich, Victor, Vladimir Popov, and Alexander Tonis (2008) ‘Mechanisms of Resource
Curse, Economic Policy and Growth’. NES Working Paper No. WP/2008/082. Moscow:
New Economic School. Available at https://pages.nes.ru/vpopov/documents/
Resource_Curse_and_Economic_Policy-English-NESFin2008-Febr.pdf.
Polterovich, Victor, Vladimir Popov, and Alexander Tonis (2010) ‘Resource Abundance: A
Curse or Blessing?’. DESA Working Paper No. 93. New York: United Nations Department
of Economic and Social Affairs. Available at http://www.un.org/esa/desa/papers/2010/
wp93_2010.pdf.
Popov, Vladimir (2000) ‘Shock Therapy versus Gradualism: The End of the Debate (Ex
plaining the Magnitude of the Transformational Recession)’, Comparative Economic Stud
ies 42(1): 1–57.
Popov, Vladimir (2007) ‘Shock Therapy versus Gradualism Reconsidered: Lessons from
Transition Economies after 15 Years of Reforms’, Comparative Economic Studies 49(1): 1–
31.
Popov, Vladimir (2010a) ‘Development Theories and Development Experience’. CEFIR and
NES Working Paper No. 153. Moscow: Centre for Economic and Financial Research at
New Economic School.
Popov, Vladimir (2010b) ‘Global Imbalances: An Unconventional View’, IDEAs, 19 October.
Popov, Vladimir (2011) ‘Стратегии экономического развития’ (‘Strategies of Economic

Development’). Moscow: The Higher School of Economics Publishing House.
Popov, Vladimir (2013) ‘An Economic Miracle in the Post-Soviet Space: How Uzbekistan
Managed to Achieve What No Other Post-Soviet State Has’. MPRA Paper No. 48723.
Available at https://mpra.ub.uni-muenchen.de/48723/1/MPRA_paper_48723.pdf.
Popov, Vladimir (2014) ‘Can Uzbekistan Economy Retain its High Growth Rate? Scenarios
of Economic Development in 2015–30’. PONARS Eurasia Working Paper. Washington, DC:
PONARS Eurasia.
Popov, Vladimir and Anis Chowdhury (2016) ‘What can Uzbekistan Tell Us about Industri
al Policy That We Did Not Already Know?’. DESA Working Paper No. 147. New York: Unit
ed Nations Department of Economic and Social Affairs. Available at http://www.un.org/
esa/desa/papers/2016/wp147_2016.pdf.
Popov, Vladimir and Kwame Sundaram Jomo (2017) ‘Are Developing Countries Catching
Up?’, Cambridge Journal of Economics 42(1): 33–46.
Prasad, E. (2004) ‘China’s Growth and Integration into the World Economy: Prospects
and Challenges’. IMF Occasional Paper No. 232. Washington, DC: IMF.
Reddy, Sandjay G. (2013) ‘Randomise This! On Poor Economics’, Review of Agrarian Stud
ies 2(2). Available at http://www.ras.org.in/randomise_this_on_poor_economics.
Page 31 of 34
rope and Asia
Rodriguez, Francisco and Dani Rodrik (2001) ‘Trade Policy and Economic Growth: A
Skeptic’s Guide to the Cross-National Evidence’, in Ben S. Bernanke and Kenneth Rogoff
(eds) NBER Book Series NBER Macroeconomics Annual, Vol. 15. Cambridge, MA: MIT
Press, pp. 261–338.
Rodrik, Dani (2006) What’s So Special about China’s Exports? Cambridge, MA: Harvard
University Press.
on Economic Activity Fall: 365–412. (p. 810)
Rodrik, Dani (2012) ‘Doing Development Better: Project Syndicate’, 11 May. Available at
https://www.project-syndicate.org/commentary/doing-development-better?
barrier=accesspaylog.
Rodrik, Dani and Arvind Subramanian (2004) ‘From Hindu Growth to Productivity Surge:
The Mystery of the Indian Growth Transition’. IMF Working Paper No. WP/04/77. Wash
ington, DC: International Monetary Fund, May. Available at: https://www.imf.org/Exter
nal/Pubs/FT/staffp/2004/00–00/rodrik.pdf.
Rodrik, Dani et al. (2016) ‘Structural Change, Fundamentals, and Growth: Overview’, in
Margaret McMillan, Dani Rodrik, and Claudia Sepúlveda (eds) Structural Change, Funda
mentals, and Growth: A Framework and Case Studies. Washington, DC: International
Food Policy Research Institute Washington, pp. 1–38.
Rose, Andrew K. (2002) ‘Do We Really Know that the WTO Increases Trade?’. NBER
Working Paper No. 9273. Cambridge, MA: NBER. Available at http://www.nber.org/pa
pers/w9273.pdf.
Sachs, Jeffrey D. and Andrew M. Warner (1995) ‘Economic Reform and the Process of
Global Integration’. Brookings Papers on Economic Activity. Washington, DC: Brookings
Institution Press.
Sachs, Jeffrey D. and Andrew M. Warner (2001) ‘Natural Resources and Economic Devel
opment: The Curse of Natural Resources’, European Economic Review 45: 827–38.
Sala-i-Martin, Xavier and Arvind Subramanian (2003) ‘Addressing the Natural Resource
Curse: An Illustration from Nigeria’. NBER Working Paper No. 9804. Available at http://
Schmidt-Hebbel, Klaus, Luis Serven, and Andres Solimano (1996) ‘Saving and Invest
ment: Paradigms, Puzzles, Policies’, The World Bank Research Observer 11(l): 87–117.
Shevtsova, Yevgeniya (2012) ‘International Trade and Productivity: Does Destination Mat
ter?’. Discussion Papers in Economics 12/18. York: University of York.
Page 32 of 34
rope and Asia
Stiglitz, Joseph E. (2005) ‘Making Natural Resources into a Blessing rather than a Curse’,
in Svetlana Tsalik and Anya Schiffrin (eds) Covering Oil: A Reporter’s Guide to Energy
and Development. New York: Open Society Institute, pp. 13–20.
Stiglitz, Joseph E. (2018) ‘From Manufacturing Led Export Growth to a 21st Century In
clusive Growth Strategy: Explaining the Demise of a Successful Growth Model and What
to Do About It’. WIDER Development Conference, September.
Stiglitz, Joseph E. and Andrew Charlton (2005) Fair Trade for All: How Trade Can Pro
mote Development. Oxford: Oxford University Press.
Syrquin, Moshe and Hollis Chenery (1989) ‘Three Decades of Industrialization’, The
World Bank Economic Review 3(2): 145–81.
UN ESCAP (2014) ‘Economic Diversification in Asian Landlocked Developing Countries:

Prospects and Challenges’. Bangkok: United Nations Economic and Social Commission
for Asia and the Pacific, United Nations. Available at https://www.unescap.org/sites/de
fault/files/Economic%20Diversification_low%20resolution.pdf.
UN WESP (2016) World Economic Situation and Prospects 2016: Global Economic Out
look. Available at https://www.un.org/development/desa/dpad/publication/world-
economic-situation-and-prospects-2016/.
Williamson, Jeffrey G. (2002) ‘Winners and Losers over Two Centuries of Globalization’.
NBER Working Paper No. 9161. Cambridge, MA: NBER. Available at http://
World Bank (1993) The East Asian Miracle: Economic Growth and Public Policy. Oxford:
Notes:
(1) One exception could be Botswana, which had one of the highest rates of per capita
GDP growth in the last fifty years (5 per cent during 1960‒2010), primarily driven by ex
ports of primary commodities (namely, diamonds) and not of high-tech goods. The other
exception may be Oman: out of twenty economies with average growth rates of GDP per
capita in 1950‒2010 of 3 per cent and more a year, Oman was the only oil-rich state
(nearly 5 per cent growth a year) (Popov and Jomo, 2017). True, many other oil and gas
exporters in the Persian Gulf and elsewhere quickly became rich in recent decades, but
not due to higher growth rates of output (it was moderate), but due to the improvement
of the terms of trade—their income from resource exports grew much faster than their
output and exports.
(2) While Vietnamese industry, excluding constantly and rapidly growing oil production,
experienced some downturn in 1989‒90 (–6 per cent in 1989 and 0 per cent in 1990) agri
Page 33 of 34
rope and Asia
cultural growth remained strong, so that GDP growth rates remained virtually stable at
5‒6 per cent a year.
(3) Unfortunately, there are no comparable statistics on the shares of mining and utilities
separately.
(4) Personal communication with Justin Lin. In general form the theory is presented in Lin
(2011).
(5) Regression analysis (Polterovich and Popov, 2005) supports these conclusions. It im
plies that FDI positively influences growth in countries with a good investment climate
and negatively in countries with a poor investment climate:
where GR—annual average growth rates of GDP per capita,
ICI—investment climate index,
FDI—average foreign direct investment inflow as a % of GDP in 1980‒99.
Coefficients are significant at the 10 per cent level or less.
This equation establishes a very high threshold of investment climate index—about 80 per
cent, which is basically the level of developed countries. Only a few developing countries
(Botswana, Kuwait) have such a good investment climate. The worse the investment cli
mate of a country, the larger may be losses from FDI, hence stronger foreign investments
should be regulated by the state.
Vladimir Popov
Vladimir Popov is a principal researcher in the Central Economics and Mathematics

Institute of the Russian Academy of Sciences. He is also professor emeritus at the
New Economic School in Moscow, and an adjunct research professor at the Institute
of European and Russian Studies at Carleton University in Ottawa. He has published
extensively on world economy and development issues; he is editor of four books, and
author of hundreds of articles and ten books including Mixed Fortunes: An Economic
History of China, Russia, and the West (Oxford University Press, 2014; also published
in Chinese).
Page 34 of 34
Latin American Industrial Policies: A Comparative Perspective
Latin American Industrial Policies: A Comparative Per

spective
José Antonio Ocampo and Gabriel Porcile
This chapter analyses the evolution of industrial policy in the four largest Latin American
economies—Brazil, Mexico, Argentina, and Colombia—using Korea as a benchmark.
Three phases in industrial policy and industrial transformation are identified: a period of
state-led industrialization between the end of the Second World War and around 1980,
which came to an end with the debt crisis of the 1980s; the 1990s, when the region aban
doned industrial policy and embraced the structural reform agenda; and the commodity
boom after 2004, when the region witnessed a timid return of industrial policy, while at
the same time world demand favoured the reprimarization of its export structure. The
outcomes of these policies in terms of building technological capabilities and diversifying
the industrial sector towards knowledge-intensive activities, are analysed using different
indicators. The importance of the interactions between structural change, industrial poli
cy, and macroeconomic policies (especially exchange rate and capital account manage
ment) are highlighted.
Keywords: Latin America, industrial policy, de-industrialization, technological capabilities, commodity

dependence, exchange rate policy
28.1 Introduction
INDUSTRIAL policy is a policy that reshapes incentives to foster technological and struc
tural change. Since, as the Schumpeterian tradition argues, economic development is dri
ven by technological innovation and learning, and associated structural change, industri
al policy is at the core of development policies. In an international system marked by
asymmetries in technological capabilities, development requires technological catching
up and diversification towards a more sophisticated (technology-intensive) pattern of spe
cialization. This defines a desirable path of structural change for development, as op
posed to ‘regressive’ structural change—a path leading to the loss of knowledge-intensive
sectors and growing dependence on primary exports. The question to be addressed in a
comparative study of industrial policy in a developing economy is in which cases such pol
Page 1 of 33
icy—or its absence—has been effective in reducing the technology gap and changing the
pattern of specialization, and in which cases it has failed to do so, and why.
What constitutes the best industrial policy for development is far from consensual among
economists—perhaps even less consensual than in other economic policy areas. There is a
strong current in neoclassical economics that argues the best industrial policy is to have
none at all, or at best that only a ‘horizontal’ policy is admissible, that is, one that pro
vides across-the-board incentives to help all industries without targeting anyone in partic
ular—such as a general subsidy for research and development (R&D). This approach con
siders that ‘vertical’ industrial policies—in which the selection of sectors to be favoured is
explicit—distort incentives, produce allocative inefficiencies, and encourage rent-seeking
activities that compromise productivity growth and capital (p. 812) accumulation in the
long run. Vertical industrial policy is described by this view as ‘picking winners’. Howev
er, there is an implicit selection of sectors (non-neutrality) in horizontal industrial policies
as well: for instance, an across-the-board subsidy to R&D will favour mostly industries
that are intensive in R&D.
The strong version of the neoclassical view was dominant in the formulation of industrial
policy after the market reforms that took place in the Southern Cone countries (notably
Chile) in the late 1970s, and on a broader basis in Latin America since the mid-1980s. In
versely, the quest for structural change was paramount in the policies adopted by several
Latin American countries during the period of ‘state-led industrialization’1 and by the suc
cessful cases of catching up, mostly in East Asia. This chapter compares industrial policy
in different countries and how they shaped different trajectories of productivity growth
and structural change. Although we present the broader Latin American trends, we focus
our analysis on the four largest Latin American economies—Argentina, Brazil, Colombia,
and Mexico—comparing these countries with the Republic of Korea (henceforth just Ko
rea). We argue that the persistence of industrial policy in Korea, combined with a macro
economic policy that avoided the appreciation and instability of the real exchange rate,
were critical for its success, while the abandonment of industrial policy in the 1980s and
the use of the real exchange rate mainly as an anti-inflationary tool and a policy instru
ment to adjust to commodity boom–bust cycles, explain why the Latin American countries
failed to catch up.
The chapter is organized into five sections. Following this introduction, we describe our
theoretical framework for the analysis of technical and structural change, combining mi
cro and macroeconomic variables in what we broadly call the structuralist tradition.
There follows a discussion of industrial and macroeconomic policies in Latin American
countries, taking Korea as a benchmark. Next we use different indicators of technological
catching up and economic transformation to show that Latin America has been lagging
behind East Asia and the world economy since the 1980s. A final section presents some
brief conclusions.
Page 2 of 33
28.2 Tacitness, Localized Technical Change,

and Structural Inertia
28.2.1 The Dynamics of Learning and Industrial Policy
Although there are many different neoclassical models of technical change (see Aghion
and Howitt, 1997), the orthodox views that are prevalent in the policy debate assume
that all sectors are equally capable of promoting technical change and that there
(p. 813)
are no significant barriers to the diffusion of technology. This perspective, however, is at

odds with the growing evidence that the velocity of innovation and technological opportu
nities varies significantly across sectors (Bogliacino and Pianta, 2016). It is also at odds
with the large international technological gaps that characterize the global economy, and
with the fact that countries that have succeeded in catching up have applied comprehen
sive industrial policies ‘distorting’ market incentives.
In contrast with the orthodox view, the structuralist tradition argues that the internation
al diffusion of technology is slow and irregular. The empirical evidence coming from firm-
level studies of technological innovation and diffusion shows much less flexibility in the
pace and direction of learning than what is implicit in neoclassical growth models (Dosi et
al., 2015). One major factor is the importance of production experience as a source of
productivity growth, a significant factor underscored by Arrow’s concept of learning by
doing and by the Kaldor–Verdoorn Law (Arrow, 2004; Kaldor, 1978: chs. 1, 3, and 4). The
key insight in both cases is that technical change cannot be separated from experience in
production and the existing set of capabilities firms and countries deploy at a certain
point in time. The evolutionary theory of technical change has offered new insights, in
particular the concept that technological capabilities are tacit, in the sense that they can
not be transmitted in a codified way (i.e. through manuals or printed instructions), and
must be incorporated into the routines of the firm and the skills of shop-floor workers and
engineers. A similar idea was suggested in a pioneering article by Atkinson and Stiglitz
(1969), who argued that technical change is localized around techniques that firms are
using.2 The production function takes the form of disconnected sets (clusters of technolo
gies) instead of a smooth, continuous isoquant along which the firm chooses the optimal
combination of factors of production.
In other words: what firms learn, the possibility of catching up, and the direction of inno
vation depend on current capabilities and experience. This also means that technological
capabilities are embedded in the production structure. The interconnection between
learning and production implies that production structures and technology co-evolve.
There is a strong element of inertia in specialization and in technological capabilities that
may lead to slow growth and learning traps. In the absence of an exogenous shock that
redefines the prevailing incentives, asymmetries in technology will endogenously repro
duce the ‘old’ structure. The role of industrial policy is to provide such an exogenous
shock to reshape incentives and overcome learning traps.
Page 3 of 33
Changes in relative prices (albeit necessary) may not suffice to bring about technical
change when learning is localized. Upgrading technological capabilities (p. 814) requires
more than adequate price signals. It demands an institutional framework for the coordi
nation of the various actors (in the private and public sectors, such as R&D agencies, uni
versities, and centres for technological training and extension) engaged in learning and
innovation, redirecting investments towards new capabilities over a long-term horizon,
and away from the most obvious short-term incentives (Katz and Stumpo, 2001; Cimoli
and Katz, 2003; Anderson and Ejermo, 2006; Bell, 2006).
A systemic approach is at the core of the evolutionary literature, especially in the concept
of national systems of innovation (Fagerberg and Verspagen, 2002; Lundvall, 2016) and
their institutional variations across countries. These variations may take the form of ‘mis
sion-oriented’ and ‘diffusion-oriented’ paradigms in industrial and technological policies
(Chiang, 1991). Mission-oriented innovation policies are based on large projects with pub
lic and private funding that seek to produce a technological breakthrough and shift the
international technological frontier, leading to global leadership in strategic sectors. Dif
fusion-oriented strategies are those focused on the absorption and improvement of exist
ing technology, especially in adapting and improving foreign technology through cumula
tive minor innovations. Countries that adopt diffusion-oriented strategies (mostly, but not
exclusively, developing economies) seek to create competitive advantages acting as smart
followers rather than as innovators, moving up (‘trickle-up’) in the technological ladder.
An important issue is to what extent the firms’ nationality matters in the effort to build a
national system of innovation. Amsden (2001) has argued that counting on national firms
was an important ingredient in Korea’s success story, which is at variance with Latin
America. This could have been a factor in the ability of the governments in Asia and Latin
America to effectively implement industrial policies. While we acknowledge the impor
tance of this point, it will not be discussed in this chapter.
28.2.2 The Links between Macroeconomic and Industrial Policies
Fiscal, monetary, and exchange rate policies are essential to stabilize inflation and
smooth business cycles. However, macroeconomic policy also has substantial effects on
structural change. As GDP growth volatility increases investment risks, avoiding sharp
cyclical fluctuations may be particularly important to promote growth and associated
structural change. This especially compromises innovative sectors, which are in them
selves subject to larger risks. Macroeconomic policies, and especially the real exchange
rate policy, may also change the composition of output. Changes in the real exchange rate
(RER)3 redefine the structure of relative prices in favour of, or against, the production of
tradable goods. To the extent that tradable and non-tradable goods (p. 815) have different
potential for learning and productivity growth, the exchange rate policy generates long-
run effects.
There is substantial evidence suggesting that a stable and slightly undervalued RER
favours structural change and economic growth in the long run (Rodrik, 2008; Razmi et
Page 4 of 33
al., 2012; Guzman et al., 2018). Two mechanisms explain this association. The first is that
growth in developing economies tends to be constrained by the external sector, as the lit
erature on balance-of-payments-constrained growth suggests. There is also a strong de
pendence of short-term economic activity on external shocks, positive and negative—an
effect that one of us has called ‘balance-of-payments dominance’ (Ocampo, 2016). A coun
try that overvalues its currency will grow at a lower rate than a country that keeps it in
equilibrium or slightly undervalued. This would be associated with short-term growth
risks (the probability of a balance-of-payments crisis), but also long-term effects, particu
larly reduced opportunities for investment and learning by doing in tradable sectors.
The second mechanism is that tradable sectors are at the heart of industrial policy be
cause high-tech sectors are usually tradable and are exposed to more intense competition
in international markets. Therefore, to be effective, industrial policy should work in line
with macroeconomic policy, especially with the exchange rate policy. If the RER is over
valued, attempts to promote tradable sectors—for instance, by improving the coordina
tion of R&D agencies and firms, improving technical training, or granting subsidies to in
novation—are doomed to fail. In other words, industrial policy cannot surmount the disin
centive for the production of tradables associated with an appreciated RER. On the other
hand, relying too much on an undervalued currency for competitiveness entails the risk of
dependence on cheap labour. Arrow’s learning by doing and the Kaldor–Verdoorn Law are
not automatic forces: the speed of learning out of growth depends on the industrial and
technological policies. In short, the macroeconomic policy cannot replace industrial poli
cy; and an industrial policy without a competitive RER will be ineffective in providing in
centives for structural change.
28.3 Policies in a Comparative Perspective: In

dustrial Policy in the Different Phases of Latin
American Development
28.3.1 Industrial Policy: From State-led Industrialization to the Wash
ington Consensus
Why did Latin America fall behind, while several Asian countries succeeded in converging
with the advanced economies? We will argue that the specific combination (p. 816) of in
dustrial and macroeconomic policies these countries pursued explain the different out
comes. First, we briefly summarize some general trends for the region and subsequently
focus on Argentina, Brazil, Colombia, and Mexico in Latin America, and Korea in Asia—
the latter as a successful benchmark.
At least three phases in the evolution of industrial policy can be identified in the post-Se
cond World War period in Latin America (simply post-war hereafter). First, from the early
post-war period4 to the late 1970s/early 1980s, state-led industrialization was the domi
nant strategy, especially in the largest economies of the region (Bértola and Ocampo,
Page 5 of 33
2012). The Latin American governments actively sought to encourage industrialization

and diversification, initially through import substitution, which was increasingly comple
mented by export diversification and regional integration. There were significant im
provements in several fields, including the building up of technological capabilities in
new sectors—frequently with a key role for publicly funded universities or R&D depart
ments in large public-sector firms. State-led industrialization, however, was challenged by
the debt crisis of the 1980s, which subsequently turned into a major fiscal crisis. There
was no room for industrial policy in the 1980s, as investment collapsed, inflation rocket
ed, major macroeconomic adjustments were in place (usually under severe restrictions
imposed by the creditor countries) and resources were transferred to foreign creditors.
Second, in the 1990s, foreign capital inflows returned to Latin America, and the external
constraint was eased. With a less restrictive external environment, there should have
been more space to move beyond the stabilization agenda. However, the predominance of
orthodox views in policymaking, and the trauma left by the high inflation levels of the
1980s, led to a retreat of the state, keeping industrial policies at bay. At best, across-the-
board policies to enhance competitiveness were admitted (sometimes focusing on small
and medium-sized enterprises; see Peres and Primi, 2019). Real exchange rate apprecia
tion, facilitated by high liquidity in global financial markets, reinforced negative signals
towards manufacturing and export diversification.
Third, after a series of external crises in the late 1990s and early 2000s, there was a re
vival of industrial policies. Disappointment with the results of the pro-market reforms of
the 1990s had led to reform fatigue by the late 1990s, as new generations of orthodox re
forms were called for but were not adopted or had no significant success. However, the
revival of industrial policies was timid and faced powerful countervailing forces. The
2004–14 commodity boom (temporarily interrupted by the 2008–09 North Atlantic finan
cial crisis) boosted exports intensive in natural resources—a reprimarization of the export
structure—particularly in South America, and generated a new wave of exchange rate ap
preciation. Some of the industrial policy instruments were mostly defensive and ill articu
lated with technological policy.
The emergence of China as a prominent global trader not only contributed to the repri
marization of Latin America’s export structure through its demand pattern (p. 817) (con
centrated in natural resources) but also by competing with the Latin American manufac
tures in Europe, the United States, and in the region itself. It became more difficult for
Latin American countries to remain competitive in industries with low or medium techno
logical intensity. As stated by Jenkins and Freitas Barbosa (2012): ‘it is clear that the con
cerns of manufacturers over Chinese competition are found throughout the region from
Mexico to Argentina. They are also prevalent in a wide range of industries from tradition
al labour-intensive activities such as clothing, footwear, and furniture to more technologi
cally advanced and capital-intensive sectors such as steel and electrical and electronic
products.’5
Page 6 of 33
During the three periods, a common feature was the weakness of the state when it came
to penalizing firms that received public support without delivering in terms of technical
change and competitiveness. Even during the years of state-led industrialization, the
state’s failure to remove rents when they did not encourage innovation contrasts with the
assertive policies adopted in Asia, where incentives were granted on the basis of perfor
mance, generating ‘reciprocal control mechanisms’, to borrow Amsden’s (2001) term.
28.3.2 From State-led Industrialization to the Debt Crisis
The period of state-led industrialization was a period of convergence (or less intense di
vergence) in relative productivity and improvements in the indices of structural change in
Argentina, Brazil, and Mexico, and to a lesser extent Colombia. The RER remained over
valued for long periods, due to the prevalence of fixed exchange rates in an inflationary
context (Mexico being an exception in this regard, as it showed low inflation). This prob
lem was partially corrected by the adoption of the crawling peg in Brazil and Colombia in
the mid-1960s. It was also counterbalanced, in terms of its effects on tradable sectors, by
high levels of protection and export subsidies on non-traditional exports.
Brazil and Mexico were more successful in convergence. The trend began earlier in Mexi
co, as a result of the policy known as stabilizing development, whereas in Brazil conver
gence gained momentum in the second half of the 1960s, leading to the ‘Brazilian mira
cle’. Unlike Argentina and Colombia, Brazil and Mexico continued to converge in the sec
ond half of the 1970s as a result of ambitious plans for building capital and intermediate
goods industries—the Second National Development Plan in Brazil (Aldrighi and Colis
tete, 2013) and the National Plan for Industrial Development in Mexico (Moreno-Brid and
Ros, 2009). Both plans sought to deepen (p. 818) import-substituting industrialization and
export diversification, using tariff and non-tariff barriers, and export subsidies. While
Brazil was heavily dependent on oil imports, partly moderated by a booming sugar-based
ethanol industry that was part of its industrial policy, Mexico had become a significant oil
exporter in the mid-1970s, a fact that also boosted economic growth. Brazil and Mexico
were seen as rising industrial powers belonging to the group of Newly Industrializing
Countries (NICs).
Economic growth could rely in the second half of the 1970s on high liquidity and low in
terest rates in the international financial markets. Still, trade liberalization and the open
ing up of the capital account in Argentina, and rapid growth in Mexico and Brazil (includ
ing the rising costs of imported oil in the case of Brazil), gave rise to an unsustainable in
crease in external debt (Cardoso and Fishlow, 1989; Dornbusch and de Pablo, 1989).
Colombia remained immune to that trend thanks to its decision to save part of the coffee
boom that it experienced in 1975–79. In turn, domestic financial liberalization generated
a domestic financial boom in all of these economies. It is important to note that the rise in
external debt occurred both in countries that were still pursuing state-led industrializa
tion (Brazil and Mexico) and in countries that had already abandoned this development
strategy (Argentina, as well as Chile, not analysed here). The debt crisis came, therefore,
Page 7 of 33
from the excessive reliance on foreign capital rather than on the specific type of industri
al policies that were in place.
The debt crisis that sparked in 1982 reversed the industrialization drive in Brazil and
Mexico and accelerated de-industrialization in Argentina. It was accompanied by major
macroeconomic adjustments in Argentina, Brazil, and Mexico in the early 1980s to man
age their massive fiscal and balance-of-payments crises, the latter leading to large ex
change rate adjustments that heightened inflationary pressures. Short-run concerns over
stabilization (how to pay the debt and tame inflation) prevailed over long-run objectives of
economic growth. The economic crisis, together with the political crises in Argentina and
Brazil, made the Latin American governments particularly vulnerable to pressure from
banks and international organizations, which acted in a coordinated way. They demanded
austerity measures and a reduction in public investment to pay for the debt. The region
implemented a drastic overhaul of trade and industrial policies along the lines of the
structural-reform agenda being pushed by the World Bank and embedded in a sequence
of stand-by agreements signed with the IMF. Although Colombia avoided an external debt
crisis, it also experienced fiscal and balance-of-payments problems and slower growth,
but not massive exchange rate adjustments and associated inflationary pressures; hence,
it was able to renew growth in the second half of the 1980s, earlier than the other coun
tries.
Debt and austerity imposed a crucial burden on the development prospects of the region,
one that would persist for many years. Until the late 1980s/early 1990s, the Latin Ameri
can countries grew at very low rates while investment rates collapsed, generating a ‘lost
decade’—the term that the Economic Commission for Latin America and the Caribbean
(ECLAC) coined to refer to the 1980s, and that has been widely used for other parts of
the world since then. The strong impact of the debt crisis on investment seems to have
produced a hysteresis phenomenon: the investment rate never (p. 819) returned to the lev
els it had attained in the three decades before the crisis. Again, due to its lower debt ra
tios, Colombia was a partial exception.
28.3.3 The Washington Consensus Years
The second phase began in the late 1980s/early 1990s and continued until the early
2000s, a period that marked the heyday of the Washington Consensus.6 It was character
ized by reforms aimed at liberalizing foreign trade, privatizing state-owned enterprises,
and deregulating markets. Many countries abandoned industrial policy (‘the best industri
al policy is no industrial policy’) or at most implemented horizontal policies to encourage
competitiveness and innovation across the board. These policies were, in theory, aimed at
making industrial development and technological change more efficient—driven by the
needs of firms rather than by the decisions of government agencies. In practice, they dis
mantled most of the old policies and put in place a poorly articulated set of incentives to
encourage productivity and innovation. Argentina created the National Agency for Scien
tific and Technological Promotion. Brazil launched the Brazilian Programme for Quality
and Productivity, and the Programmes for Technological Development in the industrial
Page 8 of 33
and agricultural sectors. Brazil’s National Development Bank (BNDES), which had been
very active in the 1970s supporting the creation of new industries, played a new role fi
nancing the privatization of public-sector firms.
In Mexico, the Economic Solidarity Plan of 1987 began to dismantle trade barriers, a
process that accelerated in the following years. By the 1990s non-tariff barriers had been
almost completely removed, with a few exceptions in the oil, automobile, and agricultural
sectors (Moreno-Brid and Dutrénit, 2018). In the transition towards more liberal policies
and the abandonment of industrial policy, trade agreements became an essential element
of the new strategy, particularly the 1994 North American Free Trade Agreement (NAF
TA). This agreement implied a commitment to eliminate barriers to trade over a period of
ten years. It was expected to allow Mexico to start a phase of export-led growth based on
its privileged access to the US market.
Colombia massively liberalized foreign trade in the early 1990s (with some agricultural
goods and car assembling as partial exceptions). It also redesigned its system of develop
ment banks (creating a bank for foreign trade, BANCOLDEX, out of a central bank redis
count facility for non-traditional exports, and keeping a development bank (p. 820) for in
dustrial development, IFI)7 and an export promotion agency (PROEXPORT). The country
also launched a series of national and regional competitiveness councils to support the
private sector (Ocampo and Arias, 2018). However, the effects of trade liberalization and
the exchange rate appreciation—generated in the latter case at least partly by the foreign
exchange and capital account liberalizations that accompanied the new trade policy—pre
vailed throughout the manufacturing sector over those of the moderate competitiveness
policies. The latter also lacked continuity across government administrations.
The 1990s also witnessed a change in the prevailing outlook on economic integration in
Latin America. While integration agreements since the 1960s had been mostly partial and
had been reached in a context of high barriers to international trade, the ‘open regional
ism’ of the 1990s sought to reduce overall trade barriers while keeping a margin of pref
erence for regional trading partners. The Andean Community (previously Andean Group)
and the Central American Common Market were relaunched in the early 1990s, and the
Southern Common Market (Mercosur, according to its Spanish acronym) was created in
1991. They served as a framework for the significant growth of intra-regional trade (par
ticularly of manufactures) in the 1990s until the crisis that struck the region at the end of
the decade.
Overall, trade liberalization and the removal of barriers to competition and regulations
failed to enhance competitiveness and speed up economic growth to the levels that had
been typical during the period of state-led industrialization. This decade was also one of
sluggish investment and strong RER appreciation. Figure 28.1 shows how the opening up
of the capital account (as measured by the Chinn–Ito Index of capital account openness
(Chinn and Ito, 2008), plotted on the left vertical axis) and the investment rate (plotted on
the right vertical axis) moved in opposite directions in Latin America (see Cherkasky and
Abeles, 2019).
Page 9 of 33
Figure 28.1 Investment ratio and capital account

openness in Latin America
Source: ECLAC based on the Chinn–Ito Index and

Penn World Tables.
There were partial exceptions to the horizontal approach to industrial policy. In all four
countries analysed, the automobile industry enjoyed a special regime, which also sought
to foster the production of auto parts. In Brazil, sixteen Sectoral Funds (Fundos Setoriais)
were created in 1998, which represented a new mechanism for funding investments in
science and technology, based on contributions from many different sources, including
royalties from oil exploration. In Argentina, a regional policy provided incentives for the
installation of assembling industries in Tierra del Fuego—though with very little incentive
for developing local capabilities; it was created in the 1980s and renewed over the years.
In Colombia, as indicated, weak competitiveness policies were put in place.
Global financial conditions in this period were favourable for capital inflows towards the
region, which took place, as we have seen, amid capital account liberalization processes
(Ffrench-Davis and Ocampo, 2001; Ffrench-Davis, 2012; Freund and Pierola, 2012;
Frenkel and Rapetti, 2011; Frieden, 2015). Capital account liberalization was moderated
by taxes on capital inflows in Brazil and by reserve requirements on such inflows in
Colombia (as well as Chile, not analysed here). Global liquidity led to an appreciation of
the RER in Argentina, Brazil, and Colombia (Figure 28.2). While this was helpful to curb
inflationary pressures in Argentina and Brazil, it depressed the competitiveness of indus
trial production. In both countries, the exchange rate became, therefore, a nominal an
chor to stabilize inflation expectations. Mexico also faced appreciation pressures, which
were interrupted, however, by a balance-of-payments crisis in late 1994.
Deregulation of capital accounts, the loss of competitiveness, and current account deficits
foreshadowed the next external crises. They occurred in 1994 (Tequila crisis) in Mexico,
and in 1997–98 in the other economies, as a contagion of the crises in East Asia and Rus
sia. Brazil experienced a crisis at the end of the Real Plan in 1999, Colombia experienced
in that year its worst recession of the post-war period, and Argentina saw the collapse of
its Convertibility Plan in 2002 after several years of difficulties.8 In all cases, these
processes were accompanied by strong exchange rate depreciations (see Figure 28.2).
Page 10 of 33
Figure 28.2 Real exchange rate (2005 = 100)
Source: ECLAC; an increase means a real exchange

rate depreciation.
28.3.4 The Timid Return of Industrial Policy in the 2000s
The external crises of 1998–2002, and the sense that pro-market reforms had failed to
bring about the expected recovery in growth, led to a gradual return of industrial policy,
mixed in part with new R&D policies and the spread of intellectual property rights
regimes adopted by WTO and free-trade agreements (Peres, 2010; Primi, 2014). The timid
return of industrial policies faced, however, a new wave of RER appreciation in South
America, the strong effects of the 2008–09 North Atlantic financial crisis on Mexico, and
a significant slowdown in South America after the 2004–14 commodity boom came to an
end.
Also, in contrast to the revitalization of the regional integration agreements in the 1990s,
which had been a significant source of intra-regional trade in manufactures, they weak
ened and became embedded in political rivalries with strong ideological content. Notably,
Venezuela left the Andean Community in 2006 and joined Mercosur, from which it was in
turn expelled in 2017 due to its lack of commitment to democracy. The split in the Andean
Community, together with the political confrontations and the economic and political cri
sis of Venezuela, led to the collapse of the second major bilateral intra-regional trade
flow, that between Colombia and Venezuela. A new integration agreement, the Pacific Al
liance, has been the most dynamic regional agreement since it was launched by four Pa
cific nations (Chile, Colombia, Peru, and Mexico) in 2011.
In Brazil, during the first presidency of Lula da Silva in 2003, the Industrial, Technologi
cal and Trade Policy (PITCE by its Portuguese acronym) was approved, (p. 823) focusing
on encouraging innovation and high-tech sectors. The PITCE included horizontal instru
ments along with fiscal and financial incentives for strategic sectors, such as capital
goods, pharmaceuticals, software, nanotechnology, and biotechnology. An interesting di
mension was the development of technologies and capital goods for deep oil exploration
to allow Brazil to fully exploit discoveries that were taking place at the time, enabling the
country to become an international leader in this field. Brazil also set out a new institu
Page 11 of 33
tional framework with a key coordinating role for the Agency of Industrial Development
(Portuguese acronym ABDI).
PITCE was subsequently replaced by the Policy of Productive Development (PPD, 2008–
10), in which the development bank (BNDES) played a major role in encouraging the in
ternationalization of large Brazilian firms. There was also a focus on coordination with
the private sector and with other public-sector programmes, notably the PAG (Pro
gramme for the Acceleration of Growth), primarily focused on investments in infrastruc
ture. While the crisis of 2008 compromised the advance of the PDP, this program was in
strumental in sustaining investments and aggregate demand in the context of an interna
tional recession of industrial countries. Finally, the PDP was replaced by the Plano Brasil
Maior (PBM) in the period 2011–14, under President Dilma Rouseff. The PBM showed a
more protectionist stance and was concerned with lowering production costs, as the ap
preciation of the RER (boosted by the commodity boom) harmed growth and employment
(see ECLAC, 2012; Bresser-Pereira, 2008; Bresser-Pereira et al., 2014; Nassif et al., 2016;
Bekerman and Dalmasso, 2014). In spite of its ambitious objectives for competitiveness
and technical change, PBM mostly played an anti-cyclical role (Laplane and Laplane,
2017). It remains to be seen whether the trade liberalization policy announced by the Bol
sonaro administration that started in 2019 will end up reversing the industrial policies of
previous administrations, which is a very likely outcome.
The Brazilian industrial policy of the 2000s had an impact on industry and R&D. Signifi
cant resources were devoted to these programmes, including, as we will see, relatively
large levels of investment in R&D relative to other Latin American countries, though
much lower than those of Korea. Notwithstanding these advances, industrial policy was
fragmented, driven by short-term political pressures, targeting in many cases sectors not
related to advanced technologies. There was no progress in terms of structural change or
reversing the de-industrialization process. Indeed, the 2004–14 commodity boom rein
forced the Brazilian specialization in natural resource-intensive exports. The growing role
of China as a trade partner of Brazil reproduced the traditional North–South (centre–pe
riphery) specialization patterns with which structuralist economists were so concerned:
Latin American countries exporting commodities in exchange for manufactures. Incen
tives for investment thus remained concentrated in traditional comparative advantages.
The lack of synchrony with the macroeconomic policy was indeed an important factor
leading to the reinforcement of the specialization pattern in natural resources (see Nassif
et al., 2015, 2018). The strong appreciation of the real was a persistent challenge for in
dustrial competitiveness, except for a short interregnum in late 2008 (p. 824) and 2009.
Since October 2009, Brazil has deployed a vast array of capital account regulations to
prevent the real from continuing appreciating. A tax on the notional amount of deriva
tives was applied, which complemented other measures to close loopholes that had al
lowed investors to bypass the regulations. However, differences in interest rates between
foreign and domestic markets remained very high in the 2000s (Kaltenbrunner and
Page 12 of 33
Paincieira, 2015).9 Capital account regulations began to be dismantled from 2012, as the
real showed increasing weakness, and monetary policy became less restrictive.
In Argentina, the Convertibility Plan collapsed in 2002, when the peso was devalued, and
GDP contracted by 11 per cent (18 per cent with respect to 1998).10 The response of the
new government that took office amidst political and economic turmoil was to advance in
both macroeconomic and industrial policies. Concerning the latter, it reinvigorated the
science and technology policy, and (especially since 2010) recovered instruments that had
become redundant in the 1990s, including investments in technology by state-owned en
terprises, using government purchases as an industrial policy instrument and some ele
ments of trade management. Policymakers sought to articulate large public projects with
investments in R&D and technological development. Sectoral technology funds and sec
toral innovation funds (FTS and FITS, according to their Spanish acronyms) were created
to encourage technological learning in biotechnology, nanotechnology, and agro-indus
tries. There were also public research programmes in satellites and nuclear energy, and a
rise in technology investments by large public firms, such as the (re)nationalized REPSOL
in the oil sector (Stumpo and Rivas, 2013).
On the macroeconomic front, and for a short time, RER management encouraged indus
trial growth. In 2002, the government adopted a system of differential taxes on exports
(retenciones) that amounted to a multiple exchange rate system, penalizing agricultural
exports and favouring manufacturing. Taxes on exports of primary goods increased from
12 per cent to 34 per cent between 2003–06 and 2010–13. Inversely, these taxes re
mained at 4 per cent for manufactures processing industrial inputs during the whole peri
od 2003–13 (Lavarello and Mancini, 2017: 94). However, RER appreciation eroded com
petitiveness over time. This made the government turn to protectionism (as happened in
Brazil), particularly after the 2008 crisis, in the form of quotas, non-automatic import li
censing (since 2009), and managed trade and export requirements. The differential taxes
were dismantled by the incoming Macri administration in late 2015 (apart from for soy
beans), but were re-established as a fiscal adjustment device in September 2018.
The launch of the Strategic Industrial Plan 2020 and Argentine Innovation 2020 in 2012
represented important steps in the quest for a coherent industrial plan for Argentina.
However, industrial and technological efforts remained dispersed and (p. 825) ill-articulat
ed. Most of the resources for industrial policy went to the automobile sector and to spe
cial regimes like in electronics in Tierra del Fuego, in which assembling activities are
dominant. The main pro-manufacturing incentives came, as we have indicated, from the
system of differential export taxes. But, as in the case of Brazil (and, as we will see,
Colombia), the 2004–14 commodity boom redefined incentives in a way that penalized the
industrial sector. Neither Brazil nor Argentina were able to prevent the RER from neutral
izing the potential gains in productivity that arose from the return of industrial policy in
struments.
The return of industrial policy has been much less assertive in Mexico than in Argentina
and Brazil. President Fox recognized the importance of sector-specific policies in his Na
Page 13 of 33
tional Development Plan (2001–06) but did not put resources behind the plan to make a
difference compared to the previous administrations (Moreno-Brid, 2016). The idea of re
covering the role of industrial policy was resuscitated by the Peña Nieto administration in
the National Development Plan 2013–18 and the quest for ‘democratizing productivity’.11
According to this plan, industrial policy should be directed at raising productivity and in
creasing value added in manufacturing production. The need for a renewed industrial pol
icy was put in the following terms by economic minister Videgaray in March 2015: ‘We
should have a country in which every year productivity grows; but for that we need an in
dustrial policy…This concept was practically forbidden in Mexico during many years…But
we have to come back and dare to pursue an industrial policy…for a more competitive in
dustry and for raising the country’s productivity’ (quoted in Moreno-Brid and Dutrénit,
2018: 28).
In practice, however, there was little progress in this direction, and few resources were
effectively allocated to industrial policy programmes. The emphasis was on policies that
should boost the existing comparative advantages rather than defying them. A Pro
gramme for Innovative Development (PRODEINN, 2013–18) was adopted to enhance pro
ductivity. But it had only a slight impact, as reflected in the stagnation of R&D expendi
tures at low levels. In the same vein, the government approved new legislation in 2016 to
create special economic zones, which would receive fiscal and financial incentives, and
investments in infrastructure aimed at reducing regional inequalities and attracting high
er-productivity activities to depressed territories, especially in the south of Mexico. As
happened with other policy initiatives, implementation was sluggish and few resources
were allocated. They failed to alter the scenario of slow productivity growth and slow
structural change of the Mexican economy.
All in all, industrial policy continues to be relegated to a secondary position in Mexico, a

problem aggravated by low rates of public investment, which dwarfs the capacity of the
government to effectively implement its development plans. By expecting that NAFTA and
trade liberalization would provide enough incentives for export-led growth, Mexico re
mained locked in its static comparative advantages (notably the maquila system in which
low wages are still a key driver of (p. 826) competitiveness), missing the key link between
trade and industrial policy as a tool for structural change in the Asian economies. A com
parison with other major manufacturing exporters, China and the Republic of Korea, indi
cates that Mexico has been able to extract less value added from its exports and is show
ing a negative rather than a positive trend in this regard.12 The absence of a national de
velopment bank as powerful as BNDES has also been a significant difference between the
experiences of industrial policy in Mexico and Argentina compared to Brazil (Santarcán
gelo et al., 2018).
For its part, and following the pattern set in the 1990s, Colombia continued with the
trade liberalization processes, now enhanced by free-trade agreements with developed
countries, starting with the United States in 2006. Proexport continued to play the role of
export-promoting agency and was transformed into Procolombia in 2014, with the addi
tional tasks of promoting tourism and foreign direct investment. The foreign trade devel
Page 14 of 33
opment bank (Bancóldex) absorbed the industrial bank (IFI) in 2003. Bancóldex also em
phasized a programme of entrepreneurial modernization, has been active in promoting fi
nancing of firms in their early stages of development, notably through the angel investor
network (Red Nacional de Ángeles Inversionistas), and has promoted private-equity ven
ture funds to support business growth in a diverse set of sectors (Ocampo and Arias,
2018).
These policies were accompanied by a sequence of national and regional competitiveness

strategies, which operated mainly as horizontal industrial policies. It also included sec
toral initiatives, largely aimed at breaking bottlenecks to development, supporting region
al clusters and strengthening regional agents promoting innovation and entrepreneurial
development. The sectoral growth covenants (Pactos por el crecimiento y la generación
de empleo), launched in 2019, belong to this family. There are also specific programmes
that provide technological, administrative, and financial support to firms—particularly in
novative firms—at the microeconomic level through Colombia Productiva (formerly Pro
grama de Transformación Productiva) and iNNpulsa (managed by Bancóldex), adopted in
2008 and 2012. These initiatives have been formulated in the context of the Production
Development Policy, which had its roots in 2008 and was improved in 2016 when it was
approved by the National Council of Economic and Social Policy (CONPES), the top na
tional economic authority (CONPES, 2016). Additional support at the firm level and for
regional clusters is implemented by chambers of commerce. A private-sector council
(Consejo Privado de Competitividad), created in 2006, provides regular analysis of com
petitiveness and productivity issues.
However, the effect of these policies has been restricted due to scale and discontinuity
across administrations. There has been a remarkable lack of coordination between na
tional and regional programmes and limited resources have been allocated to them. In
deed, some of the local programmes run by the most dynamic chambers of commerce,
which also support individual firms and regional clusters, have had greater (p. 827) conti
nuity and impact.13 Also, investment in R&D remained very low, at the lowest level of the
four countries analysed here (see Figure 28.7).
Furthermore, as in the case of Argentina and Brazil, macroeconomic policy overwhelmed

the effects of industrial policy—particularly, again, the exchange rate effects of the 2004–
14 commodity boom. Manufacturing did experience a revival in the mid-2000s, when the
balance-of-payments crisis that hit the country in 1999 was over: the exchange rate was
still competitive (see Figure 28.2), and a temporary boom in manufacturing exports to
Venezuela also helped. But the exchange rate effects of the commodity boom soon led to
the renewal of the de-industrialization trend. In contrast to 1997 and 2006–07, when re
serve requirements were imposed on foreign borrowing to limit short-term capital flows
and the exchange rate appreciation they generated, similar policies were not put in place
during the second phase of the commodity boom, 2010–14.
The paths identified for the four Latin American countries confirm the observation of
Peres and Primi (2019: 214), who argue that: ‘Latin America’s development trajectory is
Page 15 of 33
linked to the fact that these policies have never been among the top priorities for govern
ments—at least since the 1980s. Macroeconomic stability and a certain compliance with
what was considered to be a respectable and conventional economic policy have been the
main shapers of national development strategies.’ The efforts of the 2000s to revive in
dustrial policy in the region were too weak to reverse the loss of the institutional and po
litical capabilities it suffered in the 1990s. What, on paper, was a quest for industrial effi
ciency in the shape of a move from a vertical to a horizontal industrial policy, was in prac
tice a feeble and fragmented policy.
28.4 Korea as a Benchmark

In contrast to the Latin American experience in recent decades, the role of industrial poli
cy in targeting more sophisticated industries through time has been the hallmark of the
Korean development process in the post-war period.14 Korea is a paradigmatic example of
the development pattern that characterized several economies in East Asia (traditionally
called the ‘flying geese’, though China has already changed that pattern),15 which com
bined widespread government intervention with a strong presence in international mar
kets (Khan, 2000).
(p. 828) A watershed in the evolution of industrial policy in Korea occurred with the mili
tary government in the 1960s, which focused on establishing an industrial base that
would allow the country to counterbalance the military power of North Korea and reduce
its dependence on foreign finance and aid. The Economic Planning Board, created in July
1961, was in charge of designing and implementing five-year development plans. Firms’
export performance was a key criterion for resource allocation based on the leverage pro
vided by the government’s control over the financial sector (Lim, 2004: 144). In parallel,
Korea invested heavily in education, expanding the supply of highly qualified labour to an
increasingly complex industrial sector.
Comprehensive policy support, structured in successive five-year plans, helped transform

the production structure from consumer goods to heavy industries in the 1970s, and since
the 1990s towards electronics, ICT, optics, and aerospace. In the 2000s, the quest for new
engines of growth led to the promotion of green industries, high-tech convergence, and
high value-added services (Koh, 2010).
The broad transformation of the industrial structure was accompanied by a rise in R&D
spending, led at first by the public sector and subsequently, from the 2000s, by the pri
vate sector, whose share in total R&D increased from 30 per cent in the late 1970s to 75
per cent in 2011 (Koh, 2010). The liberalization of imports was gradual and followed the
path set by industrial upgrading (Koh, 2010: 45–6). Indeed, as pointed out by Nassif et al.
(2016), the gradual, step-by-step approach to industrial policy in Asia (and its consistency
with macroeconomic policy) is in sharp contrast with the shock therapy observed in Latin
America (see also Suzigan and Furtado, 2006).
Page 16 of 33
Differences between Korea and Latin America were not confined to industrial policy, but
can also be found in their financial and macroeconomic policies. Korea combined import
substitution and export promotion to varying degrees throughout its development
process, but emphasized export promotion from the early 1960s. In this transition, the
won went through a major depreciation in February and October 1960 and in May 1964
(when the price of the dollar jumped from 130 to 255 won per dollar). The system of mul
tiple exchange rates was then replaced by a single fluctuating exchange rate (Koh, 2010:
34).
Financial market and credit allocation and interest rates remained tightly controlled in
the 1960s and 1970s. Several specialized public banks directed 40–60 per cent of domes
tic loans to selected industries, particularly in the capital and intermediate goods sectors
(Yoon and Kim, 1995).16 Such a policy, combined with export promotion (the ‘acid test’ of
learning), gave Korea the opportunity to redefine its comparative advantages.
For a long time, Korea had capital account regulations in place (Noland, 2007). As set out
by the former minister of trade, industry, and energy in Korea, Joong-Kyung Choi, strict
regulations of foreign exchange flows were crucial in the early stages of (p. 829) develop
ment, and probably necessary even in the mature stages (Choi, 2013: 108–9). In the light
of the impact of the 1997 crisis, he observed that ‘a financial crisis is much more painful
than inflation…Thus, controlling the cross-border flow of capital is more important than
containing inflationary pressures’ (Choi, 2013: 201).
As part of the process of joining the OECD in 1996, Korea eased its financial and capital
account regulations (Noland, 2007). This was not matched by the setting up of a domestic
macro-prudential framework to contain short-term debt and RER appreciation.17 The
rapid increase in the current account deficit, along with the short-term maturity of debts
contracted in the external market, led to the 1997 economic crisis. The intensity of the
crisis was aggravated by a contractionary fiscal and monetary policy adopted in Decem
ber 1997 and implemented in the first half of 1998—which was part of the agreement
signed with the IMF that provided a US$57 billion rescue package. Although the trend
continued to be towards opening the capital account, Korea has, in some cases, resorted
to safeguards to protect the country against financial instability and crises similar to
those of 1997 and 2008.18 In addition, the country sought to build up its stock of foreign
reserves to reduce its vulnerability to capital flight.
All in all, although the transition to democracy in the 1980s and the decision to join the
OECD led to some changes in economic policy, industrial and technology policies re
mained highly developmentalist, as reflected more recently in very high levels of R&D.
The 1997 crisis also led to the return of more interventionist macroeconomic policies
aimed at avoiding new financial crises.
Page 17 of 33
28.5 Stylized Facts: A Comparative View on Pro

ductivity and Specialization in Latin America
28.5.1 Phases of Convergence and Divergence
The impact of industrial policy on productivity and structural change should be seen from
a long-term perspective. Between 1950 and 1980, labour productivity in Latin America
was between 35 and 40 per cent of that in the United States, with a slight upward trend
(Figure 28.3A). In contrast, there is a great divergence in productivity trends since 1980:
it dropped to about one-fifth of US productivity between 1980 and 2002, experienced a
moderate recovery during the years of the commodity boom (‘the super-cycle’) of 2004–
14, to stabilize at approximately one-fourth of the comparative (p. 830) benchmark. The
stylized fact that emerges is quite clear: a phase of convergence with the developed world
between 1950 and 1980, followed by a phase of divergence since around 1980, with a
mild spurt of convergence in the commodity-boom years (see Correa and Stumpo, 2017).
Figure 28.3 Relative productivity vs. United States
Source: Economic Commission for Latin America and

the Caribbean (ECLAC), based on Total Economy.
The relative productivity in the four largest Latin American countries—particularly those
of Brazil and Mexico—also tended to converge with that of the United States until the late
1970s/early 1980s and to diverge thereafter (Figure 28.3B). In Argentina, (p. 831) diver
gence began in the mid-1970s, a few years before Brazil and Colombia, whereas it came a
bit later in Mexico, in the very early 1980s. In turn, the short period of convergence—or
interruption of divergence—in the 2000s benefited the South American economies, which
are important commodity exporters, but not Mexico. The relative productivity of Latin
American countries vis-à-vis the United States remained, in any case, at a dismal level. In
Page 18 of 33
versely, Korea began at a much lower level of labour productivity in the 1950s but went
through a persistent process of convergence with the United States and Latin America,
surpassing Brazil and Colombia in the mid-1980s, and Argentina and Mexico in the late
1990s.
These productivity patterns were also reflected in GDP trends. Overall, Latin American
GDP growth fell by half between 1950–80 and 1990–2018 (leaving aside the lost decade):
from 5.5 per cent to 2.8 per cent a year. Growth also became more volatile, as reflected in
two prolonged crises, in 1998–2003 and 2014–18. Among the four countries analysed
here, the long-term slowdown was particularly strong in Brazil and Mexico (which grew
by 2.4 and 2.7 per cent, respectively, in 1990–2018, versus 7.0 and 6.6 per cent in the pe
riod of state-led industrialization), followed by Colombia (3.5 versus 5.1 per cent); Ar
gentina had been a poor performer during state-led industrialization but delivered an
equally poor performance in the recent period (3.3 per cent growth in both cases).
An interesting comparison in terms of productivity performance is that of Brazil and Ko

rea. In 1950, labour productivity in Brazil was about 2.7 times higher than in Korea, while
in 2017 it was less than half. Convergence (in favour of Korea) was not uniform and was
much faster after Brazil entered into an era of divergence vis-à-vis the United States.
Brazil/Korea relative labour productivity declined by about 28 per cent in the four
decades between 1950 and 1980, and by about 74 per cent in the four decades between
1980 and 2010.
A similar pattern is obtained comparing the Latin American countries with developed
economies. For instance, labour productivity in Brazil was, on average, about 37–38 per
cent of that of Denmark in the 1950s and 1960s, and 47 per cent in the 1970s, attaining a
peak of 48 per cent in 1980, but relative productivity in Brazil fell afterwards. Since 1990
it has stabilized at about one-third of Danish productivity. Even during the commodity-
boom years, Brazilian/Danish relative productivity increased just one percentage point
(from 31 to 32 per cent, as estimated from data of the Total Economy Database of the
Conference Board).
28.5.2 Structural Change and Convergence
How does the evolution of relative productivity relate to structural change? A first aspect
to highlight is the evolution of the share of manufacturing in GDP. For Latin America as a
whole, Figure 28.4 shows the relevant trend, according to the three series of national ac
counts estimated by ECLAC. The period of stable productivity vis-à-vis the United States
coincided with a significant increase in the share of manufacturing in (p. 832) GDP, which,
as shown by Bértola and Ocampo (2012: ch. 4), depended on the size of the economies, in
terms of both strength and diversification. In contrast, the period of divergence since the
1980s has been accompanied by a strong de-industrialization, which in this case, was not
interrupted, but rather speeded up by the 2004–14 commodity boom. If we make a broad
abstraction from changes in the base years of the national accounts, Latin America is es
sentially back to the levels of industrialization it had in the early 1950s. As underscored
Page 19 of 33
by Palma (2005, 2012), the process can be characterized as one of premature de-industri
alization, as manufacturing lost ground in GDP at a lower level of GDP per capita than
when this process had started to take place in the developed countries.
Figure 28.4 Latin America: Share of manufacturing

in GDP, 1950–2017
Source: ECLAC.
The share of manufacturing in GDP declined in all the larger Latin American countries
(Figure 28.5). The decline started earlier, in the mid-1970s, in Argentina and Colombia,
and in the 1980s in Brazil and Mexico. The sharp decline of manufacturing in Argentina
was remarkable: it went from representing 40 per cent of GDP in 1965 (more than 20 per
centage points above Mexico) to 13 per cent in 2017 (4 percentage points below Mexico).
The collapse of Brazilian manufacturing after 1980 was also impressive, as was to a less
er extent that of Colombia. Thanks to its strong manufacturing export sector, the reduc
tion was much more moderate in Mexico. In contrast, manufacturing’s share in Korean
GDP increased by about 10 percentage points during the same period.
Figure 28.5 Share of manufacturing in GDP: Largest

Latin American countries vs. Korea
Source: World Bank national accounts data, https://

data.worldbank.org/indicator/nv.ind.manf.cpod.
Page 20 of 33
The composition of manufacturing production also changed. The share of engineering in
dustries (deemed to be the main drivers of technical change) in total manufacturing value
added declined, while the share of less technology-intensive sectors (p. 833) increased. To
keep a comparative approach, we computed the ratio between the share of the engineer
ing industries in total manufacturing value added in Argentina, Brazil, Colombia, Mexico,
and Korea relative to the same share in the United States (Figure 28.6). This ratio is the
Index of Relative Participation of engineering industries in total manufacturing value
added (IRP;19 see ECLAC, 2012). It increased until around 1980 in Brazil and Mexico and
remained more or less stable thereafter—with a fall in the 1980s and a strong recovery in
the 1990s in the case of Mexico, and smaller fluctuations in Brazil. In Argentina, this in
dex stagnated in the 1970s at lower levels than Brazil and Mexico by the later part of that
decade and began to fall in the 1980s, in such a way that it had lost about half of the val
ue it had in the early 1970s. Colombia had a less developed engineering sector compared
to the other Latin American countries analysed, and also experienced a decline in the
1990s. In contrast, Korea experienced a steady increase of this indicator, starting from
levels slightly higher than (p. 834) those of Colombia in the early 1970s, surpassing Brazil
and Mexico by the late 1980s and even surpassing the United States in the 1990s.
Figure 28.6 Index of the share of engineering indus

tries in manufacturing value added vs. United States
Source: Authors’ estimates based on ECLAC data.
Poor economic diversification was associated with very limited investment in R&D. Latin
America’s investment in R&D is only a fraction of the average for the OECD, with only
Brazil having attained levels higher than 1 per cent of GDP (Figure 28.7). The levels of in
vestment in R&D of Argentina and Brazil are less than half of those of Brazil and re
mained minimal in Colombia. The contrast with Korea is remarkable, as Korea (p. 835) in
vests more than 4 per cent of GDP, indeed one of the highest levels in the world, and in
recent years almost twice the average for the OECD.
Page 21 of 33
Figure 28.7 Research and development as % of GDP
Source: UNESCO.
Another direct indicator of technological capabilities is the NIS5 computed by Lee and
Lee (2019) (see Figure 28.8). Using US patent data, this index aims at providing a com
prehensive picture of the innovative output of different countries, based on a set of stan
dardized indicators of the patenting activity. The NIS5 is only available since 1991, but
does not include Colombia. Econometric exercises by Lee and Lee show that this index is
a more robust predictor of economic growth than other indicators used in the literature.
The evolution of the NIS5 points in the same direction as the previous indices when the
economic performance of the Latin American countries is compared with that of Korea.
First, Korea attained higher levels of NIS5 than in Latin American countries. Second, the
trend in Korea was mostly positive, while in the case of the Latin American countries, the
trend was more irregular, and there was no positive evolution over the long run. Brazil
shows an almost flat NIS5 for the whole period. Mexico shows some fluctuation, but its
most recent level is similar to that which it achieved in the early 1990s. This captures the
lack of backward linkages that the maquila-type export specialization, followed by Mexico
within the NAFTA agreement (being replaced by the United States–Mexico–Canada
Agreement, USMCA), contributing to technological learning in a laggard economy. In Ar
gentina, there was a fall with the crisis of 2002, a recovery after that, and finally a flat
trend with some fluctuations.20
Figure 28.8 Strength of the National Systems of In

novation (NIS5)
Source: Lee and Lee (2019).
Page 22 of 33
Summing up, the pattern that emerges from relative productivity is confirmed by the evo
lution of various indices of structural change and technological capabilities. (p. 836) The
decline in relative productivity in Latin America, particularly after 1980, and the more dy
namic performance of Korea, are associated with continuous structural change in the Ko
rean case and stagnation or even reversion of industrial transformation in Latin America.
Although these outcomes are the result of many different factors, the specific combina
tion of industrial and macroeconomic policies adopted in the Latin American countries
vis-à-vis Korea, as well as the low levels of investment in R&D, are crucial to explaining
why they fell behind.

The dismantling of the state-led industrialization policies that many Latin American coun
tries pursued in the post-war period—and in most cases since the 1930s—led to a diver
gence of productivity levels and technological capabilities vis-à-vis the developed coun
tries. This was the path followed by Argentina and Colombia since the mid-1970s, and by
Brazil and Mexico since the early 1980s. The return of industrial policies in the early
twenty-first century has been hesitant, failing to prevent the loss of knowledge-intensive
sectors in the production structure or to stimulate investments in R&D, except partially in
Brazil.
Structural change and specialization are stagnant, and there has been a reprimarization
of export structure in recent decades in South America. Mexico relies on a specialization
pattern based on exports of manufactures with low levels of domestic value added rela
tive to the more successful insertion in global value chains of countries like China and Ko
rea. In turn, Argentina, Brazil, and Colombia have reinforced their specialization in natur
al resources as a result of the 2004–14 commodity boom and their growing trade with
China. These paths are in sharp contrast with the experience of Korea, where active in
dustrial policies aimed at deepening technological development and high levels of R&D
spending have been the rule.
The political economy of industrial policy was critical for the success of Korea. ‘Recipro
cal control mechanisms’ helped discipline the large Korean firms and monitor the effec
tiveness of public policy in supporting the emergence of new industries and capabilities.
On the other hand, these mechanisms were fragile in Latin America before the debt crisis
and became still more fragile after it struck the region. The debt crisis first, the uncritical
adoption of the Washington Consensus later, and a political economy which favours the
capture of rents from natural resources over the creation of rents from technical change,
have compromised the effectiveness and the willingness of economic actors to implement
industrial policies.
Macroeconomic policy, and particularly its effects on RER, has also been crucial. Recur
rent RER appreciations have also contributed to the de-industrialization process and re
current external crises. Periods of high financial liquidity in international markets have
led to massive capital inflows, particularly when associated with capital account liberal
Page 23 of 33
ization. The use of the exchange rate as an inflation anchor in many (p. 837) periods has
contributed to that result. RER appreciation also resulted from the 2004–14 commodity
boom in the South American economies. This boom had adverse effects on manufacturing
and counteracted the positive impact of the moderate return of industrial policies. Al
though Korea was not immune to the adverse effects of financial liberalization and capital
account booms in the 1990s, it returned to a macroeconomic and financial policy that
aimed at avoiding these effects after the 1997 crisis it faced.
The tasks to be accomplished by the Latin American economies to catch up are still more
daunting now than in the past. The technological revolution is shifting the international
technological frontier very quickly, which makes it urgent for the region to finance and
carry out massive investments in education, R&D, and institutional capabilities to re
spond to this more demanding competitive challenge. In addition, the socio-political con
text is in flux (at the international and domestic levels): demands for equality and sustain
ability need to be made compatible with the quest for technological change. As long as in
dustrial policies continue to be a topic of secondary importance in the policy agenda, it
will be increasingly difficult for Latin America not to fall behind.
Acknowledgements
We are very grateful to Arkebe Oqubay, Eva Paus, Dirk Willem, Juan Carlos Moreno-Brid,
and Andre Nassif for comments on a previous version of this chapter.
References
Acosta, Jaime (2017) ‘Economía de la innovación, territorios y equidad para Colombia’,
Análisis, No. 3, Friedrich Ebert Stiftung, Bogotá. Available at http://library.fes.de/pdf-
files/bueros/kolumbien/13778.pdf.
Aghion, Philippe and Peter Howitt (1997) Endogenous Growth Theory. Cambridge, MA:
MIT Press.
Aldrighi, Dante and Renato Colistete (2013) ‘Industrial Growth and Structural Change:
Brazil in a Long-run Perspective’. Working Paper Series No. 2013–10, Department of Eco
nomics, University of Sao Paulo.
Amsden, Alice (2001) The Rise of the Rest: Non-Western Economies’ Ascent in World Mar
kets. Oxford: Oxford University Press.
Anderson, Martin and Olof Ejermo (2006) ‘Technology and Trade: An Analysis of Technol
ogy Specialization and Export Flows’. Working Paper No. 2006/05, CIRCLE, Lund Univer
sity.
Page 24 of 33
Arrow, Kenneth (2004) ‘Path Dependence and Competitive Equilibrium’, in William A.

Sundstrom and Timothy Guinnane (eds) History Matters: Essays on Economic Growth,
Technology and Demographic Change. Stanford, CA: Stanford University Press, pp. 23–
35.
Atkinson, Anthony and Joseph E. Stiglitz (1969) ‘A New View of Technological Change’,
Economic Journal 79(315): 573–8. (p. 838)
Baltar, Carolina (2015) ‘Inflation and Economic Growth in an Open Developing Country:
The Case of Brazil’, Cambridge Journal of Economics 39(5): 1263–80.
Bekerman, Marta and Gonzalo Dalmasso (2014) ‘Políticas productivas y competitividad in
dustrial: El caso de Argentina y Brasil’, Revista de Economia Política 34(1): 158–81.
Bell, Martin (2006) ‘Time and Technological Learning in Industrialising Countries: How
Long Does It Take? How Fast Is It Moving (If At All)?’, International Journal of Technology
Management 36(1–3): 25–39.
Bértola, Luis and José Antonio Ocampo (2012) The Economic Development of Latin Ameri
ca since Independence. Oxford: Oxford University Press.
Bogliacino, Francesco and Mario Pianta (2016) ‘The Pavitt Taxonomy, Revisited: Patterns
of Innovation in Manufacturing and Services’, Economia Política 33: 153–80.
Bresser-Pereira, Luiz Carlos (2008) ‘Dutch Disease and its Neutralization: A Ricardian Ap
proach’, Brazilian Journal of Political Economy 28(1): 47–71.
Bresser-Pereira, Luiz Carlos, Jose Luiz Oreiro, and Nelson Marconi (2014) Developmental
Macroeconomics. London: Routledge.
Cardoso, Eliana and Albert Fishlow (1989) ‘The Macroeconomics of the Brazilian External
Debt’, in Jeffrey Sachs (ed.) Developing Country Debt and the World Economy. Chicago,
IL: Chicago University Press, pp. 81–99.
Carvalho, Bernardo and Márcio Garcia (2008) ‘Ineffective Controls on Capital Inflows un
der Sophisticated Financial Markets: Brazil in the Nineties’, in Sebastian Edwards and
Márcio Garcia (eds) Financial Markets Volatility and Performance in Emerging Markets.
Cambridge, MA: National Bureau of Economic Research, pp. 29–79.
Cherkasky, Martín and Martín Abeles (2019) ‘Régimen monetario e instituciones labo
rales: una interpretación alternativa de la tendencia a la declinación del traspaso del tipo
de cambio (pass-through) en los países periféricos’, Revista de la CEPAL August: 7–37.
Chiang, Jing-Tsong (1991) ‘From “Mission Oriented” to “Diffusion Oriented” Paradigm:

The New Trend of US Industrial Technology Policy’, Technovation 11(6): 339–56.
Chinn, Menzie and Hiro Ito (2008) ‘A New Measure of Financial Openness’, Journal of
Comparative Policy Analysis 10(3): 309–22.
Page 25 of 33
Choi, Jiung-Koon (2013) Upside-Down Success Story of Korea’s Economic Development.

Seoul: Daewon.
Cimoli, Mario and Jorge Katz (2003) ‘Structural Reforms, Technological Gaps and Eco
nomic Development: A Latin American Perspective’, Industrial and Corporate Change
12(2): 387–411.
CONPES (Consejo Nacional de Política Económica y Social) (2016) ‘Política nacional de

desarrollo productivo’, Documento CONPES 3866, Bogotá, Departamento Nacional de
Planeación, 8 August.
Correa, Felipe and Giovanni Stumpo (2017) ‘Brechas de Productividad y Cambio Estruc
tural’, in Mario Cimoli, Mario Castillo, Gabriel Porcile, and Giovanni Stumpo (eds) Políti
cas Industriales y Tecnológicas en América Latina. Santiago: ECLAC.
Dornbusch, Rudiger and Juan Carlos de Pablo (1989) ‘Debt and Macroeconomic Instabili
ty in Argentina’, in Jeffrey Sachs (ed.) Developing Country Debt and the World Economy.
Chicago, IL: University of Chicago Press, pp. 37–56.
Dosi, Giovanni, Marco Grazzi, and Daniele Moschella (2015) ‘Technology and Costs in In
ternational Competitiveness: From Countries and Sectors to Firms’, Research Policy 44:
1795–814.
ECLAC (Economic Commission for Latin America and the Caribbean) (2012) Structural
Change for Equality: An Integrated Approach to Development. Santiago de Chile: United
Nations. (p. 839)
Fagerberg, Jan and Bart Verspagen (2002) ‘Technology Gap, Innovation-Diffusion and
Transformation: An Evolutionary Interpretation’, Research Policy 31: 1291–304.
Ffrench-Davis, Ricardo (2012) ‘Employment and Real Macroeconomic Stability: The Re
gressive Role of Financial Flows in Latin America’, International Labour Review 151(1–2):
21–41.
Ffrench-Davis, Ricardo and José Antonio Ocampo (2001) ‘The Globalization of Financial
Volatility’, in Ricardo Ffrench-Davis (ed.) Financial Crises in ‘Successful’ Emerging
Economies. Washington, DC: Brookings Institution Press/ECLAC, pp. 1–37.
Frenkel, Roberto and Martin Rapetti (2011) ‘A Concise History of Exchange Rate Regimes
in Latin America’, in José Antonio Ocampo and Jaime Ros (2011), The Oxford Handbook of
Latin American Economics. Oxford: Oxford University Press, pp. 187–213.
Freund, Caroline and Martha Pierola (2012) ‘Export Surges: The Power of a Competitive
Currency’, Journal of Development Economics 97(2): 387–95.
Frieden, Jeffry (2015) Currency Politics: The Political Economy of Exchange Rate Policy.
Page 26 of 33
Goldfajn, Illan and Andre Minella (2005) ‘Capital Flows and Controls in Brazil: What Have
We Learned?’, NBER Working Paper No. 11640. Cambridge, MA: NBER.
Guzmán, Martin, José Antonio Ocampo, and Joseph E. Stiglitz (2018) ‘Real Exchange Rate
Policies for Economic Development’, World Development 110: 51–62.
IMF (International Monetary Fund) (2003) ‘Lessons from the Crisis in Argentina’. Pre
pared by the Policy Development and Review Department in consultation with the other
Departments, 8 October, p. 63.
Jenkins, Rhys and Alexandre de Freitas Barbosa (2012) ‘Fear for Manufacturing? China
and the Future of Industry in Brazil and Latin America’, The China Quarterly 209: 59–81.
Kaltenbrunner, Annina and Juan Pablo Paincieira (2015) ‘Developing Countries’ Changing
Nature of Financial Integration and New Forms of External Vulnerability: The Brazilian
Experience’, Cambridge Journal of Economics 39(5): 1281–306.
Katz, Jorge and Giovanni Stumpo (2001) ‘Sectoral Regimes, Productivity and Internation
al Competitiveness’, CEPAL Review 75: 137–59.
Khan, Mushtaq (2000) ‘Rent-seeking as Process’, in Mushtaq Khan and Jomo Kwame Sun
daram (eds) Rents, Rent-Seeking and Economic Development: Theory and Evidence in
Asia. Cambridge: Cambridge University Press, pp. 70–139.
Kim, Eun Mee (2011) ‘The Korean Development Alliance between Business, Labour and
Government’. Paper presented to the seminar organized by UNRISD and KOICA, Seoul,
South Korea, 29 September.
Kim, Linsu (1997) ‘Korea’s National Innovation System in Transition’. Science and Tech
nology Policy Institute, Korea.
Koh, Youngsun (2010) ‘The Growth of Korean Economy and the Role of Government’, in Il
SaKong and Youngsun Koh (eds) The Korean Economy: Six Decades of Growth and Devel
opment. Seoul: Korean Development Institute, pp. 9–82.
Laplane, Mariano and Andrea Laplane (2017) ‘Planes industriales y los desafíos del desar
rollo sostenible en Brasil’, in Mario Cimoli, Mario Castillo, Gabriel Porcile, and Giovanni
Stumpo (eds) Políticas Industriales y Tecnológicas en América Latina. Santiago: ECLAC,
pp. 133–74.
Lavarello, Pablo and Matías Mancini (2017) ‘Política industrial y recuperación manufac
turera en Argentina’, in Mario Cimoli, Mario Castillo, Gabriel Porcile, and Giovanni
Stumpo (eds) Políticas Industriales y Tecnológicas en América Latina. Santiago: ECLAC,
pp. 79–132. (p. 840)
Page 27 of 33
Lee, Keun (2013) Schumpeterian Analysis of Economic Catch-Up. Cambridge: Cambridge

University Press.
Lee, Keun and Jongho Lee (2019) ‘National Innovation Systems, Economic Complexity,
and Economic Growth: Country Panel Analysis Using the US Patent Data’, Journal of Evo
lutionary Economics. Available at https://doi.org/10.1007/s00191-019-00612-3.
Lim, Phillip Wonhyuk (2004) ‘Path Dependence in Action: The Adoption and Persistence of
the Korean Model of Economic Development’, in Timothy Guinnane, William Andre Sund
strom, and Warren Whatley (eds) History Matters: Essays on Economic Growth, Technolo
gy and Demographic Change. Stanford, CA: Stanford University Press, pp. 142–62.
Lo, Dic and Mei Wu (2014) ‘The State and Industrial Policy in Chinese Economic Develop
ment. Geneva: UNCTAD-ILO, pp. 307–26.
Lundvall, Bengt-Åke (2016) ‘From the Economics of Knowledge to the Learning Econo
my’, in Bengt-Åke Lundvall (ed.) The Learning Economy and the Economics of Hope. Lon
don: Anthem Press, pp. 133–151.
Moreno-Brid, Juan Carlos (2016) ‘Industrial Policy: A Missing Link in Mexico’s Quest for
Export-led Growth’, Latin American Policy 4(2): 216–37.
Moreno-Brid, Juan Carlos and Jaime Ros (2009) Development and Growth in the Mexican
Economy. A Historical Perspective. Oxford: Oxford University Press.
Moreno-Brid, Juan Carlos and Gabriela Dutrénit (eds) (2018) ‘Políticas de Desarrollo Pro
ductivo en México’, OIT Américas, Informes Técnicos, 2018/12, October.
Nassif, Andre, Luiz Carlos Bresser-Pereira, and Carmen Feijó (2018) ‘The Case for Rein
dustrialisation in Developing Countries: Towards the Connection between the Macroeco
nomic Regime and the Industrial Policy in Brazil’, Cambridge Journal of Economics 42:
355–81.
Nassif, Andre, Carmen Feijo, and Eliane Araujo (2015) ‘Structural Change and Economic
Development: Is Brazil Catching up or Falling behind?’, Cambridge Journal of Economics
39(5): 1307–32.
Nassif, Andre, Carmen Feijo, and Eliane Araujo (2016) ‘Structural Change, Catching up
and Falling behind in the BRICS: A Comparative Analysis Based on Trade Patterns and
Thirlwall’s Law’, PSL Quarterly Review 69(278): 373–421.
Noland, Marcus (2007) ‘South Korea’s Experience with International Capital Flows’, in
Sebastian Edwards (ed.) Capital Controls and Capital Flows in Emerging Economies: Poli
cies, Practices and Consequences. Chicago, IL: University of Chicago Press, pp. 481–528.
Page 28 of 33
Ocampo, José Antonio (2016) ‘Balance-of-Payments Dominance: Implications for Macro

economic Policy’, in Mario Damill, Marín Rapetti, and Guillermo Rozenwurcel (eds)
Macroeconomics and Development: Roberto Frenkel and the Economies of Latin America.
New York: Columbia University Press, pp. 211–28.
Ocampo, José Antonio and Paola Arias (2018) ‘Colombia’s System of National Develop
ment Banks’, in Stephany Griffith-Jones and José Antonio Ocampo (eds) The Future of Na
tional Development Banks. Oxford: Oxford University Press, pp. 166–95.
Ocampo, José Antonio and Carmen Astrid Romero (2015) ‘La búsqueda, larga e incon
clusa de un nuevo modelo (1981–2014)’, in José Antonio Ocampo (ed.) Historia económica
de Colombia. Bogotá: Fondo de Cultura Económica and Fedesarrollo, pp. 293–372.
OECD (Organisation for Economic Co-operation and Development) (2019) Production

Transformation Policy Review of Colombia: Unleashing Productivity. Paris: OECD Devel
opment Pathways. (p. 841)
Oreiro, José Luiz, Lionello Punzo, and Eliane Araujo (2012) ‘Macroeconomic Constraints
to Growth of the Brazilian Economy: Diagnosis and Some Policy Proposals’, Cambridge
Journal of Economics 36(4): 919–39.
Palma, José Gabriel (2005), ‘Four Sources of “De-industrialization” and a New Concept of
the “Dutch Disease”’, in José Antonio Ocampo (ed.) Beyond Reforms: Structural Dynamics
and Macroeconomic Vulnerability. Palo Alto, CA: Stanford University Press, ECLAC and
World Bank, pp. 71–116.
Palma, José Gabriel (2012) ‘How the full opening of the capital account to highly liquid fi
nancial markets led Latin America to two and a half cycles of “mania, panic and crash”’.
Cambridge Working Papers in Economics No. 1201, January.
Peres, Wilson (2010) ‘The (Slow) Return of Industrial Policies in Latin America and the
Caribbean’, in Mario Cimoli, Giovanni Dosi, and Joseph E. Stiglitz (eds) The Political Econ
omy of Capabilities Accumulation: The Past and Future of Policies for Industrial Develop
ment. Oxford: Oxford University Press, pp. 175–201.
Peres, Wilson and Annalisa Primi (2019) ‘Industrial Policy and Learning: Lessons from
Latin America’, in Arkebe Oqubay and Kenichi Ohno (eds) How Nations Learn: Technolog
ical Learning, Industrial Policy, and Catch-up. Oxford: Oxford University Press, pp. 207–
34.
Prates, Daniela, Andre Cunha, and Marcus Lelis (2009) ‘La Gestión del Régimen Cam
biario en Brasil’, Revista de la CEPAL 99: 97–118.
Primi, Annalisa (2014) Promoting Innovation in Latin America. Maastricht: Maastricht

University Press.
Razmi, Arslam, Martin Rapetti, and Peter Skott (2012) ‘The Real Exchange Rate and Eco
nomic Development’, Structural Change and Economic Dynamics 23(2): 151–69.
Page 29 of 33
Rodrik, Dani (1994) ‘Getting Interventions Right: How South Korea and Taiwan Grew
Rich’. NBER Working Paper No. 4964. Cambridge, MA: NBER.
Rodrik, Dani (2008) ‘The Real Exchange Rate and Economic Growth’, Brookings Econom
ic Studies Program 39(2): 365–439.
Salvia, Sebastian Pedro (2015) ‘The Boom and Crisis of the Convertibility Plan in Argenti
na’, Revista de Economia Política 35(2): 325–42.
Santarcángelo, Juan E., Daniel Schteingart, and Fernando Porta (2018) ‘Industrial Policy
in Argentina, Brazil, Chile and Mexico: A Comparative Approach’, Revue Interventions
économiques, Papers in Political Economy 59: 1–41.
Stalling, Barbara and Wilson Peres (2000) Growth, Employment and Equity: The Impact
of Economic Reforms in Latin America and the Caribbean. Washington, DC: Brookings In
stitution Press.
Stumpo, Giovanni and Diego Rivas (2013) ‘La industria manufacturera argentina 2003–
2010. Reconstruyendo capacidades para enfrentar los desafíos del siglo XXÏ’, in Giovanni
Stumpo and Diego Rivas (eds) La industria argentina frente a los nuevos desafíos y opor
tunidades del siglo XXI. Santiago: ECLAC, pp. 27–60.
Suzigan, Wilson and João Furtado (2006) ‘Política Industrial e Desenvolvimento’, Journal
of Political Economy 26(2): 163–85.
Wade, Robert H. (1992) ‘East Asia’s Economic Success: Conflicting Perspectives, Partial
Sights, Shaky Evidence’, World Politics 44(2): 270–320.
Yoon, Je Cho and Joon-Kyung Kim (1995) ‘Credit Policies and the Industrialization of Ko
rea’. World Bank Discussion Paper No. 286. Washington, DC: World Bank.
Notes:
(1) Following Bértola and Ocampo (2012), we use this concept rather than the more tradi
tional one of ‘import-substituting industrialization’, because the latter captures only one
element of the policy package of that period, and not necessarily the most relevant one.
(2) ‘It is sometimes argued that the problem of the allocation of resources to research is
not relevant for a present-day underdeveloped country, since it will benefit from technical
progress in the advanced countries, and any independent research would simply be a du
plication of effort. But if, as we have suggested, technical knowledge is highly specific to
particular production processes, this will not be the case’ (Atkinson and Stiglitz, 1969:
575).
(3) The RER is defined as q=PfE/P, where Pf are foreign prices and P domestic prices, and
E the nominal exchange rate, expressed as the value of the foreign currency in terms of
the domestic currency.
Page 30 of 33
(4) In some cases, the strategies went back to the 1930s or even to the late nineteenth
and early twentieth centuries. See Bértola and Ocampo (2012).
(5) The authors point out that the high share of intermediate Chinese inputs in manufac
turing production in Brazil and Mexico implies the risk of a process of ‘hollowing out’ of
the manufacturing sector if domestic policies are not changed.
(6) See Stalling and Peres (2000). The liberalization drive started earlier in Mexico, with
the De la Madrid administration (1982–88). In 1990, Mexico adopted a Programme for
Modernizing Industry and Trade (Programa Nacional de Modernización Industrial y de
Comercio Exterior, PRONAMICE, 1990–94), which would subsequently be complemented
by further liberalization, as pointed out in the main text.
(7) There were three additional development banks, not relevant for the analysis here:
one for rural development, another for regional and local government projects, and a
third one for energy investment, which was later given the responsibility to promote pub
lic–private infrastructure projects.
(8) Brazil’s Real Plan was less rigid that Argentina’s Convertibility Plan and was aban
doned earlier than the latter (see Salvia, 2015). In addition, Brazil imposed a tax on capi
tal inflows as a response to their surge in 1993–96 (Goldfajn and Minella, 2005; Carvalho
and Garcia, 2008), which failed, in any case, to prevent the appreciation of the real, given
the attractiveness of the Brazilian market due to high domestic interest rates. Brazil fell
into what has been called a ‘low RER x high interest rates trap’ (Oreiro et al., 2012; see
also Prates et al., 2009). After the 1997–99 turbulences, capital account regulations were
lifted (see Baltar, 2015). In any case, these regulations made the Brazilian crisis of Janu
ary 1999 (which marked the end of the Real Plan) milder and shorter than the Argentine
crisis of 2002. On the crisis of the late twentieth century in Colombia, in contrast to its
performance during the Latin American debt crisis, see Ocampo and Romero (2015: 324–
40).
(9) Kaltenbrunner and Paincieira (2015: 1287) observed that the composition of financial
investments was heavily concentrated in short-term maturity assets.
(10) See IMF (2003) for an analysis of this crisis.
(11) The concept was ill-defined, but probably made reference to the need to spread pro
ductivity gains from a small group of exporters to the rest of the economy.
(12) See the recent OECD-WTO data in this regard (https://www.oecd.org/sti/ind/measur

ing-trade-in-value-added.htm#access).
(13) For an evaluation of existing programmes, see Acosta (2017) and OECD (2019). The
latter proposes an interesting set of initiatives for the country in this field.
(14) The classic work is Amsden (1989). See also Kim (1997, 2011), Lee (2013), Rodrik
(1994), and Wade (1992).
Page 31 of 33
(15) China may be seen as the last (and more dramatic) example of the flying-geese strate
gy of industrialization, but it has changed the pattern, and now dominates Asian manufac
turing. See Lo and Wu (2014).
(16) Noland (2007: 486) points out that ‘modest financial-sector liberalizations that had
been undertaken in the late 1960s were reversed in 1972, when interest rates were low
ered and direct government control of the banking system was increased in order to
channel capital to preferred sectors, projects, or firms’.
(17) A Financial Supervisory Commission responsible for setting regulations and stan
dards in the financial market was established in 1998.
(18) For instance, in 2010 Korea placed some restrictions on derivative markets, but set
ting limits on the foreign-exchange derivatives relative to the capital base of the financial
institutions.
(19) The IRP is defined as IRP=(ei/mi)/(eUSA/mUSA), where e is engineering value added

and m total manufacturing value added; the superscript i denotes the country and the su
perscript USA denotes the reference country, in this case, the United States. The engi
neering industries are: metal products, non-electrical machinery, electrical machinery,
transport equipment, and scientific and professional instruments (categories 381 to 385
of the Standard Industrial Classification, SIC).
(20) Lee and Lee (2019) reach a similar conclusion running growth regressions using the
NIS5 in the set of explanatory variables: ‘economies with successful growth experiences,
such as South Korea, China, and Taiwan…mostly show upward sloping lines over time.’
These authors also note that in economies with less successful growth experiences, such
as Brazil, Mexico, Thailand, and South Africa, ‘the NIS5 index does not increase much in
certain periods or even declines’.
José Antonio Ocampo is professor at Columbia University’s School of International

and Public Affairs, and chair of the Committee for Development Policy of the United
Nations Economic and Social Council (ECOSOC). He has occupied numerous posi
tions at the United Nations and his native Colombia, including UN under-secretary-
general for economic and social affairs, executive secretary of the UN Economic
Commission for Latin America and the Caribbean (ECLAC), and member of the board
of directors of Banco de la República (Colombia’s central bank), minister of finance,
minister of agriculture and director of the National Planning Office of Colombia.
Gabriel Porcile
Gabriel Porcile is an economic affairs officer at ECLAC and professor of economics at

the Federal University of Parana, Brazil. He holds a PhD in economic history from the
London School of Economics. He has been a visiting scholar at the University of Sao
Paulo, Brazil, University of the Republic, Uruguay and Columbia University, United
Page 32 of 33
States. His main research topics are economic growth, distribution, structural
change, and the political economy of development. His latest publications are ‘New
Structuralism and the Balance-of-Payments Constraint’, Review of Keynesian Eco
nomics, 7, 2019 (with Giuliano Yajima) and ‘A Technology Gap Interpretation of
Growth Paths in Latin America and Asia’, Research Policy, 48, 2019 (with Mario
Cimoli and Joao Basilio Pereima).
Page 33 of 33
Phases and Uneven Experiences in African Industrial Policy
Phases and Uneven Experiences in African Industrial

Policy
Lindsay Whitfield and Nimrod Zalk
The chapter examines the challenges of implementing industrial policy in selected sub-Sa
haran African (SSA) countries with a particular focus on raising the capabilities of locally
owned firms. Under hyper-globalization, entry even into the lowest rungs of global value
chains is challenging, as reflected by Ethiopia’s efforts to enter apparel production and
Kenya’s and Rwanda’s experience with global value chains in knowledge-based services.
Nigeria and Ethiopia succeeded in establishing domestic cement industries dominated by
local firms, but face constraints in domestic demand and challenges with oligopolistic
structures in domestic markets, which can produce challenges for regulating markets and
redistributing wealth. South Africa reflects a case of middle-income de-industrialization
and efforts at re-industrialization. Industrial policies in automotive and apparel sectors
have had some positive impact but have been constrained by mutually reinforcing forces
of disengagement of large business groups from diversified manufacturing and rising cor
ruption exacerbated by unsupportive macroeconomic policy.
Keywords: Africa, structural transformation, industrial policy, Ethiopia, Kenya, Rwanda, Nigeria, South Africa
29.1 Introduction
SUB-SAHARAN AFRICAN (SSA) countries experienced a growth spurt from the
mid-1990s to the mid-2010s, after decades of economic stagnation and even contraction.
However, despite over two decades of growth, very few African countries created manu
facturing industries that were internationally competitive and diversified their exports
away from dependence on a few primary commodities, and most countries were still im
porting the majority of their manufactured goods (UNECA, 2016). The 2007‒08 global fi
nancial crisis and the associated collapse of the global commodity ‘supercycle’ by the
mid-2010s highlighted the pitfalls of a narrow reliance on commodities and revived inter
est in structural transformation as key to providing jobs, increasing incomes, and raising
Page 1 of 26
the standard of living. There has been a slew of publications and processes exploring the
prospects for a revival of industrial policy in SSA.
Structural transformation is not just about attracting human and physical capital out of
low-productivity and subsistence economic activities and into more productive enterpris
es. This definition does not capture the multiple, complex, and interacting processes that
drive and sustain productivity growth. Rather, transformation involves moving the econo
my away from being a set of assets based on primary products exploited by unskilled
labour towards an economy built on knowledge-based assets exploited by skilled labour
(Amsden, 2001: 2). The term ‘technological capabilities’ refers to these knowledge-based
assets: technical, managerial, and organizational skills that firms need in addition to for
mal education and scientific knowledge in order to achieve the level of (p. 843) productivi
ty that established firms have achieved, which set the (international) market standard.
The more technological capabilities that firms have, the greater their ability to sustain na
tional income growth by moving into higher-value economic activities and innovating as
well as responding flexibly to changing competitive conditions. The technological capabil
ity approach to understanding structural transformation and its concomitant focus on in
dustrial policy as promoting local firm learning and capability building stems from an evo
lutionary economics perspective (see Cimoli et al., Chapter 4 in this volume). In examin
ing industrial policies aimed at transforming African economies in the twenty-first centu
ry, this chapter focuses on how these industrial policies help locally owned firms become
internationally competitive through building these capabilities.
The chapter also examines the challenges that late-late industrializing countries in SSA
face in the context of a global economy defined by international trade in intermediate
goods through global value chains (Gereffi, 2014). Historically, the manufacturing sector
is where technological capabilities were nurtured and wealth was produced as a result of
imperfect competition, barriers to entry, and increasing returns, and this is why industri
alization was key to generating wealth in Northern European countries and the United
States (Reinert, 2007). When other countries tried to emulate this process, they used a
combination of import-substitution and export-oriented industrial policies to build similar
manufacturing industries (Schwartz, 2010). However, the expansion of manufacturing ca
pabilities to more countries in the post-Second World War period and the concomitant
outsourcing of manufacturing by Western and Japanese corporations led to the rise of
global value chains. This took place in the context of extensive liberalization of trade, a
dramatic expansion of the global workforce engaged in global trade, and rising pressure
on corporations to raise returns to shareholders. Western and Japanese corporations now
focus on economic activities in the pre- and post-manufacturing stages, such as research
and development, branding, marketing, and retail, where they can still create high barri
ers to entry through proximity to markets, first-mover advantages, and intellectual prop
erty rights. Manufacturing activities are fragmented and dispersed globally in a spatial
hierarchy of production sites depending on countries’ wage costs and their firms’ capabil
ities (Schwartz, 2010; Coe and Yeung, 2015). Production activities and functions in global
value chains are characterized by different levels of competition, generating different lev
els of wealth and income, hence the hierarchy. The lowest function in buyer-driven global
Page 2 of 26
value chains, the assembly of imported inputs, is characterized by near perfect competi
tion (including perfect substitutability). As a result, firms producing this type of manufac
tured export are price takers, and it is global buyers generally located in developed coun
tries that capture productivity increases rather than the developing-country producers, as
was the case with agricultural exports from developing countries in the twentieth century
(Kaplinsky, 2005). At the same time, agribusiness industries and knowledge-based service
industries now have characteristics that manufacturing historically did, such as sophisti
cated organization and production processes that require technological capabilities to be
built (and thus increasing returns through (p. 844) learning) and the ability to create
wealth through barriers to entry through research and development, specialization,
economies of scale, and so on.1
As a result, industrial policies can target agribusiness, manufacturing, and knowledge-

based services for international, regional, and domestic markets as a way to build the
technological capabilities of local firms, generate wealth, tap into high demand through
export markets, and create deeper linkages within the domestic economy. However, SSA
countries may not be able to use the same industrial policy strategies as previously indus
trializing countries. The import-substitution industrialization (ISI) and export-oriented in
dustrialization (EOI) policies of past decades were never mutually exclusive alternatives;
rather, they were used in combination to access foreign knowledge but also to protect lo
cal firms while they were learning. However, it has become difficult for SSA late-late in
dustrializers to implement them in the same way. Entry into a new industry and its export
markets generally requires entry into global value chains, often through assembly of im
ported parts. Thus, industrialization begins with vertically specialized industrialization
and generally low-value-added exports, then local firms upgrade within the global value
chain by producing more of the imported inputs locally (Milberg et al., 2014). In other
words, firms start with a new form of EOI, acquiring foreign tacit knowledge and social
connections to global buyers and input suppliers, and then move to a new form of ISI that
works backwards from assembly to intermediate inputs to raw materials. Thus, successful
industrialization through global value chains corresponds with declining vertical special
ization. It also increasingly involves a move from export markets to regional and domestic
markets, so that local firms can move into branding, marketing, and retailing (where
more wealth can be captured), and then sometimes a re-coupling with global value chains
where local firms enter at higher-value functions (Coe and Yeung, 2015).
However, many SSA countries are still at the early stages of diversifying their economies
away from natural resources, building knowledge-based assets, and creating deeper link
ages within the domestic economy, especially in the context of the colonial trading
economies that they inherited at independence (see Whitfield, 2018). Thus, they need in
dustrial policies focusing on domestic industries and non-tradable sectors that face differ
ent challenges from entering global value chains, such as the need to stimulate domestic
demand through government expenditure, linkages between agriculture and industrial
ization, and promoting several industries at the same time that can concurrently generate
Page 3 of 26
sustained domestic demand. These are old challenges discussed by structural develop
ment economists, but still very real for SSA countries.
The aim of this chapter is to discuss these issues through an examination of the recent ex
periences of selected SSA countries implementing industrial policies. Notably, many SSA
governments have not pursued industrial policies, reflecting policy continuity from the
period of structural adjustment (see Chitonge and Lawrence, Chapter 30 in this volume).
The cases of industrial policy examined in this chapter focus on types of (p. 845) indus
tries, markets, and industrial policies, and not on the entirety of country experiences. The
first case involves industrial policies to promote the cement industry, a non-tradable
heavy industry, in Nigeria, contrasted with Ethiopia. We focus not only on how local firms
emerged in an industry dominated by multinational corporations in most SSA countries,
but also on how domestic market demand was created and sustained. The second case
considers industrial policies in Ethiopia to create an apparel export industry. It examines
the challenges for local firms entering labour-intensive manufacturing global value
chains, and the challenges for SSA governments in implementing industrial policies that
compel local firms to build the required capabilities as well as to create more of the value
chain within the country. The third case explores industrial policies to promote knowl
edge-based service sectors in Rwanda and Kenya. In particular, it examines their govern
ments’ attempts to promote local firm participation in export-oriented knowledge services
such as business-process outsourcing. The fourth and final case assesses industrial poli
cies in South Africa aimed at diversifying manufacturing beyond heavy industries, focus
ing on the automobile and apparel sectors. In doing so, it illustrates some of the chal
lenges of using industrial policy to promote apparel production for export when local
firms are domestically oriented, and to promote greater production of inputs locally for
the export-oriented automobile sector and thus move beyond the early stage of vertical
specialization industrialization through global value chains.
Following the evolutionary economics approach, the chapter takes a firm-level perspec
tive on industrial policy theoretically and empirically. While capability-building processes
are embedded within broader industry-level, national, and international processes, they
take place at the firm level. Section 29.2 provides an overview of the limited progress in
SSA of developing manufacturing sectors. We consider why this has been the case by ex
ploring the role of industrial policy in supporting and compelling local firms to build their
technological capabilities. Industrial policy in the targeted sectors in Nigeria, Ethiopia,
Rwanda, Kenya, and South Africa is then presented. In concluding, we draw out some of
the implications of these sector experiences.
29.2 Limited Progress with Structural Transfor

mation in Sub-Saharan Africa
SSA countries struggle with industrialization more than other developing-country regions
as a result of historical factors that have been compounded, creating path dependencies
that are hard to escape. At independence these countries found themselves far behind the
Page 4 of 26
world technological frontier and lacked skill endowments, especially in terms of experi
ence in manufacturing compared to other developing countries (Amsden, 2008). This lim
ited manufacturing experience in most SSA countries at the (p. 846) start of decoloniza
tion was an important factor shaping the poor performance of import-substitution indus
trialization strategies of the 1960s and 1970s, as firms started with very low capabilities
(Kennedy, 1988). In addition, import-substitution industrialization strategies in SSA coun
tries generally focused on capital investments and often failed to build the technological
capabilities of local firms, regardless of the ownership structure: public, private, or pub
lic‒private joint ventures (Stewart et al., 1992). Despite these weaknesses, some African
countries were beginning to gain international competitiveness in areas of labour-inten
sive manufacturing by the 1980s (Sundaram and Von Arnim, 2008). However, external
debt crises and fiscal deficits led SSA countries to adopt structural adjustment policies in
the 1980s and 1990s that undermined those gains (for a review, see Whitfield, 2017).
The limited stock of technological capabilities explains why manufacturing and agro-pro
cessing industries in SSA countries were hit so hard by structural adjustment, especially
compared to other developing-country regions (Whitfield et al., 2015). Structural adjust
ment reforms removed types of state intervention that in some cases turned out to be
more harmful than helpful, but the ideological position embedded in these policies of
complete reliance on the market also meant that these reforms had important negative
economic effects. Trade liberalization under structural adjustment programmes, which
was more rapid in SSA countries than their Asian counterparts, largely wiped out low-
productivity African firms because they were given no buffer or support to become com
petitive (Lall, 1995; Sundaram and Von Arnim, 2008). SSA countries that had more com
petitive industries before structural adjustment did better in certain sectors.
The growth revival starting in the mid-1990s was driven largely by increases in interna
tional commodity prices; economic liberalization, which led to the allocation of resources
to sectors in which SSA countries already had comparative advantages; new discoveries
of natural resources, using foreign direct investments to extract them; and government
spending fuelled by foreign aid (Arbache and Page, 2009; Whitfield et al., 2015: 60‒7).
From the mid-2000s, SSA growth was increasingly driven by China’s rapid industrializa
tion, which fuelled demand for the region’s natural resources, as well as Chinese direct
investments in infrastructure, primary sectors, and manufacturing sectors in SSA
economies (Farooki and Kaplinsky, 2012). However, when growth in China slowed so did
growth in SSA countries, as the international prices of the commodities they exported be
gan to fall significantly from 2014.
The subsequent moderation of SSA growth brought into sharp focus the limited sustain
ability of growth and welfare improvements without underlying structural transformation.
Since the 1980s, SSA has undergone consistent declines in the share of manufacturing in
GDP; output and employment have fallen, and the diversity and complexity of manufactur
ing sectors have declined (Page, 2012). Thus, the quality of growth was characterized by
‘static gains’ but ‘dynamic losses’ with respect to structural transformation (de Vries et
al., 2015). Projected medium-term growth on the continent is insufficient to make a dent
Page 5 of 26
in unemployment due to population growth and economic structures that continue to de
pend heavily on primary commodities and the extractive sector (ADB, 2019).
29.3 Theoretical Perspective: Using Indus

(p. 847)
trial Policy to Escape the Learning Trap

In all developing countries, locally owned firms in industries new to the country are ini
tially often unprofitable due to their low productivity, which stems from constraints exter
nal and internal to firms. Internal constraints have to do with low technological capabili
ties. Technological capabilities can only be acquired through experience and purposive in
vestment in learning (Lall, 1992). Firms have to undergo a period of learning in which
they may well operate at a loss, and it is uncertain how long they will do so. The risk is
highest for the first investors when the knowledge and infrastructure required to become
competitive in new economic activities do not yet exist and have to be acquired and
adapted to local conditions. For local investors, the cost and uncertainty involved in learn
ing is too high to be worth the risk, leading to what Khan (2013) calls a learning trap: lo
cal firms do not invest in building their capabilities because the gap between their initial
capabilities and what is required to be internationally competitive is too large, leaving
them stuck in low-productivity activities. This learning trap is particularly acute in SSA
countries, where local firms have very low initial capabilities and limited capital, knowl
edge, and networks on which to draw, leading to a large capability gap.
In addition, all firms (local and foreign) in new industries in SSA countries initially face
constraints external to the firm that result in high market costs: high-cost and limited ac
cess to capital, land, and skilled labour; high cost and poor quality of infrastructure and
transport logistics; limited access to foreign exchange or banking instruments tailored to
exporting; and so on. These constraints drive up the cost of production for firms, but they
are costly or impossible for individual firms to address on their own. Furthermore, they
make it even harder for locally owned firms to bridge the capability gap, as it increases
the costs and risk.
Whether local entrepreneurs decide to invest in a new industry depends on their percep
tion of the risk, which in turn is shaped by the size of their capability gap plus the market
costs that all firms face, but also the normal profit rate in the domestic market or the
global value chain in a particular function and end market when global productivity
norms are achieved. Khan (2019) refers to the normal profit rate as the size of the prize,
and it is determined largely by the degree of competition (or substitutability). Local in
vestors compare the ‘prize’ in new industries to the risk/profit ratio of other economic op
portunities in the national economy. The issue in SSA countries is often not the lack of
capital, but the capability gap, which leads local investors to put their capital into eco
nomic opportunities that require lower capabilities and thus have lower risks, such as the
import trade, real estate, and hotels.
Page 6 of 26
Escaping from the learning trap requires industrial policies that reduce the risk for local
firms of investing in new manufacturing activities by subsidizing the initial costs (p. 848)
of investment, reducing market costs, and facilitating and subsidizing the learning
process. But these industrial policies are difficult to implement because they involve gov
ernment agencies creating rents and ensuring that local firms use the rents to invest in
learning (Whitfield et al., 2015; Khan, 2009). The norm is for local firms to engage in ‘sat
isficing behaviour’, and the larger the capability gap, the more likely they are to do so
(Khan, 2013). Thus, industrial policies that are essentially learning rents have to be tied
to performance targets: what Amsden (2001) called reciprocal control mechanisms. In
this context, ex post rents (conditional on the achievement of competitive success) are of
ten more successful in developing countries than ex ante rents (those provided before
performance targets are met) because they do not depend on government agencies hav
ing the technical capacity and political power to enforce them (Di Maio, 2009; Khan, 2019).
However, the process of creating and managing rents as part of industrial policy can easi
ly become entangled with the processes of rent creation and distribution linked to politi
cal stability and the survival of the ruling political elites (Khan, 2009). Furthermore, the
capacity of a state to monitor and enforce learning rents is derived from the ruling elites
who govern the state and the ruling coalition that keeps them in power. What matters is
how coalitional pressures shape the political costs of certain policies and the ability to im
plement them, given the resistance or support from powerful groups within and outside
the ruling coalition (Whitfield et al., 2015). Thus, domestic politics is just as important as
economics in determining the success of industrial policies.
29.4 Import-substitution Industrial Policy: The

Cement Sector in Nigeria and Ethiopia
Cement is an intermediate or heavy industry with large economies of scale and capital in
tensity and is a core input in the construction industry. Several Asian countries used the
cement industry as a springboard for industrialization and to develop technological capa
bilities within local firms, including professional management skills (on Korea, see Ams
den, 1989). Cement is characterized as a low-value, high-volume product, so it is general
ly not traded internationally. Cement is a classic oligopoly due to high entry and exit bar
riers, and requires certainty of market demand. Thus, government industrial policies are
necessary to help create the market demand, but also to turn an oligopolistic market to
more efficient means (Oqubay 2015: 113‒23). Cement production in SSA countries tends
to be dominated by a small number of multinational corporations. South Africa, Nigeria,
and Ethiopia are the three largest cement-producing countries in SSA, in that order, and
notably their sectors contain local firms. This section compares (p. 849) experiences of in
dustrial policy to promote the cement industry in Nigeria and Ethiopia.2
In both countries local cement production increased in the 2000s and domestic demand
was met entirely by domestic production by 2012 in Ethiopia and 2018 in Nigeria
(Akinyoade and Uche, 2018; Oqubay, 2015). In both cases, government interventions
Page 7 of 26
played a key role: directly through industrial policy and indirectly through increasing do
mestic demand. Local cement firms, which dominate the cement sectors in these coun
tries, were developed through a combination of learning rents and oligopolistic rents. Ce
ment production in both countries uses almost entirely locally sourced inputs such as
minerals from quarries and packaging. Yet, domestic demand remained too small to sus
tain more than a handful of firms, leaving them with excess production capacity.
Both Nigeria and Ethiopia had state-owned cement firms as a result of state-directed im
port-substitution industrial policies between the 1960s and the 1980s, and they had regu
lations on imported cement ranging from outright bans to import licensing schemes in or
der to conserve foreign exchange. The expansion of the cement industry in both countries
was driven by just one or two local firms as ‘first movers’. However, the role of the state
differed significantly, as a result of the political settlement, which had important implica
tions for some aspects of cement industry development in each country.3
In Nigeria, the key government policies included the privatization of state-owned cement
factories, which were bought by local and foreign investors, and the backward integra
tion policy of 2002 (Akinyoade and Uche, 2018). Together, they effected a shift from im
porting cement to domestic production. The backward integration policy banned import
ed bagged cement and involved a scheme for phasing out bulk cement imports by allow
ing investors in local cement production to continue importing bulk cement as a means to
finance their investments in cement factories. New import licences linked the volume that
could be imported in proportion to the demonstrated investment in local plant production
capacity. Dangote, a private firm, was the only local firm initially to take advantage of this
incentive, and became the first mover. In contrast, in Ethiopia the first movers were a
state-owned firm and a greenfield investment by a firm linked to the TPLF, the dominant
party within the EPRDF ruling coalition at the time (Oqubay, 2015).
The backward integration policy in Nigeria also included several fiscal incentives. Firms
investing in cement factories were given ‘pioneer’ status, which meant a five-year tax hol
iday, and paid smaller duties on equipment for mining quarries and making cement, as
well as no VAT on equipment and reduced VAT on factory construction (p. 850) materials
not available in Nigeria. The industrial policies did not include financial incentives such
as access to cheap long-term loans for investing in expensive cement factories where the
return on investment is only realized over the medium to long term. Dangote was the first
mover among local firms because they already held a licence to import cement through
political connections and thus could use capital accumulated from oligopoly rents in the
regulated import trade to purchase a state-owned cement factory in 2000. They could al
so use their political connections to reduce the risk by ensuring policies necessary to the
expansion of the domestic market would be put in place and enforced (Itaman and Wolf,
2019). As a result of these industrial policies, there were four cement-producing firms in
Nigeria by 2015, including two local and two foreign firms. Dangote controlled 65 per
cent of the domestic cement market, its closest competitor being Lafarge (a French multi
national) with 30 per cent.
Page 8 of 26
In Ethiopia, the government started promoting the cement industry in 1992 when it
opened the sector to foreign direct investment and offered a standard package of fiscal
incentives including a three-year tax holiday, duty-free equipment imports, and income-
tax exemption for expatriates for two years. It also facilitated access to factory land and
quarries for minerals at low prices, and provided low-cost electricity (Oqubay, 2015). The
state-owned development bank offered investment loans covering 70 per cent of the in
vestment over a fifteen-year period at 5 per cent interest, and cement factories had pref
erential access to foreign currency (in the context of government capital controls). Be
tween 1992 and 2002, there were only two firms manufacturing cement in Ethiopia, the
state-owned firm and the party-owned firm mentioned above. However, between 2003
and 2012, several additional local investments were made. The larger investments includ
ed a local business group that acquired a state-owned factory and invested in upgrading
and expanding it; a greenfield investment by former managers of the state-owned cement
factory that became the first mover; and several factories that emerged from joint ven
tures between Ethiopian and Chinese firms. By 2012, there were sixteen firms, including
mini cement plants as well as medium and large ones. The mini cement plants began op
erating in 2005, but they had insignificant influence in the industry due to their small
market shares. The state-owned and party-owned cement factories had a de facto monop
oly until 2011, when three large cement factories became operational, shifting it to an oli
gopolistic market structure. Notably, local firms accounted for half of total installed ca
pacity in Ethiopia.
These industrial policies targeting the cement sector were necessary, but not sufficient,
for local investments; expanding domestic market demand was key. In both countries,
government spending on infrastructure projects accounted for over 50 per cent of the de
mand and up to two-thirds in Ethiopia. Thus, the government was the main buyer. In turn,
government infrastructure projects depended significantly on Chinese financing and Chi
nese contractors. Itaman and Wolf (2019) note that the average annual value of construc
tion projects completed by Chinese firms in sub-Saharan Africa between 2010 and 2016
was the highest in Angola, followed by Ethiopia and then Nigeria. In Ethiopia, the govern
ment had an explicit strategy of using investments in public housing, roads, and universi
ties as an industrial policy tool to (p. 851) increase domestic construction and thus stimu
late demand for the manufacturing of cement as well as other building materials such as
glass and steel (Oqubay, 2015). The cement industries in both countries ran into domestic
demand constraints as the governments reduced public investments, in the context of
lower global oil prices in Nigeria and unsustainable external debt from Chinese financing
in Ethiopia.
In Nigeria, Dangote began exporting to the region in 2013 as well as opening cement fac
tories in other African countries in response to saturated domestic demand. Itaman and
Wolf (2019) calculate that Dangote Cement had profit margins of 50‒80 per cent during
2010 to 2016, due to the continued import tariffs and low tax rates, as the company con
tinued to benefit from its pioneer status with a tax exemption on export sales. Dangote al
so made investments in economies of scale and scope through an explicit strategy to be
come the market leader and sustain high profit margins. Due to the political connections
Page 9 of 26
of Dangote’s owner with subsequent ruling coalitions in Nigeria, Dangote continues to

pay very little tax, and as Itaman and Wolf (2019) show, the firm has not significantly
raised the wages of its workers despite productivity gains. Thus, little of the value cap
tured in the cement industry has been redistributed to workers or to other parts of the
population through government expenditure based on taxes, which would have had multi
plier effects in terms of increasing domestic demand for cement. In Ethiopia, in response
to the excess supply of cement, the government banned new FDI in the sector and re
duced incentives for local firms. However, it has not pursued industrial policies to encour
age cement exports or to stimulate domestic market demand. Oqubay (2015) notes that
local firms need to increase quality and productivity in order to be competitive in regional
exports. The price of cement is lower in Ethiopia than in Nigeria and expected to fall fur
ther as more investments come on stream. There are more large firms in Ethiopia and no
single firm wields as much influence over the market, and thus pricing, as Dangote does
in Nigeria.
29.5 Industrial Policy in Labour-intensive Ex

port Manufactures: Apparel Exports in Ethiopia
The apparel export sector has been a first step for most countries in the early stages of
industrialization. The globalized apparel industry became a buyer-driven GVC character
ized by decentralized, globally dispersed production networks coordinated by lead firms
who control higher-value activities such as design, branding, and marketing (due to high
er entry barriers), but often outsource all or most of the manufacturing process to a glob
al network of suppliers. Buyers have tended to increase the manufacturing and non-man
ufacturing functions they require from their suppliers. Furthermore, the asymmetric pow
er between global buyers and an increasing number of apparel (p. 852) suppliers has led
to a ‘squeeze’ on supplier firms, which are facing low prices and stringent buyer require
ments (Gibbon and Ponte, 2005).
Madagascar is the number one apparel exporter in SSA, overtaking Mauritius in 2017,
followed by Lesotho, Kenya, Swaziland, and then Ethiopia, but Ethiopia is the new rising
star with impressive growth rates. Apparel exports rose in Ethiopia from less than US$1
million in 2001 to US$13 million in 2010 and US$117 million in 2017. In contrast to the
other SSA apparel exporters, only the Ethiopian government has pursued industrial poli
cy targeting the apparel sector. The Ethiopian government’s apparel sector-specific indus
trial policies have evolved, importantly, through trial and error and can be grouped into
three largely chronological waves of industrial policy (Staritz and Whitfield, 2019).4
Ethiopia had a basic national supply chain from cotton to textiles to apparel for the do
mestic market dominated by state-owned firms (Staritz and Whitfield, 2017). In the early
to mid-2000s, the government privatized the state-owned firms and offered incentives for
local firms to invest in the apparel export sector, as part of the first wave of industrial pol
icy. These incentives included concessional credit from the state-owned Development
Bank of Ethiopia, income-tax exemptions, access to land or industry sheds at concessional
Page 10 of 26
rates, as well as general export incentives such as duty-free imported inputs, export cred
it guarantees, and partial retention of foreign exchange (in a context of capital controls)
as incentives for local and foreign investors in apparel exports. However, there were no
performance targets attached to these incentives, and the government created general in
centives to encourage investment in manufacturing whether for export or the domestic
market. Thus, the incentives to export were only marginal additions to the general bene
fits offered to manufacturers (Gebreyesus and Demile, 2017). Only fifteen local investors
established new apparel factories between 2005 and 2007. These pioneer firms struggled
to export and only a few other local investors had entered the apparel export sector by
2010.
As part of a second wave of industrial policy, the government created the Textile Industry
Development Institute (TIDI) to support local firms. TIDI offered free benchmarking stud
ies that came with foreign experts providing advice on how to improve production
processes and training for newly hired sewing-machine operators, and it helped to direct
global buyers to local firms. Again, these industrial policy rents had very weak or no per
formance standards attached to them, which was also the case with the donor agency
programmes that provided local firms with foreign experts, attendance at trade fairs, and
grants for buying equipment through matching grant schemes.
(p. 853) The local firms struggled to learn, and by 2016, many local firms had exited the
apparel sector or shifted to producing only for the domestic market, while those that con
tinued with apparel exports were operating at a loss or just breaking even. Local firms
that survived in apparel exports were part of diversified business groups, which meant
that their other businesses subsidized the cost of learning and made up for low profit
margins, while the export industry provided the foreign exchange required for the other
businesses. In general, the local apparel firms put limited effort into building their capa
bilities in order to become competitive in the export market, but their decisions on how
much effort to put in and the result of that effort were shaped by external factors linked
to the structural country context and GVC dynamics that created disincentives for invest
ing in learning.
Local investors had limited knowledge about the apparel global value chain and what was
required to be competitive. Furthermore, there was a large gap between their initial ca
pabilities and the capabilities required to enter and remain competitive in apparel assem
bly: ‘cut-make-trim’ (CMT), the lowest node in the apparel GVC. Thus, the local firm own
ers underestimated the time, resources, and effort they would have to put into this sector.
At the same time, apparel exports on a CMT basis had very low profit margins, due to the
high level of global competition, which created a limited ‘prize’ for investors willing to
take the risk, especially compared to production for the domestic market or other eco
nomic opportunities. Furthermore, the bargaining power of supplier firms in terms of
meeting cost, quality, time, and flexibility requirements of buyers was very limited, which
created a narrow ‘margin for error’ and led to loss of orders when the buyers’ terms were
not fulfilled, further driving up the risk and costs. Production based on full package,
where suppliers are responsible for sourcing all inputs and delivery of the product, brings
Page 11 of 26
higher unit prices but also requires higher merchandizing, financing, and logistics capa
bilities and comes with additional risks. Moreover, local firms faced challenges outside
the firm that made it difficult for them to develop capabilities, which reduced their incen
tives to invest in learning given that these external factors would undermine the results.
The external factors included costly and slow transport of imported and exported goods
through the Djibouti port, limited inputs available locally, scarcity of foreign exchange in
the country and high labour turnover.
A high level of sticks and carrots would be required to incentivize local investments and
compel local firms to invest in learning. However, the industrial policy rents for apparel
exporters were not significant compared to broad investment incentives for manufactur
ing, which did not distinguish between exporters and non-exporters (Gebreyesus and De
mile, 2017). The limited additional export-related rents contained in the industrial policy
measures were ex ante and came with weak compulsions to invest in learning as no clear
performance targets were attached to them. TIDI required that all exporting firms submit
annual export plans, but staff had no authority to take action against firms that failed to
meet their targets. Furthermore, TIDI staff initially had limited knowledge of the apparel
export sector and thus lacked the ability to monitor local firms’ performance and under
stand their challenges. Research on (p. 854) industrial policy and local firms in Ethiopia’s
floriculture sector comes up with similar findings (Melese, 2017, 2019).
Furthermore, the government’s other industrial policy measures to protect domestic ap
parel production had contradictory effects on promoting exports.5 The new apparel firms
used profits from the domestic market to subsidize learning in the export sector, but the
protected domestic market also undermined incentives to invest in learning. Firms that
originally only exported started diversifying into the domestic market, while other firms
always straddled domestic and export markets or moved from the domestic market into
exports to get access to foreign exchange but remained focused on their domestic market
business. Most of the local firms that were producing for the domestic market decided
not to even try exporting given the higher profits and lower requirements of the domestic
market. There were forty-nine local textile and apparel firms in 2017, and only fourteen
exported part or all of their production.
The third wave of industrial policy involved the creation of apparel-specific industrial
parks and a more targeted promotion of foreign direct investment involving linkages with
global buyers who in turn encouraged their first-tier suppliers to invest in Ethiopia. The
industrial parks were crucial for attracting these first-tier foreign apparel firms, as they
made investing in a ‘greenfield’ apparel supplier country easier. Attracting big foreign ap
parel firms was pursued as a way to quickly increase exports and employment, but also as
a means for local firms to leverage access to global markets, technology, and skills. Thus,
the government reserved some units in the industrial parks for local firms, which were to
receive access to investment loans from state-owned banks, subsidies for worker training
and hiring expatriate managers, and support linking to buyers. However, by mid-2019,
there were only three local investors with premises in Hawassa Industrial Park, the first
park to be completed—two apparel firms and one input supplier—but they had not yet
Page 12 of 26
started exporting apparel by mid-2019. This industrial policy approach to developing an

apparel export sector and supporting local firms may turn out differently, because the in
dustrial parks provide internationally compliant factory premises and close proximity to
internationally competitive foreign firms. Furthermore, industrial policy around the parks
was implemented by the Ethiopian Investment Commission, where government officials
have a greater understanding of the apparel export sector and greater enforcement capa
bilities. Local firms that invest in the parks also face more compulsion to invest in learn
ing than was the case under the previous waves of industrial policy, as they must export
100 per cent of their production. However, the outcome depends on how many local firms
invest in the parks as well as the implementation of policy measures intended to provide
more direct support to local firms, not only by subsidizing the cost of foreign experts and
training workers, but also supporting linkages with global buyers and foreign firms.
The Ethiopia apparel export case shows that a core focus of industrial policy to
(p. 855)
support local firms in new export sectors has to be supporting and compelling them to in
vest in learning. This must include measures that subsidize the learning costs, especially
given the ‘small’ prize but high level of efficiency and capabilities required to enter the
lowest node of global value chains (where competition is highest). This kind of industrial
policy to support learning is also very difficult, because it requires technical knowledge to
support learning, coupled with the political ability to compel it with reciprocal control
mechanisms. TIDI had neither the technical knowledge nor the political authority, and it
remains to be seen how the Ethiopia Investment Commission will support more local
firms in the industrial parks and their learning process. Nevertheless, the Ethiopian gov
ernment has demonstrated learning in the evolution of its industrial policy targeting the
apparel export sector, and it is the only SSA government to pursue industrial policies to
support manufactured exports that go beyond creating general export-processing zones
with their standard package of fiscal and financial incentives (see Whitfield and Staritz,
2020).
29.6 Leapfrogging Manufacturing? Knowledge-

based Services in Kenya and Rwanda
The governments of Kenya and Rwanda pursued development strategies in the 2000s that
emphasized ‘knowledge-based services’ as the key economic driver, with a relative de-em
phasis on manufacturing. In Rwanda, this included its Vision 2020, adopted in 2000,
which highlighted finance, ICT, and tourism as priority sectors to drive growth (Govern
ment of Rwanda, 2000). In Kenya, the government’s efforts have been more targeted, sin
gling out the business-process outsourcing (BPO) sector in its Vision 2030 as the key ser
vice sector targeted to drive growth. This sector was expected to be export oriented, gen
erating jobs for young professionals, and driven by investments from Kenyan firms (Mann
and Graham, 2016). This section assesses the experiences of Kenya and Rwanda with pro
moting services, especially export-oriented services and services like tourism that gener
ate foreign exchange, rather than manufacturing.
Page 13 of 26
Rwanda staged a substantial economic recovery after the 1994 genocide with GDP
growth averaging 8.2 per cent between 1996 and 2018, driven chiefly by a combination of
public investment, agricultural productivity upgrading, and promotion of tourism
(Ggombe and Newfarmer, 2019). Rwanda made substantial gains through its industrial
policy for the tourism sector, which involved public investment in the national airline and
airport infrastructure, international marketing, skills development, and investment incen
tives (UNECA, 2016). Tourism became the largest contributor to foreign exchange earn
ings (23 per cent) in 2016 (Ggombe and Newfarmer, 2019). It is also a (p. 856) significant
employer, accounting for 6.4 per cent of total employment in 2012 (UNECA, 2016). How
ever, the tourism sector remains heavily dependent on a specific segment of the market
which can only be expanded to a limited extent: visits to view its rare mountain gorillas
(Government of Rwanda, 2011). Thus, the government also promoted the meetings, incen
tives, conferences, and events (MICE) sector as part of its tourism strategy, aiming to at
tract business tourism to conferences and events. The centrepiece of state support was
the construction of the Kigali Convention Centre and concurrent investments in hotels in
and outside the capital, complemented by the expansion of state-owned Rwandair’s fleet
of aircraft (Behuria and Goodfellow, 2019).
As part of efforts to foster knowledge-based services, the government also targeted the fi
nance, insurance, and real estate (FIRE) sector. The main policy measures to promote in
vestments in this sector included financial liberalization aimed at fostering the entry of
foreign banks and selling shares in struggling state-owned commercial banks. The num
ber of commercial banks increased significantly, directing a large proportion of invest
ment finance towards the construction sector. While this partly reflects the building of ho
tels linked to the tradable tourism and MICE sectors, it also includes construction of high-
end real estate for relatively affluent domestic purchasers. The government also invested
directly, with other groups, in large real-estate projects. Private and public investments in
the MICE and FIRE sectors have focused on high-end, luxury international markets, but
there was evidence of an oversupply in the high-end market and a chronic housing short
age for the mass market (Behuria and Goodfellow, 2019).
Lastly, the Rwandan government also promoted the information communication and tech
nology (ICT) sector through investment in 4G fibre-optic and mobile broadband infra
structure and the extension of ‘e-government’ services (Government of Rwanda, 2000).
Substantial increases in Internet penetration were recorded, with mobile broadband ac
cess at 28 per cent by 2017, as well as the digitization of a number of government ser
vices and other innovative ways to render public services, such as the Zipline programme
using drones to deliver medical supplies (World Bank, 2018). In particular, the govern
ment attempted to promote the BPO sector with two broad policies (Mann et al., 2014: 8‒
9). The first was to encourage local investors in the sector through public investment in
training. The government established the Kigali Institute of Science and Technology, sent
students abroad for training in advanced engineering, and convinced Carnegie Mellon
University to open a graduate degree programme in Kigali. It also financed a technology
innovation hub with the goal of encouraging tech start-ups. The second policy was to at
tract large multinational firms to set up regional headquarters in the country and help up
Page 14 of 26
grade the country’s infrastructure, and it was hoped that Rwandan workers and business
es would benefit through knowledge transfer. However, Mann and colleagues found
Rwanda’s BPO sector to be at an incipient stage in 2014. There was a limited amount of
business-process outsourcing for international markets; rather, firms focused on domestic
and regional markets, and local firms were usually part of a regional network of compa
nies with headquarters in Nairobi. In fact, many Rwandan-based financial companies and
call-centre and (p. 857) customer-service firms were headed by Kenyan nationals. The only
firms that were able to attract international BPO work had personal and family connec
tions abroad that allowed them to compete for foreign clients (Mann et al., 2014: 16). In
short, socio-economic connections were necessary for local firms to access international
BPO markets. This point is taken up below when discussing Kenya’s experience with pro
moting the BPO sector.
Notwithstanding significant progress, a number of limitations are apparent in Rwanda’s

attempt to leapfrog the industrialization stage straight into modern services. The finance,
insurance, and real-estate as well as ICT service sectors are largely skill intensive and
have not been able to generate large-scale employment (Behuria, 2019; Behuria and
Goodfellow, 2019). The BPO sector has more potential to generate semi-skilled jobs, but it
has not expanded as rapidly as envisioned. Thus, claims by Ggombe and Newfarmer
(2019: 317) that this services-led strategy has ‘inverted the normal sequence of structural
change’ appear exaggerated and ignore fundamental weaknesses. For instance, these
scholars include government services, accounting for 15 per cent of exports in 2016, in
their calculations to bolster the claim that Rwanda ‘is rapidly becoming a services-export
ing economy’. However, the growth in government services predominantly reflects the
government’s peace-keeping activities in the East Africa region. Financial and ICT ser
vices each only contributed 2 per cent to export revenues in 2016 (Ggombe and New
farmer, 2019: 330). Furthermore, promoting non-tradable real-estate construction is like
ly to exacerbate the balance-of-payments constraint to growth, rather than reduce it. In
deed, balance-of-payments constraints and the need to generate broader sources of em
ployment prompted the government to change course by the late 2010s and place greater
policy emphasis on manufacturing (Behuria, 2019).
As Mann and colleagues (2014: 9) note, ‘Kenya is widely considered to be an East African
success story in the realm of ICTs because of its status as a Pan-African “ICT Hub”.’ How
ever, as these researchers show, the government’s efforts to promote international in
flows of BPO work—equivalent to service exports—have been less successful, with most
firms focusing on the domestic and regional markets. The government’s main policy mea
sures included public investment in fibre-optic Internet cables alongside private invest
ments, which were completed in 2009, together with the creation of a government agency
to promote the sector and a proactive marketing campaign highlighting Kenya as a place
for BPO work and encouraging foreign firms to invest in Kenya’s BPO sector. While the
cables were under construction, the government also subsidized the cost of satellite
bandwidth for BPO firms, as the cost was much higher than in existing BPO countries
such as India and the Philippines. However, both the government and locally owned BPO
firms, which set up in response to the government’s policies, underestimated the difficul
Page 15 of 26
ty of entering and becoming competitive in the BPO global value chain (Mann and Gra
ham, 2016). In contrast to international discourses about ICT as a flat or ‘disintermediat
ed’ economy that made it easier for developing-country firms to enter, Kenyan-owned
BPO firms discovered that the BPO global value chain was an ‘extremely personal, social
ly connected economy’ (Mann and Graham, 2016: 535). Kenyan firms did not have the
tacit knowledge, social (p. 858) networks, or trust to secure direct connections to foreign
clients, nor did they have the economies of scale and ability to correctly value and price
BPO work. A few large firms hired foreign experts to advise them on how to structure
their operations and to train their workers, but smaller firms could not do this and thus
failed to meet deadlines and to make profits. In general, the latter had to depend on sub-
contracted work from BPO firms in India or elsewhere that secured large contracts di
rectly from foreign clients in the United States and then gave Kenyan firms small parts of
it, but usually the least valuable parts. Research on small local firms in South Africa
comes to the same conclusions, showing that these local firms were dependent on sub-
contracting as they did not have the reputation to find work directly from international
clients, and thus the BPO sector was largely oriented towards the domestic market (An
war and Graham, 2019).
29.7 Combining Export-oriented and Import-re

placement Policies: Automobiles and Apparel
in South Africa
Apartheid industrial policies bequeathed South Africa an industrial structure built around
a mineral-energy-complex (MEC) dependent on mining and heavy industry exports, within
a highly concentrated conglomerate structure that failed to develop a more diversified
manufacturing base (Fine and Rustomjee, 1996; Freund 2019). In the post-apartheid peri
od, trade liberalization and other orthodox economic policies facilitated corporate re
structuring that prompted rapid de-industrialization, with the share of manufacturing in
GDP falling from 23 per cent in 1990 to 14 per cent in 2007 (World Bank, n.d.). Over this
period there were only limited pockets of industrial policy interventions. This section
evaluates the record of industrial policy in the only two sectors outside heavy industry
that have benefited from sector-specific industrial policies: automobiles and clothing/tex
tiles. It also briefly assesses government efforts since 2007 to mobilize a more coherent
overarching industrial policy, as well as constraints on its implementation.
Emerging from the apartheid period, the automobile sector was focused on the domestic
market and was characterized by a proliferation of the range of vehicles produced, with
limited economies of scale. In 1995, the Motor Industry Development Programme aimed
to consolidate the industry to produce a narrower range of vehicles at considerably
greater scale and focused on supplying the export market, led by foreign firms that were
already transnational producers in the automobile global value chain as original equip
ment manufacturers (OEMs). OEMs are the global system integrators that control the de
sign, assembly, and marketing channels of relatively (p. 859) sophisticated multi-compo
Page 16 of 26
nent manufactures such as vehicles. Within a framework of declining tariffs, the Pro
gramme allowed OEMs and component suppliers to earn import rebate credit certificates
when they exported vehicles and components. OEMs could use these import rebate credit
certificates to import vehicles that they did not produce in South Africa or components
for their own production, or to sell on the certificates. As tariffs on both vehicles and com
ponents were lowered, more exports were required to secure a given value of import re
bate credit certificates. This system acted as an effective reciprocal control mechanism to
compel assemblers to double production to around 600,000 vehicles per year and to raise
the share of vehicles exported from under 5 per cent in 1995 to over 50 per cent since
2007 (Barnes and Black, 2013). Through the Programme, automobiles became the lead
ing manufacturing export outside heavy industry, contributing 14 per cent of total exports
in 2017 (Bell et al., 2018).
However, the benefits of the programme relative to its costs have been contested (e.g.
Flatters, 2005), and some weaknesses are apparent. Chief among the weaknesses are the
relatively low local content at 40 per cent, large-scale imports of components and a corre
spondingly large trade deficit in vehicles and components, little employment growth, and
limited participation by black-owned component suppliers, even at the second- and third-
tier levels (Barnes et al., 2018). In 2012, the Automotive Production and Development
Programme (APDP) replaced the Motor Industry Development Programme, with the ob
jective of addressing these weaknesses, particularly by raising volumes and local content.
Rather than being entirely reliant on import credits (which fuelled the trade deficit), the
new programme supported fixed investment by OEMs and component suppliers through
an on-budget grant, while import rebate credit certificates were earned for production
rather than exports. While production volumes have increased since the introduction of
the APDP, local content has remained at approximately 40 per cent, with limited develop
ment of second- and third-tier component suppliers, particularly black-owned component
suppliers. The limited gains under the APDP reflect in large part the dominance of
transnational OEMs in securing a programme design even more financially favourable to
them, relative to component suppliers. OEM bargaining power was bolstered by the fact
that the APDP was concluded just as the shockwaves of the global financial crisis caused
a slump in automobile demand, with US OEMs in particular financial distress. Limited
employment growth and perceived lack of progress by the automobile industry in promot
ing black economic empowerment since the introduction of the APDP has also tempered
political support for the industry. The government’s Automotive Masterplan 2035 seeks to
address these weaknesses, inter alia by increasing local content to 60 per cent, doubling
employment, and developing black-owned tier-two and tier-three component suppliers.
This involves amendments to the APDP that raise direct levels of support for domestic
component manufacturing and shift support for assembly from sales to local value addi
tion (Barnes et al., 2018; NAACAM, 2019).
The clothing and textile sector provides an example of the switch from an inappropriately
designed industrial policy to a considerably more effective one. Thus, this (p. 860) sector
reflects a case of industrial policy learning, where a programme that was unable to stem
a large-scale flood of clothing imports and associated job losses was replaced by a more
Page 17 of 26
effective one, albeit after a long time lag. The South African clothing and textile industry
in the early 1990s was characterized by production almost exclusively for the domestic
market behind significant tariffs. The thinking behind the immediate post-apartheid eco
nomic strategy was that rapid trade liberalization was all that was required to compel
manufacturers producing for the domestic market to raise their competitiveness to com
pete internationally. The 1993 Duty Credit Certificate Scheme was meant to provide a
boost to exporting firms, along similar lines to the MIDP. Exporting apparel firms could
earn Duty Credit Certificates (DCCs) that they could then sell, typically to retailers who
used them to lower the costs of their clothing imports. However, the scheme led to about
5 per cent of apparel firms exporting, which also fuelled imports as domestic retailers
used DCCs to realize even lower effective tariffs than the applied rate. Furthermore, the
applied tariff of 40 per cent on clothing was dramatically undercut by widespread under-
invoicing and customs mis-declaration (NEDLAC, 2009).
An additional weakness was that the industrial policy for the clothing and textile sector,
which consisted almost entirely of the DCC policy instrument, did not take into considera
tion the looming expiry in 2005 of the WTO Agreement on Textiles and Clothing which im
posed quotas on developing-country exports to industrialized economies. Upon its expiry
and China’s entry into the World Trade Organization, Chinese exports of clothing and tex
tile products surged to 40 per cent of world trade and rapidly became the largest source
of imported clothing in South Africa. Thus, the Duty Credit Certificate Scheme ultimately
failed. It contributed to the flood of imports and associated job losses, while doing little to
support sustainable exports. It effectively operated as a substantial subsidy to the small
group of exporters that were largely unable to compete without it.
When the Duty Credit Certificate Scheme expired in 2009, it was replaced with an on-
budget support programme called the Clothing Textiles Competitiveness Programme
(CTCP), similar to the automobile sector. This programme allowed manufacturers to earn
a value-added-based production incentive in the form of credits that could be redeemed
only through investments in specific competitiveness and upgrading activities. A number
of retailers supported the programme and sought to help revive their domestic supply
base, motivated both by more competitive and responsive domestic manufacturers and
rising Chinese wages. The Programme has broadly stabilized employment in the sector,
while supporting the highest levels of labour-productivity growth within the overall manu
facturing sector in South Africa.
The government’s clothing and textile industrial policies have been contested by acade
mics. A major theme is that tariff reductions should have been accompanied by labour-
market deregulation to allow wages to fall (Kaplan, 2015; Nattrass, 2014). However, these
arguments fail to engage with important empirical realities as well as political economy
impediments to a labour-deregulation path. First, as measured by the World Bank, South
Africa has an intermediate level of labour-market flexibility (p. 861) relative to peer mid
dle-income economies, which cannot fully account for massive job losses in clothing and
South Africa’s overall extreme unemployment levels (DTI, 2019; Zalk, 2017). Second,
these arguments do not engage with the political feasibility of labour-market deregulation
Page 18 of 26
in the context of the strong alliance between the ANC government and COSATU, the
largest trade union federation.
It was only after South Africa had experienced profound de-industrialization that the Na
tional Industrial Policy Framework (NIPF) was adopted in 2007, followed by associated
annual Industrial Policy Action Plans (IPAPs) setting out strategies for other sectors as
well as overhauling the automobile and clothing and textile strategies, as described
above. However, notwithstanding formal adoption of the NIPF, industrial policy has been
undermined in practice by a number of factors (DTI, 2019; Zalk, 2014, 2017). Large pri
vate business groups have focused on sectors in which they have been able to achieve
high levels of profitability, predominantly in non-tradable service sectors and capital-in
tensive tradable sectors over which market dominance could be asserted, but with limit
ed economy-wide linkages and scope for structural transformation. With high financial re
turns available in these sectors, capital has not flowed to diversified manufacturing indus
tries that are subject to higher levels of competition from imports due to trade liberaliza
tion. The misalignment of macroeconomic policy with structural transformation, including
pervasive currency overvaluation and falling real budget allocations for industrial financ
ing, undermines efforts to promote manufactured exports and creates further disincen
tives for firms to invest in manufacturing. Increasing patronage and corruption since
2008, itself arguably a consequence of the failure to develop a more productive and em
ployment-generating economy, has further undermined conditions for investment in diver
sified manufacturing. Large-scale corruption at strategic state-owned enterprises Eskom
and Transnet has fundamentally undermined their formal mandate to provide reliable and
cost-effective electricity, and rail and ports services.
29.8 Conclusion
While we agree to some extent with Milberg and colleagues (2014) on the need for verti
cal specialization and industrial policies tailored towards global value chains, they under
estimate the challenges for local firms in SSA countries to enter even the lowest function
in global value chains, as the cases of apparel exports in Ethiopia and business-process
outsourcing in Kenya and Rwanda show. SSA countries need industrial policies targeting
local firms to help them build their capabilities even at the lower end of value chain activ
ities. Such industrial policies involve not only creating and monitoring learning rents, but
also bringing foreign knowledge into the country through various means such as foreign
direct investment, joint ventures between foreign and local firms, or other schemes
through which local firms can access foreign tacit knowledge.
The case of the cement industry in Nigeria and Ethiopia illustrates that the con
(p. 862)
straint of domestic demand is real, and as a result, domestic markets tend towards oli
gopolistic structures, even more so in capital-intensive sectors where there are higher
barriers to entry and large first-mover advantages. Yet Nigeria and Ethiopia are some of
the most populous and thus largest economies in sub-Saharan Africa. Without expanding
purchasing power through manufactured exports (which are minimal in Nigeria) or redis
Page 19 of 26
tribution of wealth, domestic markets cannot support competitive intermediate and heavy
industries that do not export. Thus, domestic market and export-oriented manufacturing
that depends on sources of demand external to national economies have to go hand in
hand.
The cases of business-process outsourcing in Kenya and Rwanda show that global value
chains in knowledge-based services operate just as manufacturing global value chains do.
Entering requires much more than just low wages to become profitable. In particular, it
requires tacit knowledge that can only be acquired outside the country from people who
have experience working in the global industry, and it requires contact and trust to be es
tablished with clients across distance, which is not easy. Sub-contracting can be an easier
way to enter export markets, but it comes with very low prices and thus firms cannot stay
in that node of the global value chain for very long. Furthermore, as Kenyan firms and the
government came to realize, the poor performance of some local firms can affect the
overall image and reputation of the country as a supplier country. As a result, local BPO
firms in Kenya realized that they needed to build up knowledge and experience by supply
ing the domestic and regional markets first, before trying again in the global value chain
(Mann and Graham, 2016). The government then shifted its policy to focus on promoting
higher-value IT-enabled services for the domestic market, and on attracting big multina
tional firms in order to build Kenya’s reputation in the eyes of foreign clients. Mann and
Graham (2016: 543) conclude that the synergy between the BPO sector and local soft
ware developers could allow Kenya to develop a longer-term potential for higher-value
BPO. This strategy resembles what Coe and Yeung (2015) discuss as coupling, de-cou
pling, and re-coupling with global value chains in order to build capabilities but also even
tually capture great value from participation in global value chains. Thus, export-oriented
and domestic market strategies have to be combined.
Industrial policies in the South African automotive and apparel sectors have achieved
some successes in raising firm capabilities, albeit with limitations. However, a confluence
of economy-wide factors has prevented the reversal of de-industrialization. Macroeco
nomic and financial-sector policies have hindered rather than supported structural trans
formation and industrial policy. Large private business groups have restructured and
shifted capital to sectors over which they can exert market dominance, particularly non-
tradable service sectors, allowing for high profitability with limited commitment of fixed
investment, particularly in diversified manufacturing sectors. Low fixed investment cou
pled with high unemployment and inequality has contributed to conditions for patronage
networks and corruption to flourish, further impeding the accumulation of productive ca
pabilities.
This analysis implies that SSA countries need to draw upon a wide range of indus
(p. 863)
trial policy instruments. These include mobilizing investment in traditional and modern
infrastructure; deepening intermediate inputs including cement, metals, and chemicals;
exploring ways to stimulate domestic demand through government expenditure; forging
stronger linkages between agriculture and manufacturing; promoting a range of indus
Page 20 of 26
tries that can concurrently generate sustained domestic demand and relieve the balance-
of-payments constraint through exports.
References
African Development Bank Group (ADB) (2019) African Economic Outlook 2019. Abidjan.
Akinyoade, Akinyinka and Chibuike Uche (2018) ‘Development Built on Crony Capitalism?
The Case of Dangote Cement’, Business History 60(6): 833–58.
Amsden, Alice (2001) The Rise of ‘the Rest’: Challenges to the West from Late Industrial
izing Economies. Oxford: Oxford University Press.
Amsden, Alice (2008) ‘The Wild Ones: Industrial Policies in the Developing World’, in Nar
cis Serra and Joseph E. Stiglitz (eds) The Washington Consensus Reconsidered: Towards a
New Global Governance. Oxford: Oxford University Press, pp. 95–118.
Anwar, Mohammad Amir and Mark Graham (2019) ‘Does Economic Upgrading Lead to
Social Upgrading in Contact Centers? Evidence from South Africa’, African Geographical
Review 38(3): 209–26.
Arbache, Jorge Saba and John Page (2009) ‘How Fragile is Africa’s Recent Growth?’, Jour
nal of African Economies 19(1): 1–24.
Barnes, Justin and Anthony Black (2013) ‘The Motor Industry Development Programme
1995–2012: What Have We Learned?’, in International Conference on Manufacturing-led
Growth for Employment and Equality, pp. 1–38.
Barnes, Justin, Anthony Black, Douglas Comrie, and Tamryn Hartogh (2018) ‘Geared for
Growth: South Africa’s Automotive Industry Masterplan to 2035’. Available at https://
www.thedti.gov.za/news2019/SA_Auto_Masterplan.pdf.
Behuria, Pritish (2019) ‘The Domestic Political Economy of Upgrading in Global Value
Chains: How Politics Shapes Pathways for Upgrading in Rwanda’s Coffee Sector’, Review
Behuria, Pritish and Tom Goodfellow (2019) ‘Leapfrogging Manufacturing? Rwanda’s At
tempt to Build a Services-led “Developmental State”’, The European Journal of Develop
ment Research 31(3): 581–603.
Bell, Jason, Sumayya Goga, Pamela Mondliwa, and Simon Roberts (2018) ‘Structural
Transformation in South Africa: Moving towards a Smart, Open Economy for All’. Avail
able at https://static1.squarespace.com/static/52246331e4b0a46e5f1b8ce5/t/
Page 21 of 26
5be17018575d1f0f5f3333f0/1541501005773/
CCRED+Working+Paper+9_2018_Structural+Transformation_BellGogaMondliwaRoberts.pd
.
Coe, Neil and Henry Wai-chung Yeung (2015) Global Production Networks: Theorizing
Economic Development in an Interconnected World. Oxford: Oxford University Press.
(p. 864)
Di Maio, Michele (2009) ‘Industrial Policies in Developing Countries: History and Perspec
tives’, in Mario Cimoli, Giovanni Dosi, and Joseph E. Stiglitz (eds) Industrial Policy and
Development: The Political Economy of Capabilities Accumulation. Oxford: Oxford Univer
DTI (2019) ‘Concept Note: Placing Structural Transformation at the Centre of Economic
Revival under the New Dawn’, Department of Trade and Industry. Available at https://
gallery.mailchimp.com/269d03e82d31aba2461044c20/files/
3fa4ea4d-5be1-4b79-934d-fcb57d0216ed/
Concept_Note_Structural_Transformation_under_the_New_Dawn_DraftV5.pdf.
Farooki, Masuma and Raphael Kaplinsky (2012) The Impact of China on Global Commodi
ty Prices: The Global Reshaping of the Resource Sector. London: Routledge.
Minerals-energy Complex to Industrialisation. London: C. Hurst & Co.
Flatters, Frank (2005) ‘The Economics of MIDP and the South African Motor Industry’.
Presented at the TIPS/NEDLAC South Africa Trade and Poverty Programme (SATPP) Poli
cy Dialogue Workshop, November 2. Johannesburg: Queens University, Canada.
Freund, Bill (2019) Twentieth-century South Africa: A Developmental History. Cambridge:

Cambridge University Press.
Gebreyesus, Mulu and Ashagrie Demile (2017) ‘Why Export Promotion Efforts Failed to
Deliver? Assessment of the Export Incentives and their Implementation in Ethiopia’. EDRI
Working Paper No. 17. Addis Ababa: Ethiopian Development Research Institute.
Ggombe, Kasim Munyegera and Richard S. Newfarmer (2019) ‘Rwanda: From Devasta
tion to Services-first Transformation’, in Richard S. Newfarmer, John Page, and Finn Tarp
(eds) Industries without Smokestacks: Industrialization in Africa Reconsidered. Oxford:
Gibbon, Peter and Stefano Ponte (2005) Trading Down: Africa, Value Chains and the Glob
al Economy. Philadelphia, PA: Temple Press University.
Page 22 of 26
Government of Rwanda (2000) Rwanda: Vision 2020. Kigali: Ministry of Finance and Eco
nomic Planning.
Government of Rwanda (2011) Rwanda National Export Strategy. Kigali: Ministry of

Trade and Industry.
Itaman, Richard and Christina Wolf (2019) ‘Industrial Policy and Monopoly Capitalism in
Nigeria: Lessons from the Dangote Business Conglomerate’. CAE Working Paper No.
2019:1. Roskilde University, Denmark: CAE. Available at http://www.ruc.dk/cae.
Kaplan, David (2015) ‘The Structure and Performance of Manufacturing in South Africa’,
in Wim Naudé, Adam Szirmai, and Nobuya Haraguchi (eds) Structural Change and Indus
trial Development in the BRICS. Oxford: Oxford University Press, pp. 244–64.
Kaplinsky, Raphael (2005) Globalization, Poverty and Inequality: Between a Rock and a
Hard Place. Cambridge: Polity Press.
Kennedy, Paul (1988) African Capitalism: The Struggle for Ascendency. Cambridge: Cam
Khan, Mushtaq (2009) ‘Learning, Technology Acquisition and Governance Challenges in

Developing Countries’. Available at mercury.soas.ac.uk/users/mk17.
Khan, Mushtaq (2013) ‘Political Settlements and the Design of Technology Policy’, in
Joseph E. Stiglitz, Justin Yifu Lin, and Ebrahim Patel (eds) The Industrial Policy Revolu
tion II: Africa in the Twenty-first Century. Basingstoke: Palgrave Macmillan, pp. 243–80.
(p. 865)
Khan, Mushtaq (2019) ‘Knowledge, Skills and Organizational Capabilities for Structural
Transformation’, Structural Change and Economic Dynamics 48: 42–52.
Lall, Sanjaya (1992) ‘Technological Capabilities and Industrialization’, World Development

20(2): 165–86.
Lall, Sanjaya (1995) ‘Structural Adjustment and African Industry’, World Development
23(12): 2019–31.
Mann, Laura and Mark Graham (2016) ‘The Domestic Turn: Business-process Outsourc
ing and the Growing Automation of Kenyan Organisations’, Journal of Development Stud
ies 52(4): 530–48.
Mann, Laura, Mark Graham, and Nicolas Friederici (2014) ‘The Internet and Business-
process Outsourcing in East Africa: Value Chains and Networks of Connectivity-based En
terprises in Kenya and Rwanda’. Report of the Oxford Internet Institute. Available at
http://www.oii.ox.ac.uk.
Page 23 of 26
Melese, Ayelech Tiruwha (2017) ‘Ethiopian-owned Firms in the Floriculture Global Value
Chain: With What Capabilities?’, CAE Working Paper No. 2017: 2. Roskilde University,
Denmark: Center of African Economies.
Melese, Ayelech Tiruwha (2019) ‘African-owned Firms and Investment in Learning: Local
Firms in the Ethiopian Floriculture Export Sector’. PhD thesis, Department of Social
Sciences and Business, Roskilde University, Denmark.
Milberg, William, Xiao Jiang, and Gary Gereffi (2014) ‘Industrial Policy in the Era of Verti
cally Specialized Industrialization’, in Irmgard Nubler, José Manuel Salazar-Xirinachs, and
Richard Kozul-Wright (eds) Transforming Economies: Making Industrial Policy Work for
Growth, Jobs and Development. Geneva: International Labour Office, pp. 151–78.
NAACAM (2019) ‘Business Guide’. National Association of Automotive Component & Al
lied Manufacturers. Available at https://www.naacam.co.za/assets/content/
page18.pdf.
Nattrass, Nicoli (2014) ‘A South African Variety of Capitalism?’, New Political Economy
19(1): 56–78.
NEDLAC (2009) ‘Report on Customs Fraud in Respect of Textile, Clothing and Footwear’.
Trade and Industry Chamber. Available at http://nedlac.org.za/wp-content/uploads/
2017/10/NEDLAC-REPORT-ON-CUSTOMS-FRAUD-IN-RESPECT-OF-TEXTILE-
CLOTHING-AND-FOOTWEAR.pdf.
Newfarmer, Richard S., John Page, and Finn Tarp (eds) (2018) Industries without Smoke
stacks: Industrialization in Africa Reconsidered. Oxford: Oxford University Press.
versity Press.
Page, John (2012) ‘Can Africa Industrialize?’, Journal of African Economies

21(Supplement 2): ii86–125.
London: Constable.
Schwartz, Herman Mark (2010) States versus Markets: Understanding the Global Econo
my. 3rd edition. London: Palgrave Macmillan.
Startiz, Cornelia and Lindsay Whitfield (2017) ‘Made in Ethiopia: The Emergence and
Evolution of the Ethiopian Apparel Export Sector’. CAE Working Paper No. 2017:3. Cen
tre of African Economies, Roskilde University, Denmark. Available at http://www.ruc.dk/
cae.
Staritz, Cornelia and Lindsay Whitfield (2019) ‘Light Manufacturing in Ethiopia: The Case
of the Apparel Export Industry’, in Fantu Cheru, Christopher Cramer, and Arkebe Oqubay
Page 24 of 26
(eds) The Oxford Handbook of the Ethiopian Economy. Oxford: Oxford University
(p. 866)
Press, pp. 704–20.
Stewart, Francis, Sanjaya Lall, and Samuel Wangwe (eds) (1992) Alternative Develop
ment Strategies in Sub-Saharan Africa. London: Macmillan.
Sundaram, Jomo and Rudiger Von Arnim (2008) ‘Economic Liberalization and Constraints
to Development in Sub-Saharan Africa’. Working Paper No. 67. UN, Department of Eco
nomics and Social Affairs.
UNECA (United Nations Economic Commission for Africa) (2016) Transformative Indus
trial Policy for Africa. Addis Ababa: Economic Commission for Africa.
de Vries, Gaaitzen, Marcel Timmer, and Klaas de Vries (2015) ‘Structural Transformation
in Africa: Static Gains, Dynamic Losses’, The Journal of Development Studies 51(6): 674–
88.
Whitfield, Lindsay (2017) ‘An Introduction to the Political Economy of Development’, in

Nic Cheeseman, Lindsay Whitfield, and Carl Death (eds) The African Affairs Reader: Key
Texts in Politics, Development, and International Relations. Oxford: Oxford University
Press, pp. 115–28.
Whitfield, Lindsay (2018) Economies after Colonialism: Ghana and the Struggle for Power.
Whitfield, Lindsay and Cornelia Staritz (2020) ‘Industrial Hubs and Technology Transfer
in Africa’s Apparel Export Sector’, in Arkebe Oqubay and Justin Yifu Lin (eds) The Oxford
Handbook of Industrial Hubs and Economic Development. Oxford: Oxford University
Press.
Whitfield, Lindsay, Ole Therkildsen, Lars Buur, and Anne Mette Kjær (2015) The Politics
of African Industrial Policy: A Comparative Perspective. Cambridge: Cambridge Universi
ty Press.
World Bank (2018) Future Drivers of Growth in Rwanda: Innovation, Integration, Agglom
eration, and Competition. Washington, DC: World Bank Group.
World Bank (n.d.) World Development Indicators. Washington, DC: World Bank.
Zalk, Nimrod (2014) ‘Industrial Policy in a Harsh Climate: The Case of South Africa’, in
José M. Salazar-Xirinachs, Irmgard Nübler, and Richard Kozul-Wright (eds) Transforming
Economies: Making Industrial Policy Work for Growth, Jobs and Development. Geneva:
ILO, pp. 327–53.
Zalk, Nimrod (2017) ‘The Things We Lost in the Fire: The Political Economy of Post-
apartheid Restructuring of the South African Steel and Engineering Sectors’, PhD thesis,
University of London, School of Oriental and African Studies.
Page 25 of 26
Notes:
(1) For further details, see the review in Whitfield et al. (2015), and the discussion on
agribusiness and services in Newfarmer, Page, and Tarp (2018).
(2) This section draws primarily on research by Richard Itaman and Christina Wolf (2019)
on the cement industry in Nigeria, and the research of Arkebe Oqubay (2015) on the ce
ment industry in Ethiopia.
(3) On the concept of political settlement and its application to sub-Saharan Africa, see
Whitfield et al. (2015).
(4) This section draws on research carried out by Lindsay Whitfield in collaboration with
Cornelia Staritz, as part of the African-owned Firms Building Capabilities in Global Value
Chains (Africap) project funded by the Danish Council for Independent Research, Grant
number DFF–4182–00099. The research included a survey on technological capabilities
carried out with all locally owned apparel-exporting firms, and interviews with staff in rel
evant government agencies and industry associations as well as global buyers and buying
agents.
(5) The domestic market for textiles and apparel was still protected with a tariff rate of 35
per cent and an additional 10 per cent surcharge and 10 per cent excise taxes, as well as
a ban on second-hand clothing imports.
Lindsay Whitfield
Lindsay Whitfield is professor (with special responsibilities) in Global Studies and

leader of the Centre of African Economies in the Department of Social Sciences and
Business, Roskilde University, Denmark. She is author of several books on African
politics and economies, including The Politics of African Industrial Policy: A Compar
ative Perspective (Cambridge University Press, 2015) and Economies after Colonial
ism: Ghana and the Struggle for Power (Cambridge University Press, 2018).
Nimrod Zalk
Nimrod Zalk is industrial development adviser at the South African Department of

Trade and Industry (DTI). Prior to this he was deputy director-general of the Industri
al Development Division (IDD) of the DTI. He sits on the board of the South African
Industrial Development Corporation (IDC) and chairs the steering committee of the
Industrial Development Think Tank (ITT) at the University of Johannesburg.
Page 26 of 26
The Political Economy of Industrialization and Industrial Policy in Africa,
1960‒2018
The Political Economy of Industrialization and Industri

al Policy in Africa, 1960‒2018
Horman Chitonge and Peter Lawrence
This chapter examines industrial policy (IP) in Africa from late colonial times to the
present. It focuses on explanations of the failure of African countries to transcend the
first stage of import substitution and effect the structural transformation of their
economies. These explanations lie in the nature of a country’s power relations, their inter
action with foreign capital and international institutions, and their effect on the ability of
the state to implement IP. We point to the continuities from colonial IP to the post-inde
pendence adoption of more rapid import substitution, then to the effects of liberalization
through structural adjustment programmes on IP and industrialization, and finally to the
return of a more focused IP. We show the policy effects on industrial growth and econom
ic structure in each period and argue that particular policies were followed because of
the power of specific class and state coalitions to prevent or effect change.
Keywords: power, class, state, structure, capital, coalitions, liberalization, delinking
30.1 Introduction
AFRICAN countries have enjoyed relatively high rates of GDP growth over the last two
decades, largely as a result of the commodity booms that have been driven by the rapid
industrial growth of China and other Asian countries. However, this has not resulted in
the structural transformation of African economies, that is, a transition from low- to high-
productivity activities, most rapidly achieved by the growth of manufacturing industry,
but rather the opposite (McMillan and Rodrik, 2011; Weiss and Jalilian, 2016). In Africa,
industrial policy (IP), a deliberate attempt by governments to promote manufacturing in
dustries through specific economic and financial policies, began to a limited extent under
colonial rule, continued through the early years of independence, and was effectively
abandoned under structural adjustment. It is now back in favour as the central role of the
state in past industrializations has been increasingly acknowledged (Chang, 2013; Rodrik,
Page 1 of 40
1960‒2018
2008; UNECA, 2016; Weiss, 2016) and seen for Africa as ‘a necessary economic tool…to
correct major sectoral and other misallocations’ (Stiglitz et al., 2013).
This chapter examines IP in post-independence Africa through the lens of power relations
and the state, traditionally the focus of political economy. We find the origins of IP in
Africa lying in the colonial period and analyse policy before and after independence, fo
cusing on changes in the strategies adopted and the reasons why. (p. 868) We examine the
main IP of import substitution, a strategy begun under colonial rule and intensified after
independence and explain why it did not result in deeper industrialization. We then dis
cuss IP, or its absence, during the phase of economic liberalization, following that with a
discussion of current ideas together with some case examples of the return of IP in prac
tice as a consequence of the failure of liberalization to deliver the transformation that it
promised. Although we acknowledge the various organizational and technical explana
tions usually given for the failure of industrial transformation in Africa, our focus, unlike
much of the literature in this area, is on issues of the character of state power, class rela
tions, and the role of global capital and international financial institutions in determining
the trajectory of African industrial policy and industrialization.
30.2 Colonial Industrial Policy: The Birth of

Import-substitution Industrialization
Colonial economic policy was geared to ensuring that supplies of agricultural and mineral
primary products were secured for processing and consumption in the imperial countries
while the colonies provided markets for the metropolis’s manufactures (Rodney, 1972;
Chitonge, 2019). The protection of the imperial powers’ manufacturing industry by not
engaging in such production in the colonies was the accompanying arm of economic poli
cy—‘the metropolitan and colonial economies were complementary not
competitive’ (Havinden and Meredeth, 2010: 169).
There is, however, some debate about the degree to which manufacturing in Africa was
already established at independence. Although it is usual to think of colonial regimes not
having an IP, the literature on the subject suggests that by the 1940s they began to turn
attention to diversifying economic activity through industrialization (Butler, 1991).
Whether this change of direction was the result of realizing that the colonies would even
tually become independent, or whether it was to use such diversification to keep the pop
ulations content with colonial rule is debatable. Yet in spite of this apparent change in
policy, the main strategy, even by the latter part of the decade, remained to restore colo
nial infrastructure neglected during the Second World War by securing imports of such
critical products as cement and steel, such infrastructure being the result of initial colo
nial investment policy to facilitate the exports of primary products (Butler, 1991). In the
case of the French colonies, however, the shortages they endured during the Second
World War stimulated African manufacturing activity which by the end of the war was suf
ficiently well established to withstand competition from French-based companies and was
Page 2 of 40
1960‒2018
also helped by French government policy of supporting industrialization in the colonies
(Thompson and Adloff, 1975).
The British vehicle for developing industrial activity was the UK government’s
(p. 869)
Colonial Development Corporation (CDC), deemed to be more locally acceptable than pri
vate companies such as the United Africa Company (UAC). However, by this time (around
1947) the government view had changed from one supporting diversification through in
dustrialization to one arguing for specialization in agricultural exports. As a result, the in
vestment by the CDC was directed to increasing colonial exports to US-dollar-earning
destinations and producing primary products that could go to the United Kingdom; both
measures aimed to ease the United Kingdom’s chronic balance-of-payments deficits (But
ler, 1991).
The settler regimes of South Africa and Southern Rhodesia were exceptions and estab
lished significant manufacturing capabilities. There was a strong European constituency
in favour of industrialization, especially in South Africa whose main IP was to protect do
mestic manufacturing through tariffs and to subsidize electricity and steel production.
The strength of the mining industry meant that export revenues enabled the import of
goods not produced domestically while also building up the necessary infrastructure for
manufacturing to flourish (Fine and Rustomjee, 1996). In Southern Rhodesia, the govern
ment started intervening in the economy in 1942, nationalizing iron and steel production,
and taking steps to establish a textile industry. IP was based less on protective tariffs and
more on subsidizing losses, partly the result of poor management skills and a greater in
terest among white Rhodesians in trade and property (Kilby, 1975).
Other parts of Africa where there were significant but not large settler populations also
developed some manufacturing: Senegal, the Belgian Congo, and Kenya, aided to some
extent by subsidies in the case of Senegal. Angola and Mozambique, Portuguese colonies
until the early 1970s, developed some manufacturing capacity, especially in the later
years, although in the case of Mozambique, this was hampered by the proximity of an in
dustrialized South Africa (Abshire, 1969). Much of manufacturing across the continent
was in the hands of European and Asian settlers (Kilby, 1975) and colonial administra
tions assisted such enterprises, as, for example, with the formation of the Uganda Devel
opment Corporation in 1952. There were also policies to develop technical education, for
example, in Nigeria, and also to encourage import-substitution industries with some in
vestment support in collaboration with private companies such as the UAC (Pedler, 1975;
Kilby, 1975). Transfers of funds, a precursor of development aid, became much more sig
nificant after the Second World War (Munro, 1976). However, the absence of an industrial
labour force in the peasant economies and the shortage of skilled labour in the settler
economies of South Africa, especially because of its colour bar and job reservation policy,
ensured that investments in manufacturing were capital intensive (Austin, 2010).
By the end of the 1940s and the early 1950s imperial development policy saw the colonies
as a source of primary-product export revenue used to import capital goods and develop
infrastructure to attract private investment. As Havinden and Meredeth (2010) note, this
Page 3 of 40
1960‒2018
approach was problematic. Fluctuating export income from unstable commodity prices,
import expenditure on ‘luxury’ consumer goods for the (p. 870) high-income settler popu
lation, poor linkages between the export sector and the rest of the economy, and profit
outflows from the activities of foreign trading enterprises and foreign banks all made the
prospect of export-led growth and development unlikely and anticipated the problems
that would be faced by the newly independent African countries in the following decade.
Nonetheless in some colonies, development funds were made available which allowed the
beginnings of import-substitution industrialization (ISI), anticipating the industrial policy
and strategy of the post-colonial governments (Kilby, 1975).
Leaving South Africa aside, by the end of the colonial period the Belgian Congo and
Southern Rhodesia had the largest manufacturing sectors in terms of their contribution
to GDP, followed by Kenya and Senegal (Kilby, 1975). Production was centred on such
products as cement, soap, textiles, baked foods, breweries, and food processing (Austin et
al., 2016). The policy of promoting import-substituting manufacturing by largely private
metropolitan companies was supplemented variously by tariff protection, subsidies, in
dustrial training and development funds (Munro, 1976). These policies formed the basis
of post-colonial industrial strategies in Africa.
30.3 Import-substitution Industrialization in

Africa: The Early Post-colonial Experience
In spite of the low level of industrial development, its overall neglect during colonial rule
in Africa was confirmed at independence. Typically, African economies were dominated by
the export of agricultural and mineral primary products (GATT, 1966). The agricultural
sector itself, including both cash crops for export and food crops for domestic consump
tion, accounted for 60‒70 per cent of GDP. Manufacturing value added as a share of GDP
in 1963 ranged from 2 per cent to 18 per cent but for most countries it was below 10 per
cent (see Table 30.1). Capital and consumer goods, even non-durables, were mostly im
ported (Ackah et al., 2016: 48).
Page 4 of 40
The Political Economy of Industrialization and Industrial Policy in Africa, 1960‒2018
Table 30.1 Manufacturing performance during early ISI, 1960–80
Manufactur Average annu Manufactur Average annu

ing share of al growth ing share of al growth
GDP rates of MVA GDP rates of MVA
Coun 1963 1980/ 1960 1971 Coun 1963 1980/ 1960 1971
try 81 –70 –80 try 81 –70 –80
Countries with high MVA share in 1963 Countries with medium MVA share in 1963 (cont’d)
Alge 10.0 17.0 7.7 11.6 Ethio 9.6 4.4 8.2 2.8
ria pia
Burki 11.3 14.7 18.3 4.6 Gabo 7.2 4.6 10.9 14.9
na-Fa n
so
Cen 12.1 7.2 5.4 –4.3 Gam 3.0 4.9 6.1 4.2
tral bia
Africa
n Rep.
Côte 11.6 8.1 11.6 5.8 Ghan 9.7 7.8 6.9 1

d’Ivoi a
re
Page 5 of 40
Eswa 18.2 18.0 18.1 6.7 Kenya 8.7 10.9 5.6 12.2
tini
(Swaz
iland)
Mada 10.6 11.2 9 –0.4 Mali 5.1 6.5 4.0 2.0

gas
car
Mala 10.4 12.3 14.9 6.7 Moza 8.5 7.4 4.8 3.8
wi mbiqu
e
Mau 13.5 13.0 2.3 9.5 Nami 5.8 8.5 6.6 -6.1
ritius bia
Mo 15.7 18.0 8.0 5.6 Sierra 5.7 4.9 4.5 3.7
rocco Leone
Sene 12.8 13.5 4.2 –4.5 Tan 9.7 9.1 10.2 2.9
gal zania
South 20.6 20.5 8.7 4.5 Togo 9.7 7.8 14 –10.4

Africa
Sudan 14.4 6.8 5.6 1.5 Tunisi 8.0 11.8 7.6 11.7
a
Page 6 of 40
Zim 16.0 20.7 11.8 2.8 Ugan 8.2 4.3 5.3 –9.3
bab da
we
Zam 5.5 16.9 11.5 0.3

bia
Countries with medium MVA share in 1963 Countries with low MVA share in 1963
Ango 6.2 9.7 7.2 –12.8

la
Benin 7.4 8.0 6 –0.6 Guine 4.2 4.4 3.3 2.6

a
Botsw 5.0 4.4 6.2 17.3 Guine 1.3 1.4 8.4 3.4
ana a Bis
sau
Bu 9.3 6.9 13.8 5.9 Lesot 0.8 7.7 3.4 7.2
rundi ho
Came 9.9 8.8 7.4 7.3 Liberi 4.3 9.4 12.8 5.6
roon a
Cape 5.5 — 9 3.2 Libya 2.5 2.5 8.4 18.4

Verde
Page 7 of 40
Chad 6.2 11.1 5.4 5.7 Mau 4.8 4.8 9.2 4.6
rita
nia
Congo 6.8 7.5 7.4 –1.8 Niger 3.1 3.7 9.6 4.7
Con 8.0 7.6 12.5 0.6 Nige 4.4 4.7 7.6 13.4
go DR ria
Equa 5.2 5.5 5.1 –16.1 Rwan 2.1 15.3 2.2 6.3
torial da
Guine
a
Notes: ISI: import-substitution industrialization; SAPs: structural adjustment programmes; MVA share: manufactur
ing value added as a share of GDP. High = 10 per cent+; medium= >5<10 per cent; low = <5 per cent.
Sources: World Development Indicators (online); African Statistical Yearbook (various years).
Page 8 of 40
1960‒2018
Generally, ISI during the first decade of independence was implemented as part of the
broad development strategy elaborated in the national development plans (NDPs) which
almost every independent African country formulated. Countries did not develop a sepa
rate IP in which they outlined the ISI strategy; ISI was embedded in the NDPs (Chitonge,
2019). For most post-colonial African governments, the establishment of domestic indus
trial capabilities was not only seen as a means of lessening dependence on former colo
nial powers but as a way of achieving economic and social development.
ISI as a long-term strategy has been understood as a two- or three-stage process. The
first ‘easy’ phase (Hirschman, 1958) was the local production of previously imported
(p. 871) (p. 872) consumer goods, using foreign capital investment and as much of domes
tic raw materials as possible. The second phase involved learning from the experience of
industrial activity together with directed education and training to develop skills in man
agement and engineering. The second phase would increase productivity and lead to
greater competitiveness in export markets, allowing for the exploitation of economies of
scale. The third stage would involve the production of producer goods: machines and in
termediate production induced by the backward linkages from final goods output. These
stages could overlap (Hirschman, 1958).
30.4 The Political and Economic Objectives of

Import-substitution Industrialization in Africa
African leaders saw the growth of the industrial sector as the quickest way to catch up
with the developed countries and reduce their dependence on former colonial rulers.
Demonstrating their independence was expressed concretely by not embracing capital
ism, which was a system closely associated with colonial powers. The perception of capi
talism as the system of the colonisers was clearly articulated by Leopold Senghor, the
president of Senegal, who argued that ‘there is naturally, in the mind of proletarian na
tions, complete identification between capitalism and colonialism, between the political
system and the economic system’, hence the rejection of capitalism by ‘the most con
scious leader’ (cited in Mohan, 1966: 221–2). Berg (1964: 550) made a similar observa
tion that all African leaders view capitalism with suspicion as the ‘economy of the coloniz
ers’ and see the two as mechanisms for oppressing African societies. Some countries
therefore pursued variants of ‘African socialism’, some of which were a thinly disguised
acceptance of capitalism, as in Kenya, or overtly state socialist, as in, for example, Ghana,
Guinea, and Tanzania and later Mozambique (Akyeampong, 2018; Mohan, 1966; Saul,
1985).
In the absence of a surplus-accumulating domestic capitalist class, the state took on the
task of mobilizing domestic resources and attracting foreign direct investments to com
pensate for the low levels of domestic savings and technology. ISI required a strong inter
ventionist state that would protect and support new industries by determining tariff lev
els on competing imports and, in many cases, take part or total ownership of industrial
enterprises in partnership with the foreign multinationals that managed the firms under
Page 9 of 40
1960‒2018
agreements between states and multinationals. Through these policies, many African
states created considerable latitude for intervening in the economy (Chitonge, 2019).
The economic arguments which justified ISI included the need to overcome the domina
tion of export-enclave activities concentrated on resource extraction with little (p. 873)
connection to other sectors of the economy (Mhone, 2001; Ake, 1981). Conversely, ISI
would promote linkages between the different sectors of the economy, ensuring that the
natural resources were used in the industrial sector to manufacture consumer and inter
mediate goods (Ogujiuba et al., 2011). ISI would thus lead to the saving of foreign curren
cy which could be used to import the intermediate and heavy-duty capital goods required
for developing the manufacturing and agriculture sectors, thus promoting the growth of
local industries and achieving faster social and economic transformation. Promoting the
diversification of exports by developing extensive industrial sectors was the ultimate ob
jective of ISI (Ogujiuba et al., 2011).
30.5 Industrial Policies for Import-substitution

Industrialization
Although African countries implemented ISI differently during the first two decades of in
dependence, one common feature was the adoption of protectionist policies against com
peting imports. These included tariffs, import quotas, subsidies, exchange-rate manage
ment, and special preference licences aimed at supporting the infant domestic industrial
sector (Ogujiuba et al., 2011). The idea was to protect infant industries up to the point
where they were able to withstand competition from outside (Westphal, 1982).
For example, in East Africa, the average weighted tariff rate for the region was 27.8 per
cent in 1986 before the liberalization policy was adopted, with some countries such as
Kenya imposing higher than the average tariff rate of 39 per cent: Tanzania with 32 per
cent, Uganda 30 per cent, and Burundi 37 per cent having higher than average tariff
rates (UNECA, 2013). The average weighted tariff rate for North Africa in the 1980‒85
period was 35.2 per cent, 38.5 per cent for West Africa, 33 per cent for Central Africa,
and 19.5 per cent for Southern Africa, and 32.8 per cent for Africa as a whole (Ackah and
Morrissey, 2005).
Contrary to what might be expected from a policy protecting manufactures, tariffs on

these products were 28.1 per cent, with those on agricultural products at 40.2 per cent
and mining/resources at 50.5 per cent1 over the same reference period (Ackah and Mor
rissey, 2005). In countries where the ISI strategy was effectively implemented, these tar
iffs were meant to protect local producers from foreign competition. When implementa
tion was weak tariffs did not offer any form of effective protection (Mendes et al., 2014);
in some cases where protective tariffs existed, imported products were still cheaper and,
if not, they were of higher quality, ultimately displacing local products. (p. 874) Many
African governments used other measures to support infant industries deemed unable to
withstand international competition. To curb imports there were quantitative restrictions
Page 10 of 40
1960‒2018
such as quotas, outright prohibition, and restrictions on foreign currency allocation to
specific importers (Bienen, 1990). In a number of countries, other measures such as the
provision of subsidized loans, preferential licensing, and special foreign currency alloca
tions to firms in strategic sectors were employed to promote the growth of local manufac
turing capability.
The protectionist element of ISI was further justified on the basis that the tariffs imposed
on imported goods provided a substantial source of revenue for the state. In Sudan, for
example, it has been reported that taxes on external trade amounted to 50 per cent of to
tal government revenue during the 1980s (Bienen, 1990: 714). As we shall see later, some
African economies were reluctant to liberalize their markets even under extreme pres
sure from the IMF, the World Bank, and donors, because this entailed the loss of a signifi
cant portion of public revenue.
In many countries state-owned enterprises (SOEs) were created to promote the growth of
local manufacturing (Chitonge, 2019). SOEs became a common feature in most African
countries prior to the liberalization programme of the 1980s (Bates, 1981), and covered a
wide range of sectors from transport to produce marketing to retail chain stores. They
were not always fully state owned: the foreign partners often owned up to 49 per cent of
the shares. The advantage of bringing in foreign capital was ostensibly to raise levels of
capital, technology, and skills. Although there were some choices when it came to tech
nology, in general, the choice of product to manufacture, and its history of technological
development, determined the choice of technique used in production (Stewart, 1972). As
a result, the manufacturing processes for consumer goods were capital rather than
labour intensive and thus unable to make use of the surplus unskilled labour available in
most countries. Usually the SOE strategy was complemented by a suite of institutions, in
cluding those dedicated to mobilizing and providing development finance to the broader
industrial sector (Chitonge, 2019).
In spite of the identification of capitalism with colonialism, as noted above, it was not long
before a small but significant domestic capitalist class was found to be emerging with the
spread of capitalist social relations in agriculture and industry (Sender and Smith,
1986b). This was the case even in those countries where industrialization was initially
dominated by SOEs and across countries with different political ideologies. In the agricul
tural sector, there were increasing numbers of large farmers employing permanent and
seasonal labour; in manufacturing, there were domestic capitalist enterprises employing
wage labour, in some cases looking for collaboration with foreign investment, and there
was a trading class, accumulating capital and investing in productive activities (Swain
son, 1977; Kennedy, 1977; Sender and Smith, 1986b). In spite of the domination of for
eign capital in ISI and the small size of the emerging capitalist class, it was the increas
ing capital accumulation by this class-in-formation that was critical. It allied with foreign
capital to gain the technological capacity and competed against foreign capital, later us
ing its connections to elements of the state to achieve its goals (Leys, 1980). (p. 875)
Page 11 of 40
1960‒2018
30.6 Industrial Performance Under Import-sub

stitution Industrialization
Manufacturing output initially grew rapidly with a large number of countries reporting
high growth rates of manufacturing value added (MVA) (see Table 30.1). MVA growth con
tinued through the 1970s and 1980s with most countries peaking in the 1980s, though
Madagascar and Swaziland’s MVA peaked as late as 2010, and Lesotho even later in
2017. For parts of the continent, in the first two decades considerable progress was
recorded in replacing imports with locally manufactured goods (Bates, 1981) and MVA
growth rates in some cases reached double figures. The dominant sectors in most
economies were food and drink and textiles and clothing, with only South Africa having a
significant capital goods sector.2
However, the ISI strategy did not succeed in Africa for various interrelated reasons. First,
it did not foster the learning process which would have moved ISI on to its second and
third stages. The advanced education and training required to produce a highly skilled
engineering and technically skilled labour force failed to materialize (Bienen, 1990; Sen
der and Smith, 1986b). Second, there was an absence of long-term coordination by a ca
pable interventionist state to plan and resource ISI and to ensure that it matched the
country’s supply of natural resources, rather than using imported raw materials used by
the foreign investors in their activities elsewhere in the world: a sign of quality but also a
reflection of the imbalance in power and knowledge between the state and foreign capital
(Hirschman, 1958; Sender and Smith, 1986b). Low capacity and corruption (Ogujiuba et
al., 2011) further undermined attempts to have such a domestically oriented ISI. Third,
the small size of domestic markets coupled with the inability to compete in export mar
kets further limited ISI (Mendes et al., 2014). Fourth, there was a general failure to moni
tor and evaluate the impact of protection on the economy and to identify strategic sectors
and firms for support (Bienen, 1990). Fifth, state expenditure was arguably directed less
to productive investment and more to generating state employment (Sender and Smith,
1986b).
In sum, the industries supported through ISI required imported machinery and spare
parts and a sufficiently large domestic market to take advantage of economies of scale to
gether with enough foreign exchange earnings to pay for the imports, earnings being
heavily dependent on the exports of primary commodities and subject to fluctuating world
prices. The strategy’s success also required an overvalued domestic currency to be main
tained in order to keep imports relatively cheap. But this adversely affected non-tradition
al exports in particular, restricting foreign exchange inflows, especially when primary
commodity prices sharply declined after 1973‒4. It also adversely affected prices re
ceived by farmers for their export crops and this, together (p. 876) with a tendency to
write off the primary-product export sector, resulted in a decrease in export volumes and
thus a further exacerbation of the foreign exchange constraint (Sender and Smith,
1986a). As a consequence, ISI became unsustainable by the end of the 1970s and was an
Page 12 of 40
1960‒2018
easy target during the 1980s when pressure from the IMF and World Bank to restructure
African economies mounted.
30.7 Resources- and Needs-based Basic Indus

try Policy: An Alternative Strategy?
What was the alternative for the African elites who had been elected on promises that as
independent countries they would do better for their populations than the colonial pow
ers? Was there an alternative to ISI as it was executed? Was there an alternative to the
‘extraverted periphery’ (Amin, 1974a: 600) that would continue colonial policies? One in
dustrial strategy that was proposed in the mid-1970s was to ‘effect a convergence of the
pattern and rate of domestic resource use and domestic demand’ (Thomas, 1974: 125): a
more introverted strategy of development of both agriculture and industry. This required
the domestic needs of the ‘broad mass of the people’ to be defined (Thomas, 1974: 260):
Generalizing from human experience, it appears as if the level of material welfare

hinges largely on the output and consumption of certain tangible and intangible
products. The intangibles are mainly education, recreation, and travel, together
with the usual array of medical, cultural and social services. The tangibles are ad
equate housing, food, clothing, a proper living environment, and the service of
such important consumer durables as transportation, electronic entertainment
equipment and durable household goods furniture, washing machines, refrigera
tors, etc.
(Thomas, 1974: 260‒1)
The closest any countries came to this was in the processing of domestic agricultural
crops such as grains, beverages, and some fruits and vegetables.
Once the population’s needs were defined, production could be oriented towards them us
ing, wherever possible, domestic resources. This required a considerable degree of state
planning. Thomas was writing at a time when in spite of the many failings of socialist
states in Europe and Asia, there was evidence that state planning was geared to the basic
needs of the mass of the population. Where ISI began with producing goods previously
imported for a narrow, largely urban elite, industrialization based on the needs of the
mass of the population would start with planning to manufacture products which would
satisfy those basic needs. This would require the use of domestic rather than foreign nat
ural-resource inputs. Large- and small-scale supply industries would (p. 877) need to be
established, based wherever possible on local deposits of mineral inputs. Countries would
pursue an active science and technology and research and development policy to adapt
imported technologies to local needs and make full use of the interlinkages between the
different sectors of manufacturing, and between them and the agricultural and mineral
primary products.
Page 13 of 40
1960‒2018
Such a strategy would have put a strain on the limited capacity of the state and therefore
would have needed the building of institutions to make such intervention and direction ef
fective. In particular, the planning of manufacturing industry in an integrated, rather than
‘shopping list’ fashion would have required substantial organization and the planned se
quencing of production. The use of foreign expertise to negotiate, where necessary, with
foreign capital might also have helped to produce more effective manufacturing invest
ment (Thomas, 1974). This strategy of industrialization was closely related to that of
‘delinking’ from the global economy (Amin, 1990: 62). Recognizing the difficulties coun
tries of the global South faced in effecting structural transformation because of their inte
gration into a world economy dominated by global capital, Amin argued for a strategy
based on agriculture, on control of foreign capital to import a mixture of technologies
which matched existing levels of development, and on an egalitarian distribution of in
come (Amin, 1990: 63)
Although attractive in theory, a key problem with these alternatives was identifying the
forces that were likely to be able to push states to pursue such policies. Even if states
adopted such strategies, it was unlikely that they had the capabilities, let alone power
and control, to plan industrialization in the way proposed. As for introverted strategies,
there were strong arguments to suggest that an outward-looking approach that generat
ed increased output of primary-product exports would be effective in creating backward
linkages to support both a growing export sector and second- and third-stage ISI (Sender
and Smith, 1986a).
30.8 Liberalization and Industrial Policy

While the ISI strategy had the opportunity to transform African economies, its implemen
tation was cut short when structural adjustment programmes (SAPs) were introduced in
Africa, leading to the abandonment not just of ISI, but IP in general (Lall, 2004; Olukoshi,
2004; Mkandawire and Soludo, 1999). Sundaram (2008), for instance, argues that when
other countries adopted the ISI strategy, African countries were under colonial rule, and
when independent African countries adopted ISI, it was only implanted for about a
decade before liberalization policies were introduced.
These policies, which began during the 1980s, had both economic and political dimen
sions, and seriously impacted on the industrialization of Africa. At first, the focus was on
addressing the apparent economic crisis in Africa manifested by unsustainable (p. 878)
levels of foreign debt, shortages of foreign currency and macroeconomic instability. The
political dimension of the liberalization programme was largely a consequence of the eco
nomic dimension; it focused mainly on reducing the role of the state, reforming the reput
edly inefficient and bloated public sector and promoting competitive politics to mirror
competitive markets through multi-party democracy and free elections (Olukoshi, 2004).
Market liberalization and a diminished role for the state effectively meant that there was
no need for IP: competition in the global market could determine the level and scope of
industrial activity and which industrial sectors to promote. In this can be seen the ‘ex
Page 14 of 40
1960‒2018
treme neoliberal position of the early 1990s which denied any role for the government be
yond providing essential public goods and security’ (Lall, 2004: 75). Further underlying
the World Bank’s approach was a degree of pessimism about the prospects for industrial
ization in Africa, in spite of the variety of experience and performance across the conti
nent and especially in those cases where manufacturing growth did not slow down (see
Tables 30.1 and 30.2; Sender and Smith 1986a).
During the 1980s, almost all African countries implemented SAPs to varying degrees
(Kirkpatrick and Weiss, 1995). The programmes’ rationale was that African economies
had structural problems resulting from extensive state intervention in the economy
through industrial and infrastructure enterprises, the marketing and price-setting of agri
culture produce, the restrictions on importing goods and services, the imposition of tar
iffs on imports, and the rigid exchange-rate policies resulting in overvalued local curren
cies (World Bank, 1981). The structural reforms undertaken by African governments fo
cused on liberalizing trade and currency markets, reducing the role and size of the state
through privatization and retrenchment, and stabilizing the macroeconomy.
Trade liberalization, floating exchange rates, and privatization replaced industrial strate
gy under the SAPs. The first required the protectionist policies introduced as part of ISI
during the 1960s and 1970s to be dismantled (Ackah and Morrissey, 2005) in order to re
move any form of restriction on trade. Liberalizing imports in particular would promote
efficiency in the local economy by subjecting local producers to international competition.
For example, while in most African countries import tariffs averaged 30 per cent prior to
the liberalization programme, after liberalization, average tariffs fell sharply. In Kenya,
where the trade liberalization programme was vigorously implemented in 1993, the aver
age weighted tariff declined from 39.2 per cent in 1986 to 19 per cent in 1997 and to 12
per cent by 2007 (UNECA, 2013: 5). For Africa as a whole, average tariffs declined from
32.8 per cent over the 1980‒85 period to 16.1 per cent in 2002. Average tariffs on agri
culture and manufacturing declined from 40.2 and 50.5 per cent to 14.5 and 13.2 per
cent respectively over the reference period (Ackah and Morrissey, 2005: 14).
Floating exchange rates resulted in severe depreciation of local currencies against the US
dollar. Prior to the SAPs, foreign currency in most African countries was allocated to
strategic sectors by dedicated state institutions to promote the growth of local industries.
When SAPs were introduced, foreign currencies were auctioned to the highest bidder; ac
cording to the theory, this was the most transparent and efficient (p. 879) (p. 880) way of
allocating resources. For the manufacturing industry, currency depreciation increased the
costs of imported inputs and therefore market prices and they were unable to compete
with the now liberalized imports (Bienen, 1990; Kirkpatrick and Weiss, 1995).
Under privatization, the focus was on selling inefficient and uneconomic SOEs. Using con
ditionality attached to loans, the World Bank and IMF forced many African countries to
sell off SOEs, sometimes at give away prices. The privatization programme was not only
restricted to the industrial SOEs, it included the service sector such as water and other
Page 15 of 40
1960‒2018
utilities, transport, and many other social services (Nafziger, 1993; Sundaram, 2008).
Through privatization, most of the SOEs created during the time of ISI were dismantled.
The stabilization programme was mainly focused on prudent macroeconomic manage

ment, on the fiscal side by drastically cutting public expenditure through a rigid commit
ment to a balanced budget and on the monetary side by competitive market-determined
interest rates. Both fiscal and monetary elements affected the industrial sector, the first
by reducing the state’s capacity to subsidize industry and the second by increasing bor
rowing costs. The SAP stabilization programme was mainly handled by the IMF (Chi
tonge, 2015).
For countries that implemented SAPs, this marked the end of an active IP, and perhaps
the prospect of industrializing the continent. Not only that, most African countries effec
tively gave up the sovereign right to formulate and implement their own policies, includ
ing those relating to methods of achieving industrialization (Olukoshi, 2004). IP, no longer
used as a tool for promoting economic growth and transformation, was declared redun
dant by market fundamentalism. As has been observed, ‘no region has been more suscep
tible to swings in the pendulum of ideas of economic development than Africa—from ISI-
planning models of the 1950s to the 1970s, to the swing into massive liberalization and
structural adjustment programmes from the 1980s to date’ (Soludo et al., 2004: 23).
In terms of outcomes, these programmes did not achieve most of the objectives envi
sioned by those who formulated them. It has been observed, for instance, that although
‘price controls, exchange rate misalignments, and budget deficits contributed to Africa’s
crisis, the immediate freeing of markets and contraction of [public] spending could not re
solve the disequilibrium’ (Nafziger, 1993: 432). Other scholars have argued that the rapid
liberalization of African markets and the downsizing of the public service resulted in the
decline of many industrial activities with the momentum of the earlier periods lost
(Olukoshi, 2004; Sundaram, 2008). Economic liberalization’s overall effect on MVA’s
shares of GDP and growth rates varied across the continent with some countries showing
evidence of de-industrialization (Jalilian and Weiss, 2000), while there were also cases,
though very few and sometimes from a low base, which experienced double-digit growth
rates and an increase in manufacturing MVA’s share of GDP (see Table 30.2).
Page 16 of 40
Table 30.2 Manufacturing performance during SAPs, 1981–2000
Page 17 of 40
Manufacturing Average annual Manufacturing Average annual

share of GDP growth rates of share of GDP growth rates of
MVA MVA
Coun 1981 2000 1981– 1991– Coun 1981 2000 1981– 1991–
try 90 2000 try 90 2000
Countries with high MVA share in 1963 Countries with medium MVA share in 1963
(cont’d)
Algeria 11.0 4.6 4.4 –1.5 Ethiopia 4.4 5.6 3.0 1.9
Burki 14.4 12.3 2.6 3.1 Gabon 4.0 4.5 2.5 2.7
na-Faso
Central 13.1 6.1 6.5 0.4 Gambia 5.7 6.8 7.5 1.4
African
Rep.
Côte 15.6 17.2 2.4 –0.5 Ghana 14.1 9.0 4 –1.0

d’Ivoire
Eswati 23.9 38.7 16.8 2.6 Kenya 13.3 10.3 4.8 1.2
ni
(Swazi
land)
Page 18 of 40
Mada 10.1 12.0 1.5 –0.9 Mali 8.6 7.2 5.9 –0.7
gascar
Malawi 12.7 11.6 4.2 0.5 Mozam 7.4 14.8 — 9.2

bique
Mauri 13.5 19.7 9.8 5.3 Namib 9.7 10.0 3.9 2.9
tius ia
Moroc 19.1 18.5 3.2 2.7 Sierra 5.6 3.3 — -1.5

co Leone
Senegal 13.8 14.9 4.8 3.1 Tanza 5.8 9.4 –9.4 2.9
nia
South 22.6 17.5 1.4 1.4 Togo 7.1 7.5 3.5 4.2
Africa
Sudan 7.8 5.7 — — Tunisia 11.9 16.2 3.4 5.7
Zimbab 20.4 13.4 3.5 –0.5 Uganda 1.9 7.1 3.4 5.7
we
Zambia 17.7 9.5 4.6 1.6
Page 19 of 40
Angola 4.0 3.0 –4.6 2.6
Benin 6.7 21.0 7.4 5.9 Guinea 4.2 3.8 3.3 2.6
Botswa 6.1 5.6 12.2 6.8 Guinea 1.3 10.1 8.4 3.4
na Bissau
Burundi 7.3 10.9 5.5 –3.1 Lesotho 9.5 13.6 14.9 7.5
Camero 9.7 20.8 7.4 0.2 Liberia 4.3 3.0 — —

on
Cape 12.6 9.2 — 9.7 Libya 2.5 3.0 — 18.4

Verde
Chad 7.5 8.9 — –2.1 Mauri 6.3 10.6 –1.8 2.8

tania
Congo 6.3 3.5 8.2 –2.4 Niger 3.4 6.8 8 3.1
Congo 7.6 2.9 0.2 –9.1 Nigeria 20.3 13.9 –3.4 –2.0
DR
Equato 6.2 18.3 — — Rwanda 15.7 7.0 2.9 –0.8

rial
Guinea
Page 20 of 40
Page 21 of 40
1960‒2018
30.9 Renewed Potential for Second- and

(p. 881)
Third-Stage Import-substitution Industrializa

tion?
The need for IP is beginning to find traction with the revival of the state as an economic
actor giving direction to industrial development, examples of which will be given later in
this chapter in the cases of Ethiopia, Nigeria, and Rwanda (see also Chapter 29, Whitfield
and Zalk, this volume). The technical possibilities of effecting a more auto-centric indus
trialization strategy have always existed. African countries’ natural resource base in agri
culture and minerals was originally expected to form the basis of its industrialization.
There are, and have always been, strong possibilities for both backward and forward link
ages. The processing rather than the direct export of primary products, both agricultural
and mineral, was undertaken to a limited extent for food and beverages during the first
stage of ISI with some success. Industrial activities which supply products to the agricul
tural sector are relatively less well developed and have considerable potential. The extent
of linkages that have been made varies greatly across the continent. In those countries
where linkages have been more strongly developed, they are more vertical than horizon
tal: skills and competencies learned in the early industrializing sectors have rarely been
transferred to other sectors. However, South Africa, Zambia, and Nigeria, countries with
reportedly deeper vertical linkages, show some signs of horizontal linkages (Morris et al.,
2012).
The need to base successful industrialization on export markets because of the relatively
small size of domestic markets has to some extent been overtaken by the ‘shrinking space
for export-led industrialization’ (Ovadia and Wolf, 2018: 1057) so that directing invest
ment to industries that supply basic goods to a domestic market now enlarged by the
rapid increases in population will require a reversal of trends in income distribution to
create a mass market for basic goods, which will also lead to a growth in intersectoral as
well as consumer demand (Ovadia and Wolf, 2018) and stimulate linkages both forward
and backward. One way to have dealt with issues of scale and market size in the past
would have been a continent-wide or at least region-wide strategy to allocate industries
across countries so that markets were larger and industries internationally competitive
and capable of exporting to the global North. At its inception, the East African Communi
ty was an attempt to do just that, but it fell apart not just because of the 1971 Uganda
coup but because of inter-country disputes resulting in duplication of investments and un
equal sharing of benefits (Mwase, 1979). More recently, the development of new regional
groupings presents possibilities for renewed industrial development taking advantage of
much larger markets (UNECA, 2017).
Successful ISI will also depend on fostering linkages to go beyond its first stage. One ex
ample is the case of the projected iron and steel complex in Tanzania based on (p. 882) do
mestic resources. The construction of the Tanzania‒Zambia railway in the early 1970s
Page 22 of 40
1960‒2018
opened up a part of the south-west of the country that was known to contain deposits of
iron and coal. This could have formed the basis of a steel industry with possibilities for
machine-tool and other downstream production (Luttrell, 1986). Although the possibility
had some political support, it faced what Luttrell termed an ‘alliance’ of state bureaucra
cy, foreign capital, foreign consultancies, and donor agencies which maintained the em
phasis on not having an IP and concentrating on the export of agricultural raw materials
and minerals.3 In 2010, a Chinese company won the contract to develop an iron and steel
complex, and agreed incentives with the Tanzanian government in 2015, but it was re
ported that by November 2018, the project had not been rolled out because these incen
tives had not been forthcoming (Mirondo, 2018), raising the issue of how far the so-called
alliance is holding up the project. Yet this is exactly the kind of project which would have
substantial forward linkages to a range of capital goods manufacturing sectors with on
ward supply to the consumer goods sector.
An important part of this domestically oriented or ‘introverted’ strategy of industrializa

tion would be the need to promote small and medium-sized enterprises (SMEs) to supply
the larger scale. Much of the research on SMEs in Africa has noted their capacity to gen
erate employment but also their lack of connection to larger-scale enterprises, or even to
agriculture, through the supply chains that would be found in developed industrialized
countries (Livingstone, 1991). Many of these enterprises are classified as part of the ‘in
formal sector’ and devoted to the production of cheap consumer goods. An IP that mobi
lized the self-taught skills of the employees of these enterprises with additional training
for conversion into SMEs that could supply intermediate and end production would help
to generate a more integrated manufacturing sector. A further question is why some of
these small entrepreneurs have rarely become larger and constituted the embryo of a sig
nificant domestic capitalist class that could enter into collaborative arrangements with
global corporates to expand the manufacturing base.
Renewed possibilities for IP and industrial growth in Africa have to be viewed against the
backdrop of developments in global manufacturing over the last four decades. Rapid tech
nological change resulting from the revolution in information and communications tech
nology (ICT) offers many different routes to structural transformation. Industrial process
es are increasingly robotized with the result that investment in manufacturing generates
fewer and fewer jobs. The future of capitalism as a system of generating profit for accu
mulation (Amin, 1974b) has been called into question as goods become cheaper to manu
facture with relatively little human labour (Mason, 2015). With a rising African population
of working age, this creates major problems for increasingly urbanized societies. Howev
er, other employment opportunities have been created by the significant growth in ser
vices fuelled by the increasing ‘servicification’ of the manufacturing sector, which subcon
tracts its non-manufacturing activities such as accounting, legal, transport, ICT, and man
agement consulting (Hoekman, 2018).
A further development has been the growth of global value chains in which agri
(p. 883)
cultural and service activities become part of an integrated production process in which
many African economies are closely involved. The examples commonly given, however, re
Page 23 of 40
1960‒2018
fer to such activities as horticulture and tourism, with the manufactured inputs for these
sectors, and for the manufacturing sector itself, being imported (Newfarmer et al., 2018;
Spray and Wolf, 2018). These developments and the changes in trade rules governed by
the WTO have shrunk the policy space available for developing countries to pursue an in
tegrated and domestically oriented IP, although there are still many policy options open to
them because of their developing country status (UNECA, 2016). The development of
technology to exploit sustainable solar, wind, and hydropower energy sources has offered
new opportunities for reducing the costs of manufacturing investment as well as creating
manufacturing linkages (Murray, 2018). The growth of ‘industries without smokestacks’
has been aided by government policy, suggesting movement towards a ‘developmental
state’.4
Finally, global capital has become increasingly concentrated in the hands of a relatively
small number of large corporates dominated by the financial sector, increasing the power
inequality between developing countries and global capital (Vitali et al., 2011). This has
unintentionally created space for Chinese enterprises to invest in other parts of Asia and
in Africa. Chinese investment in Africa, although still a small part of its global direct for
eign investment, has grown sharply in the last few years. It is increasingly driven, less by
the need to satisfy its demand for raw materials, and more by the relocation of its more
labour-intensive industries to areas of the world where labour is cheaper (Shen, 2013).
However, as evidenced by the history of the iron and steel project in Tanzania described
above, how far African countries take advantage of this newer source of foreign invest
ment is dependent on the extent to which the state has, and is able to pursue, an effective
IP.
30.10 Towards an African Developmental State

and Effective Industrial Policy?
The history of the last five decades of African industrialization suggests that there is no
shortage of technical solutions to generate rapid manufacturing industrialization across
the African continent going beyond basic agricultural processing and other light indus
tries. Theories of maximizing backward and forward linkages go back to Hirschman’s
seminal work (Hirschman, 1958) on vertical and horizontal linkages connecting agricul
ture and natural resources to light and heavy manufacturing industry. (p. 884) Explana
tions of poor performance have included comparative disadvantage in the choice of indus
tries; SOE inefficiency; corruption and a lack of control over rent seeking; a low level of
managerial and technical capacities; a low level of technological and planning expertise
in the economics ministries; a lack of coordination between different ministries, SOEs,
and foreign and domestic private enterprise; poor power and communications infrastruc
ture making industrial investment and operations prohibitively expensive; and a lack of
political will (Mendes et al., 2014; Morris et al., 2012; Chitonge, 2019). However, the key
questions are why, irrespective of any of these possible explanations, governments have
not addressed them during the last fifty years, and why IP has not changed to a more do
Page 24 of 40
1960‒2018
mestically based or regional/continental one? Answers require some understanding of the
politics of decision-making and, more broadly, a political economy analysis of power rela
tions and the state, and the distribution of such power among different sections of the rul
ing political class (Soludo et al., 2004; Whitfield et al., 2015).
The political economy explanations include the absence, or small size, of a domestic in
dustrial bourgeoisie, and of an agricultural one capable of branching out into industrial
activities, in spite of the spread of capitalist relations; the existence of a political class
that allied itself with what became global capital rather than with the largely rural popu
lation; a ‘bureaucratic bourgeoisie’ (Shivji, 1976) that even in state socialist countries pri
oritized consumption over production in decisions about investment; and finally, pressure
from the international financial institutions to prioritize primary commodity exports to ex
ploit their ‘comparative advantage’ and essentially abandon IP. This in effect meant a poli
cy of supporting the extractive export industries which had an established power base
within the state rather than developing manufacturing. Even in the most industrialized
African country, South Africa, manufacturing has been dominated by the mineral and en
ergy sectors to the detriment of other manufacturing sectors (Fine and Rustomjee, 1996;
Ashman et al., 2012).
The character of the state is also critical. Under capitalism, a powerful and dominant cap
italist class has a degree of control over the ‘executive of the modern state (which) is but
a committee for managing the common affairs of the whole bourgeoisie’ (Marx and En
gels, 1962: 36). Of course there are fractions of the capitalist class with different inter
ests (manufacturing and mineral extraction or small and large scale), so that the state bu
reaucracy itself may mirror those fractions or be captured by one of them. Where there is
a strong labour movement this can make the state an arena of class struggle.5 Where
there is no one dominant class, establishing a coherent IP may be more difficult. For
African countries, the different classes or strata of farmers, the limited, although signifi
cant, development of an industrial bourgeoisie and a sizeable organized working class,
the existence of a substantial petit-bourgeoisie of retailing and trade, as well as the devel
opment of a powerful state and parastatal bureaucracy has meant that no one group has
dominant political control of the state. The theory of (p. 885) political settlements encap
sulates this view of the distribution of ‘organizational power’ (Khan, 2017: 637) within so
ciety.
This political settlement reproduces itself over time, though it may be ruptured as one
dominant group is replaced by another through a coup d’état, for example. To understand
the political settlement it is necessary to study the history of the origin and development
of organizations or institutions. The distribution of power between organizations is sub
ject to formal rules, more so in the global North, and/or informal rules, more so in coun
tries like those of Africa.
The importance of such an approach lies in its power to explain, not only the differences
between individual countries’ experiences, but also why IP might be effective in some
countries but not in others. The criticism of other approaches, such as the widely accept
Page 25 of 40
1960‒2018
ed neo-patrimonialist thesis, is that their universalism does not explain differences in
country outcomes (Behuria et al., 2017). However, as Kelsall observes, ‘there is no short
age of evidence that neo-patrimonialism, or at least its associated phenomena of big-man
politics, clientelism, corruption and rent seeking is widespread on the continent’ (Kelsall,
2013: 13). However, as Kelsall concedes, neo-patrimonialism does not explain everything,
nor is it confined to Africa. Nor is it necessarily the case that it militates against an effec
tive development, let alone IP. The question is, what is done with the rents, however
gained? If the prospect of earning rents is an incentive to invest in an industry the state is
trying to promote through its IP, then this could generate industrial growth and structural
change. However, this will only happen with an effective IP.
For an effective IP, policymakers need to be knowledgeable about the industry or indus
tries they are promoting, but also to have the support of the ruling elite, or of a powerful
faction within it, thus creating a ‘pocket of efficiency’ (Whitfield et al., 2015: 20). There
must be mutual interests between a faction(s) of the state and domestic capitalists but
their relations cannot be too close to avoid state capture. IP is difficult to pursue in coun
tries where governing elites are fragmented and where there is limited development of a
domestic capitalist class with technological capability, Tanzania being a case in point.
Here, the domestic capitalist class that might have spearheaded industrial investment
with state support was largely associated with a particular ethnic group, those of Asian
origin, so that it was politically unpopular to support this group with aid as part of an IP.6
Corruption on a grand scale became a means of channelling such support to domestic
capitalists, some of whom had a history of high-level involvement in the ruling party,
which still retained strong majority support among the electorate, despite no longer be
ing the only party (Whitfield et al., 2015; Gray, 2013, 2015).
The state’s role in economic development was always seen at a minimum as the provision
of economic infrastructure, such as roads, railways, and power, but also of (p. 886) social
infrastructure, especially schools and hospitals. Much of Africa’s transport infrastructure
was geared to getting primary products to ports, and while this is still an important part
of provision for African economies, building the infrastructure to enable the efficient
transport of domestically produced final and intermediate goods is a key part of any pro
gramme by a putative developmental state for industrial growth and structural change.
As part of Ethiopia’s state-led programme for economic transformation, for example:
The government has been directly and aggressively investing in infrastructure de
velopment and human resource development towards this end. For instance, its in
terventions to increase electricity generation capacity, expand railways and roads
(over 90,000 kilometres of roads, including universal access to rural roads), and
create capacity to train half a million university students amply demonstrate its
developmental orientation…Enrolment in primary schools has increased from be
low 20 per cent in the early 1990s to about 95.3 per cent in 2013.
(Oqubay, 2015: 75)
Page 26 of 40
1960‒2018
However, the putative developmental state does not confine its activities to social and
economic infrastructure. It has an IP that involves the promotion of specific industries
and sectors, and related infrastructure, either directing resources to nascent domestic
capitalist investors and producers or directly participating in such activities through
SOEs. State direction does not necessarily mean neglect of the private sector. On the con
trary it might involve some privatization of SOEs. A state-led IP can use privatization as a
means of strengthening the domestic capitalist class and enabling an IP to be effective
without allowing the state to be ‘captured’ by any one faction of capitalists (Oqubay, 2015).
Nor does IP have to neglect agriculture (ACET, 2017). Developing agriculture by invest
ing to raise productivity and increase farmers’ incomes creates increasing demand for
consumer goods and for producer goods for agriculture, feeding into a general strategy of
auto-centric development. IP is also inextricably linked with trade policy, which both iden
tifies the best way of inserting domestic output into global value chains and targets tariffs
to ensure that necessary imported inputs to the industries so identified are appropriately
priced. These ‘smart’ trade policies (UNECA, 2017) also include stimulating cross-border
trade, especially within regional economic groupings.
The state can also promote industrial policies which increase the share of local content in
final production. Nigeria’s Content Development and Monitoring Board:
[i]s now leading a set of initiatives that very much resemble the interventionist
strategies of East Asian developmental states. They have been involved in promot
ing infant industries such as steel pipe and subsea equipment manufacturers,
spurred the building of new fabrication and yards and led other initiatives that
have created thousands of jobs and, as of late 2015 prior to a change in executive
leadership, were spearheading three industrial parks in the Niger Delta states of
Imo, Cross River and Bayelsa that will specialise in manufacturing of low-cost
equipment, component parts, spare parts and chemicals.
(Ovadia and Wolf, 2018: 1067)
Explanations of more recently successful IP have focused on countries like

(p. 887)
Ethiopia and Rwanda which have clear priority sectors and have made impressive
progress in floriculture and leather goods (Ethiopia) and horticulture and tourism (Rwan
da). In both cases there is evidence of a dominant faction giving state direction and coor
dination and an effective management of rents with close monitoring of performance
(Oqubay, 2015; Kelsall, 2013). However, in countries where power was relatively equally
distributed across organizations, partly as a consequence of political and economic liber
alization, such as Ghana and Tanzania, rent management is less effective as short-term
electoral considerations have led to greater tolerance of corruption and poor perfor
mance (Kelsall, 2013).
Cement production in Ethiopia, Nigeria, and Tanzania further exemplifies issues of rela
tive power. The Ethiopian industry has grown, with state support, to become one of the
top three cement producers in Africa. The most effective state support to the industry
was indirect, through the demand created by its housing and infrastructure programme
Page 27 of 40
1960‒2018
(Oqubay, 2015). The Nigerian industry is dominated by two producers, one domestic and
a foreign cement importer, that have almost twice the market share of the third, foreign-
owned, producer. The larger producer started out as a cement importer but was incen
tivized to go into production under a government IP offering tax breaks, licences, and
protective import duties. Apart from helping to make Nigeria the second-largest cement
producer in Africa, the larger domestically owned company has reinvested in other manu
facturing sectors in Nigeria and in cement production in other African countries, making
its own linkages to agro-alimentary and petrochemical sectors (Ovadia and Wolf, 2018).
Mutual interests can also be found with foreign capital and this again depends on the dis
tribution of power within the ruling elite.
In Tanzania, however, although the government has tried to build up the cement industry
and link it back to construction, the context is very different from Nigeria. The industry is
dominated by foreign companies and while one part of the bureaucracy has incentivized
one firm to invest in cement production, another ministry which seeks to promote the en
ergy industry is disincentivizing the cement company by determining that it buys its ener
gy from higher-pricing suppliers (Ovadia and Wolf, 2018). This is an example of a lack of
coordination between sectors of the state which have conflicting interests governed by
their links with different fractions of capital. It also raises questions of linkages and
choice of technique. As Ovadia and Wolf highlight, ‘what is noteworthy is that govern
ment support for building materials manufacturing in both countries [Nigeria and Tanza
nia] primarily extends to the capital-intensive cement sector. The developmental potential
of linkages between the construction and building materials sector could be enhanced by
targeting the production of less capital-intensive building materials more systematical
ly’ (2018: 1069). This comment raises again the question of why such enhancement does
not take place, a question which theories of the state and its relations with capital might
help to explain.
Page 28 of 40
Table 30.3 Manufacturing performance post SAPs, 2001–17
Page 29 of 40
Manufacturing share of Average Manufacturing share of Average

GDP annual GDP annual
growth growth
rates of rates of
MVA MVA
Country 2010 2017 2001–17 Country 2010 2017 2001–17
Countries with high MVA share in 1963 Countries with medium MVA share in 1963
(cont’d)
Algeria 40.0 35.2 1.0 Ethiopia 4.0 6.2 11.5
Burkina- 7.2 5.5 2.6 Gabon 17.1 17.6 25.8

Faso
Central 18.4 15.7 –1.9 Gambia 14.7 4.5 1.1

African
Rep.
Côte 12.6 12.3 3.6 Ghana 6.4 10.9 4.2

d’Ivoire
Eswatini 32.5 30.1 3.3 Kenya 11.3 8.1 3.2

(Swazi
land)
Page 30 of 40
Madagas 14.5 14.4 4.7 Mali 5.4 6 –0.5

car
Malawi 11.3 9.0 5.1 Mozam 10.5 9.0 6.1

bique
Mauritius 14.2 11.8 1.6 Namibia 12.5 11.2 3.8
Morocco 15.6 15.7 2.9 Sierra 2.2 2.0 2.7

Leone
Senegal 15.5 19.0 3.7 Tanzania 8.7 7.7 7.8
South 13.1 12.0 1.9 Togo 7.2 8.1 8.8

Africa
Sudan 5.6 10.8 7.1 Tunisia 6.5 14.5 2.2
Zimbabwe 9.2 10.8 2.0 Uganda 8.5 8.6 5.3
Zambia 7.6 8.1 5.1
Angola 4.5 6.6 7.1
Benin 13.8 12.3 1.5 Guinea 10.6 10.2 3.2
Page 31 of 40
Botswana 6.4 5.1 4.4 Guinea 11.4 10.5 3.1

Bissau
Burundi 9.5 9.4 –1.0 Lesotho 12.7 13.7 4.8
Cameroon 14.5 15.1 4.2 Liberia 2.6 1.8 –0.2
Cape 5.4 6.4 2.5 Libya 5.0 3.7 –1.3

Verde
Chad 1.0 3.0 9.3 Mauritania 7.1 8.1 1.4
Congo 3.4 6.5 9.2 Niger 4.8 5.7 8.5
Congo DR 16.2 19.5 1.9 Nigeria 6.6 8.7 4.8
Equatorial 13.3 20 19.3 Rwanda 6.2 5.9 7.7

Guinea
Page 32 of 40
1960‒2018
How far have these developments of a more interventionist IP affected performance
across the continent? As with the previous two tables, Table 30.3 shows considerable vari
ation in the share of manufacturing MVA in GDP and in MVA growth rates. (p. 888) Coun
tries such as Ethiopia and Rwanda, which were highlighted above as examples of movers
towards a developmental state with a clear IP, have posted high MVA growth rates while
others such as Ghana moving in the same direction as regards IP have also shown rela
tively high rates of manufacturing growth, and their shares of MVA in GDP have risen sig
nificantly. Tanzania, in spite of the difficulties of advancing to a higher (p. 889) stage of ISI
as regards the political settlement, has also enjoyed a high rate of MVA growth. Morocco
has had a relatively steady but low MVA growth rate but has benefited from its proximity
to European markets and its insertion into Europe’s value chains. More recently, through
a focused IP, Morocco has encouraged investment in aeronautics by companies such as
Boeing and Airbus and is promoting investment in other areas of final-stage manufactur
ing to locate input suppliers locally (UNECA, 2017). However, there is still little sign of
the development of a capital goods sector to supply producer goods in most African coun
tries (Lawrence, 2016).
30.11 Conclusion
This chapter has set out to survey IP in Africa from colonial rule to the present, broadly
within a political economy framework. We have argued that colonial rule in its later
stages saw the genesis of ISI which then became, on a larger scale, the strategy of the
newly independent states. We discussed the reasons why ISI was not successful in achiev
ing the structural transformation of the African economies, placing emphasis on the de
pendence on imported inputs, on primary-product exports to pay for them, and on the
failure to create backward linkages to the rest of the economy and to build on such indus
trial development as there was to orient production to the needs of the majority of the
population. The effective coalition of the ruling elite with foreign capital partly explained
why the path from ISI to a deeper industrial structure was not taken. Any IP was aban
doned during the period of liberalization and structural adjustment as SAPs determined
that inefficient industries should not be supported by the state. Since the late 1990s we
have seen the return of IP to some extent, but also new theories as to why IP has been un
successful or had limited success. In particular, political settlements theory appears to of
fer more country-specific explanations than the more generalized theories associated
with neo-patrimonialism. It is certainly clear that paying attention to the power relations
within the state and wider society, and the political behaviours and struggles around state
power, can offer deeper explanations as to why such unresolved struggles have impeded
effective policies and why, in the few cases where the struggles have been resolved in
favour of a dominant group, policies can be successfully effected.
Page 33 of 40
1960‒2018
References
Abshire, David M. (1969) ‘Minerals, Manufacturing, Power and Communications’, in
David M. Abshire and Michael A. Samuels (eds) Portuguese Africa: A Handbook. London:
Pall Mall Press, pp. 48–69.
Ackah, Charles and Oliver Morrissey (2005) ‘Trade Policy and Performance in Sub-Saha
ran Africa since the 1980s’. Working Paper No. 05/13. University of Nottingham: Centre
for Research in Economic Development and International Trades. (p. 890)
Ackah, Charles, Charles Adjasi, and Festus Turkson (2016) ‘IP in Ghana’, in Carol New
man, John Page, John Rand, Abebe Shimeles, Mans Soderbom, and Finn Tarp (eds) Manu
facturing Transformation: Comparative Studies in Industrial Development in Africa and
Emerging Asia. Oxford: Oxford University Press, pp. 48–69.
ACET (Africa Centre for Economic Transformation) (2017) African Transformation Report
2017. Agriculture Powering Africa’s Economic Transformation. Accra: ACET. Available at
http://acetforafrica.org/acet/wp-content/uploads/publications/2017/10/ATR17-
full-report.pdf.
Ake, Claude (1981) A Political Economy of Africa. Lagos: Longman Nigeria.
Akyeampong, Emmanuel (2018) ‘African Socialism; Or, the Search for an Indigenous Mod
el of Economic Development?’, Economic History of Developing Regions 33: 1–19.
Amin, Samir (1974a) Accumulation on a World Scale. New York: Monthly Review Press.
Amin, Samir (1974b) ‘Accumulation and Development: A Theoretical Model’, Review of

African Political Economy 1(1): 9–26.
Amin, Samir (1990) Delinking: Towards a Polycentric World. London: Zed Books.
Ashman, Sam, Ben Fine, and Susan Newman (2012) ‘Systems of Accumulation and the
Evolving MEC’, in Ben Fine, Jyoti Saraswati, and Daniela Tavasci (eds) Beyond the Devel
opmental State: Industrial Policy into the 21st Century. London: Pluto, pp. 245–67.
Austin, Gareth (2010) ‘African Economic Development and Colonial Legacies’, Interna
tional Development Policy Revue/Internationale de politique de développement 1: 11–32.
Austin, Gareth, Ewout Frankema, and Morten Jerven (2016) ‘Patterns of Manufacturing
Growth in Sub-Saharan Africa: From Colonization to the Present’, in Kevin Hjortshøj
O’Rourke and Jeffrey Gale Williamson (eds) The Spread of Modern Industry to the Periph
ery since 1871. Oxford: Oxford University Press, pp. 345–74.
Bates, Robert (1981) Markets and States in Tropical Africa: The Political Basis of Agricul
tural Policies. Berkeley, CA: University of California Press.
Page 34 of 40
1960‒2018
Behuria, Pritish, Lars Buur, and Hazel Gray (2017) ‘Studying Political Settlements in
Africa’, African Affairs 116(464): 508–25.
Berg, Elliot J. (1964) ‘Socialism and Economic Development in Tropical Africa’, The Quar
terly Journal of Economics 78(4): 549–73.
Bienen, Henry (1990) ‘The Politics of Trade Liberalization in Africa’, Economic Develop
ment and Cultural Change 38(4): 713–32.
Butler, Larry J. (1991) ‘The Ambiguities of British Colonial Development Policy, 1938–48’,
in Anthony Gorst, Lewis Johnman, and W. Scott Lucas (eds) Contemporary British History,
1931–61: Politics and the Limits of Policy. London: Pinter Press, pp. 119–40.
Chang, Ha-Joon (2013) ‘Industrial Policy: Can Africa Do It?’, in Joseph E. Stiglitz, Justin
Yifu Lin, and Ebrahim Patel (eds) The Industrial Policy Revolution II: Africa in the 21st
Century. Basingstoke: Palgrave Macmillan, pp. 114–34.
Chitonge, Horman (2015) Economic Growth and Development in Africa: Understanding

Trends and Prospects. New York: Routledge.
Chitonge, Horman (2019) Industrialising Africa: Unlocking the Economic Potential of the
Continent. New York: Peter Lang.
Minerals-energy Complex to Industrialization. London: Hurst.
General Agreement on Trade and Tariffs (GATT) (1966) Uganda Development Plan. Gene
va: GATT.
Gray, Hazel (2013) ‘Industrial Policy and the Political Settlement in Tanzania: Aspects of
Continuity and Change since Independence’, Review of African Political Economy 40(136):
185–201. (p. 891)
Gray, Hazel S. (2015) ‘The Political Economy of Grand Corruption in Tanzania’, African Af
fairs 114(456): 382–403.
Havinden, Michael and David Meredeth (2010) Colonialism and Development, Britain and
its Tropical Colonies 1850–1960. London: Routledge.
Hoekman, Bernard (2018) ‘Trade in Services: Opening Markets to Create Opportunities’

in Richard S. Newfarmer, John Page, and Finn Tarp (eds) Industries without Smokestacks:
Industrialization in Africa Reconsidered. Oxford: Oxford University Press, pp. 151–69.
Jalilian, Hossein and John Weiss (2000) ‘De-industrialization in Sub-Saharan Africa: Myth
or Crisis?’, in Hossein Jalilian, Michael Tribe, and John Weiss (eds) Industrial Develop
ment and Policy in Africa. Cheltenham: Edward Elgar, pp. 137–58.
Page 35 of 40
1960‒2018
Kelsall, Tim (2013) Business, Politics, and the State in Africa: Challenging the Orthodox
ies on Growth and Transformation. London: Zed Books.
Kennedy, Paul (1977) ‘Indigenous Capitalism in Ghana’, Review of African Political Econo
my 4(8): 21–38.
Khan, Mushtaq H. (2017) ‘Political Settlements and the Analysis of Institutions’, African
Affairs 117(469): 636–55.
Kilby, Peter (1975) ‘Manufacturing in Colonial Africa’, in Peter Duignan and Lewis H.
Gann (eds) Colonialism in Africa 1870–1960. Cambridge: Cambridge University Press, pp.
470–520.
Kirkpatrick, Colin and John Weiss (1995) ‘Trade Policy Reforms and Performance, in
Africa in the 1990s’, Journal of Modern African Studies 33(2): 285–98.
Lall, Sanjaya (2004) ‘Selective Industrial and Trade Policies in Developing Countries: The
oretical and Empirical Issues’, in Charles Soludo, M. Osita Ogbu, and Ha-Joon Chang
(eds) The Politics of Trade and Industrial Policy in Africa. Trenton/Asmara: Africa World
Press, pp. 75–110.
Lawrence, Peter (2005) ‘Explaining Sub-Saharan Africa’s Manufacturing Performance’.

Development and Change 36(6): 1121–41.
Lawrence, Peter (2016) ‘Is There an Africa Problem?’, in John Weiss and Michael Tribe
(eds) Routledge Handbook of Industry and Development. London: Routledge, pp. 350–81.
Leys, Colin (1980) ‘Kenya: What Does “Dependency” Explain?’, Review of African Political
Economy 7(17): 108–13.
Livingstone, Ian (1991) ‘A Reassessment of Kenya’s Rural and Urban Informal Sector’,
World Development 19(6): 651–70.
Luttrell, William L. (1986) Post-Capitalist Industrialization: Planning Economic Indepen

dence in Tanzania. New York: Praeger.
Marx, Karl and Friedrich Engels (1962) ‘Manifesto of the Communist Party’, in Karl Marx
and Friedrich Engels, Selected Works Vol 1. Moscow: Foreign Languages Publishing
House, pp. 33–65.
Mason, Paul (2015) Postcapitalism: A Guide to our Future. London: Allen Lane.
McMillan, Margaret S. and Dani Rodrik (2011) ‘Globalization, Structural Change and Pro
ductivity Growth’. NBER Working Paper No. 17143. Cambridge, MA: National Bureau of
Economic Research.
Mendes, Ana Paula F., Mário Augusto Bertella, and Rudolph F. A. P. Teixeira (2014) ‘Indus
trialization in Sub-Saharan Africa and Import Substitution Policy’, Brazilian Journal of Po
litical Economy 34(1): 120–38. (p. 892)
Page 36 of 40
1960‒2018
Mhone, Guy C. Z. (2001) ‘Enclavity and Constrained Labour Absorptive Capacity in South
ern African Economies’. Paper prepared for the UNRISD meeting on Rethinking Develop
ment Economics, 7–8 September 2001, Cape Town, South Africa.
Mirondo, Rosemary (2018) ‘Chinese Company in $3 billion Liganga-Mchuchuma Steel

Project Reacts’, The Citizen (Tanzania), 27 November.
Mkandawire, Thandika and Charles Soludo (1999). Our Continent, our Future: African
Perspectives on Structural Adjustment. Dakar: CODESRIA.
Mohan, Jitendra (1966) ‘Varieties of African Socialism’, Socialist Register 3: 220–66.
Morris, Michael, Raphael Kaplinsky, and David Kaplan (2012) One Thing Leads to Anoth
er: Promoting Industrialization by Making the Most of the Commodity Boom in Sub-Saha
ran Africa. Available at http://www.prism.uct.ac.za/Downloads/MMCP%20Book.pdf.
Munro, J. Forbes (1976) Africa and the International Economy 1800–1960. London: J. M.
Dent and Sons.
Murray, Sally (2018) ‘New Technologies Create Opportunities’ in Richard S. Newfarmer,

John Page, and Finn Tarp (eds) Industries without Smokestacks. Industrialization in Africa
Reconsidered. Oxford: Oxford University Press, pp. 27–47.
Mwase, Ngila (1979) ‘Regional Economic Integration and the Unequal Sharing of Bene
fits: Background to the Disintegration and Collapse of the East African Community’,
Africa Development/Afrique et Développement 4(2–3): 1–29.
Nafziger, Wayne (1993) ‘Debt, Adjustment and Economic Liberalization in Africa’, Journal
of Economic Issues 27(2): 429–39.
Newfarmer, Richard S., John Page, and Finn Tarp (eds) (2018) Industries without Smoke
stacks: Industrialization in Africa Reconsidered. Oxford: Oxford University Press.
Ogujiuba, Kanayo, Uche C. Nwogwugwu, and Enwere Dike (2011) ‘Import-substitution In
dustrialization as Learning Process: Sub-Saharan African Experience as Distortion of the
“Good” Business Model’, Business and Management Review 1(6): 8–21.
Olukoshi, Adebayo (2004) ‘Democratization, Globalization and Effective Policy-making in

Africa’, in Charles Soludo, M. Osita Ogbu, and Ha-Joon Chang (eds) The Politics of Trade
and Industrial Policy in Africa. Trenton/Asmara: Africa World Press, pp. 43–73.
Oqubay, Arkebe (2015). Made in Africa: Industrial Policy in Ethiopia. Oxford: Oxford Uni
versity Press.
Ovadia, Jesse S. and Christina Wolf (2018) ‘Studying the Developmental State: Theory
and Method in Research on Industrial Policy and State-led Development in Africa’, Third
World Quarterly 39(6): 1056–76.
Page 37 of 40
1960‒2018
Pedler, Frederick (1975) ‘British Planning and Private Enterprise in Colonial Africa’, in
Peter Duignan and Lewis H. Gann (eds) Colonialism in Africa 1870–1960. Cambridge:
Cambridge University Press, pp. 95–126.
Rodney, Walter (1972) How Europe Underdeveloped Africa. Washington, DC: Howard Uni
versity Press.
Rodrik, Dani (2008) ‘Normalizing Industrial Policy’. Commission on Growth and Develop
ment Working Paper No. 3. Washington, DC: World Bank. Available at https://
openknowledge.worldbank.org/handle/10986/28009.
Saul, John S. (1985) A Difficult Road: The Transition to Socialism in Mozambique. New
York: Monthly Review Press.
Sender, John and Sheila Smith (1986a) ‘What’s Right with the Berg Report and What’s
Left of its Criticisms?’, in Peter Lawrence (ed.) World Recession and the Food Crisis in
Africa. London: Review of African Political Economy and James Currey, pp. 114–28. (p. 893)
Sender, John and Sheila Smith (1986b) The Development of Capitalism in Africa. London:
Methuen.
Shen, Xiaofeng (2013) ‘Private Chinese Investment in Africa: Myths and Realities’. Policy
Research Working Paper No. 6311. Washington, DC: World Bank.
Shivji, Issa (1976) Class Struggles in Tanzania. London: Heinemann.
Soludo, Charles, M. Osita Ogbu, and Ha-Joon Chang (2004) The Politics of Trade and In
dustrial Policy in Africa. Trenton/Asmara: Africa World Press.
Spray, John and Sebastian Wolf (2018) ‘Industries without Smokestacks in Uganda and
Rwanda’, in Richard S. Newfarmer, John Page, and Finn Tarp (eds) Industries without
Smokestacks: Industrialization in Africa Reconsidered. Oxford: Oxford University Press,
pp. 341–63.
Stewart, Frances (1972) ‘Choice of Technique in Developing Countries’, Journal of Devel

opment Studies 9(1): 99–121.
Stiglitz, Joseph E., Justin Yifu Lin, Célestin Monga, and Ebrahim Patel (2013) ‘Introduc
tion: Industrial Policy in the African Context’, in Joseph E. Stiglitz, Justin Yifu Lin, and
Ebrahim Patel (eds) The Industrial Policy Revolution II: Africa in the 21st Century. Bas
Sundaram, Jomo K. (2008) Economic Liberalization and Development in Africa.

CODESRIA Monograph Series. Dakar: CODESRIA.
Swainson, Nicola (1977) ‘The Rise of a National Bourgeoisie in Kenya’, Review of African
Political Economy 8: 39–55.
Page 38 of 40
1960‒2018
Thomas, Clive (1974) Dependence and Transformation: The Economics of the Transition
to Socialism. New York: Monthly Review Press.
Thompson, Virginia and Richard Adloff (1975) ‘French Economic Policy in Tropical
Africa’, in Peter Duignan and Lewis H. Gann (eds) Colonialism in Africa 1870–1960.
Cambridge: Cambridge University Press, pp. 127–64.
UNECA (United Nations Economic Commission for Africa) (2013) Two Decades of Trade
Liberalization and Market Expansion in Eastern Africa: Towards a New Economic Geogra
phy. Addis Ababa: UNECA.
UNECA (2016) Transformative Industrial Policy for Africa. Addis Ababa: UNECA.
UNECA (2017) Transforming African Economies through Smart Trade and Industrial Poli
cy. Addis Ababa: UNECA.
Vitali, Stefania, James B. Glattfelder, and Stefano Battiston (2011) ‘The Network of Global
Corporate Control’, PLoS ONE 6(10): e25995.
Weiss, John (2016) ‘Industrial Policy; Back on the Agenda’, in John Weiss and Michael
Tribe (eds) Routledge Handbook of Industry and Development. London: Routledge, pp.
135–50.
Weiss, John and Hossein Jalilian (2016) ‘Manufacturing as an Engine of Growth’, in John
Weiss and Michael Tribe (eds) Routledge Handbook of Industry and Development. Lon
don: Routledge, pp. 26–37.
Westphal, Larry E. (1982) ‘Fostering Technological Mastery by Means of Selective Infant-

industry Protection’, in Moshe Syrquin and Simon Teitel (eds) Trade, Stability, Technolo
gy, and Equity in Latin America. New York: Academic Press, pp. 255–79.
Whitfield, Lindsay, Ole Therkildsen, Lars Buur, and Anne M. Kjaer (2015) The Politics of
African Industrial Policy: A Comparative Perspective. Cambridge: Cambridge University
Press.
World Bank (1981) Accelerated Development in Sub-Saharan Africa: An Agenda for Ac
tion. New York: IBRD/World Bank.
Notes:
(1) These last two tariffs targeted the import of crops and minerals for re-export, which
was common in many African countries. For example, a large share of copper exported
from Zimbabwe, Kenya, and South Africa was actually imported from the DRC and Zam
bia.
(2) For more detail and explanation of African countries’ manufacturing performance, see
Lawrence (2005, 2016).
Page 39 of 40
1960‒2018
(3) The reference to this kind of alliance is suggestive of what would now be referred to as
a ‘political settlement’ (section 30.10).
(4) We will not attempt a definition of this much-used term here. For a succinct account of
the concept and history of the ‘developmental state’ and its usage, see Whitfield et al.
(2015).
(5) For an early discussion of the nature of the state in Africa, see Review of African Politi
cal Economy (1976) 5(3): The State in Africa.
(6) In this context it is worth recalling Hirschman’s observation for Latin America that
ownership of trading and industrial capital by a minority group gave that group much less
influence in persuading the state to support its expansion beyond first-stage ISI
(Hirschman, 1958).
Horman Chitonge
Horman Chitonge is a professor at the Centre for African Studies, University of Cape
Town (UCT). His research interests include agrarian political economy, social wel
fare, and alternative strategies for economic growth in Africa. His most recent books
include: Industrialising Africa: Unlocking the Economic Potential of the Continent
(Peter Lang, 2019), Social Welfare Policy in South Africa: From the Poor White Prob
lem to a Digitised Social Contract (Peter Lang, 2018), Contemporary Customary Land
Issues in Africa: Navigating the Contours of Change (Cambridge Scholars Publishing,
2017) and Economic Growth and Development in Africa: Understanding Trends and
Prospects (Routledge, 2015).
Peter Lawrence
Peter Lawrence is emeritus professor of economics at Keele University, United King

dom. He holds a BA and MA from the University of Sussex and a PhD from the Uni
versity of Leeds. He has taught and researched in several African countries and pub
lished variously on rural and industrial development, macroeconomic policy, and de
velopment strategy.
Page 40 of 40
Index of Names
Index of Names

(p. 894) (p. 895) Index of Names

Note: ‘n.’ after a page number indicates the footnote number.
A
Abramovitz, Moses 32, 52, 112, 683
Acemoğlu, Daron 539–40
Aghion, Philippe 668
Akamatsu, Kaname 259, 299
Akkemik, Ali 119
Akyüz, Yilmaz 25, 26
Aldrighi, Dante 352, 354n.20
Allen, Robert C. 600
Almeida, Mansueto 357
Althoff, Friedrich 318
Altmaier, Peter 612
Amin, Samir 290
Amsden, Alice 26, 32, 97, 181, 217, 350, 684, 814
late industrialization 346, 683
reciprocity/reciprocal control mechanisms 50, 84, 817, 848
Anderson, Philip 671
Andreff, Wladimir 751, 764–5
Andreoni, Antonio 133n.16, 136n.22, 144n.39, 371, 380
Ansu, Yaw 144n.40
Antràs, Pol 247
Aristotle 511
Arocena, Rodrigo 301
Arrow, Kenneth 34n.41, 315, 813, 815
Azad, Rohit 413n.22, 414n.25, 418–19n.31
B
Babbage, Charles 47n.63, 153, 181n.3, 587, 589n.30
On the Economy of Machinery and Manufactures 588–9, 589n.29
Bacha, Edmar 468
Bacon, Francis 514, 516, 538, 539
Page 1 of 14
Index of Names
Bailey, David 487, 613

Bairoch, Paul 68, 599
Balassa, Bela 128n.6, 129n.9
Baldwin, Richard 358, 650
Bangemann, Martin 602, 608
Barca, Fabrizio 495
Barzotto, Mariachiara 496
Baumol, William 241
Beason, Richard 667–8
Becattini, Giacomo 372, 374, 375, 376, 520
Bello, Walden 290
Bendell, Jem 189
Berg, Elliot J. 872
Bértola, Luis 812n.1, 832
Best, Michael 31, 35, 47, 155–6
Bhagwati, Jagdish 127
Bianchi, Patrizio 611, 614, 624
Bianco, Magda 603
Blankenburg, Stephanie 41, 117
Blecker, Robert A. 452, 453, 460
Bodin, Jean 514n.13
Bolsonaro, Jair 823
Bonvesin de la Riva 514n.12
Borensztein, Eduardo 662
Boskin, Michael 541
Botero, Giovanni 370, 512, 513–21, 522, 546
Della Ragion di Stato 317, 514, [link] , 516
economic development 522
economic geography 517, 520
increasing value added 544
influence of 516, 520, 521, 538
a Jesuit 515
On the Greatnesse of Cities 510, 514, 523, 537
(p. 896) Relazioni Universali 514
translations of Botero’s works 516, 517, 537

Bouhia, Rashid 439
Bourdieu, Pierre 285, 302
Brander, James A. 596
Braunstein, Elissa 437, 439
Breschi, Stefano 31, 36, 46, 47n.64
Broecke, Stijn 474
Bruschi, Claudia 357
Buchanan, James 538
Bücher, Karl 538, 540
Burkett, Paul 187
Bush, Vannevar 564, 589n.30
C
Campanella, Tommaso 516
Page 2 of 14
Index of Names
Carlton, David 581–2
Carter, Jimmy 568, 583
Cary, John 534
Castellacci, Fulvio 374
Castro, Antonio Barros de 118
Cecchini, Paulo 546
Chakraborty, Shouvik 413n.22, 414n.25, 418–19n.31
Chandler, Alfred 96, 153, 154, 157
Chandra, Vandana 662
Chang, Ha-Joon 112, 133n.16, 138n.25, 144n.39, 181, 683
CAD industrial strategy 791
Chauvin, Jasmina 142nn.33–4
Chenery, Hollis 66, 137, 191, 780, 789
Childs, Marquis William 510
Chirac, Jacques 608
Choi, Joong-Kyung 828–9
Chung, Moon Young 31, 663, 664, 666–7
Cimoli, Mario 112, 117, 472, 750
Clapham, John Harold 371
Clinton, Bill 541, 568, 583
Coady, David 416n.28
Coase, Ronald 152
Coclanis, Peter 581–2
Coe, Neil 862
Cohen, Stephen 255
Cohen, Wesley 31, 36, 663
Colistete, Renato P. 354n.20
Corden, Max 127–8, 132, 139n.26
Cramer, Christopher 181
Crespi, Gustavo 144
Cripps, T. Francis 67n.10
Cristelli, Matthieu 112
Cui, Yu 751
Curzon Price, Victoria 596
Czempiel, Ernst-Otto 289
D
Dahlman, Carl J. 664
Dallas, Mark 160
Darwin, Charles 548
Dasgupta, Sukti 67
Davies, Rob 295
De Gaulle, Charles 542
De Gregorio, Jose R. 662
De Propris, Lisa 487
De Santis, MarcAntonio 523
Dedrick, Jason 253–4
Defoe, Daniel 512
Degaut, Marcos 757, 770
Page 3 of 14
Index of Names
DeLong, Bradford 255
Delucchi, Mark 278n.17
Deming, Edwards 563, 570, 585
Deng Xiaoping 20n.10, 563, 585, 730
Di Maio, Michele 751–3, 757, 766
Dominguez Lacasa, Iciar 757, 761–2
Doner, Richard 8
Donges, Juergen 596
Dosi, Giovanni 30, 117, 750
Draghi, Mario 523–4
Dunning, John 163
Durkheim, Émile 285
E
Easterly, William 69
Einhorn, Robin 576–7
Elbadawi, Ibrahim A. 460n.6
Elizabeth I, Queen of England and Ireland 513
Elson, Diane 439n.16
Erhard, Ludwig 563, 585
Erten, Bilge 77, 86
F
Fabricant, Salomon [link] , 529–30
Fally, Thibault 247
(p. 897) Federico, Giovanni 603
Feldenkirchen, Wilfried 604
Felipe, Jesus 137n.24
Fernandez-Stark, Karina 491
Ferrari, Enzo 8–9
Fine, Ben 185–6
Fisher, Allan 372
Flanders, Ralph 563, 585
Foner, Eric 575–6
Ford, Gerald 584
Ford, Henry 524n.31, 525, 541, 560, 581
Fourastie, Jean 373
Fox Quesada, Vicente 825
Frederick the Great, King of Germany 600
Freeman, Chris 671
Freitas Barbosa, Alexandre de 817
Friedman, Milton 562
G
Galiani, Ferdinando 510
Gallagher, Kevin P. 338n.2
Gandhi, Mahatma 295
Gee, Joshua 533–4
George, Gerard 667
Gereffi, Gary 164, 491
Geroski, Paul 159–60
Page 4 of 14
Index of Names
Gerschenkron, Alexander 66, 186, 340, 649, 683, 720

advantages of backwardness 32, 797
Gianelle, Carlo 496
Goldthwaite, Richard 511
Gomory, Ralph 241
Gould, Stephen Jay 542, [link] , 544–5
Gourinchas, Pierre-Olivier 81
Grabel, Ilene 418
Graham, Frank 524
Graham, Mark 862
Gramsci, Antonio 287
Granovetter, Mark 315
Greenspan, Alan 583–4
Greenwald, Bruce 780
Griffith-Jones, Stephany 82, 338n.5, 417, 418, 795
Grilli, Enzo R. 77
Guzman, Martin 133n.15, 451
H
Hall, Peter 40
Hallward-Driemeier, Mary 131n.11
Hamilton, Alexander 3, 27, 112, 242, 534, 573
Report on the Subject of Manufactures 27, 317, 337–8
Hanlon, W. Walker 113
Hart-Landsberg, Martin 187
Harvey, David 227n.45
Hassine, Haithem-Ben 616
Hausmann, Ricardo 115, 131–4, 141–2, 142nn.33–4, 791
exports 244, 463, 797–8
‘industrial policy in the small’ 132, 135
innovation 133, 144, 646–7
product space 140
selectivity for resource allocation 136
Havinden, Michael 869
Hayes, Jack Irby, Jr 577
Head, Keith 113
Hely-Hutchison, John 534
Henry VII, King of England 512–13
Hikino, Takashi 97, 154
Hirschman, Albert O. 26, 35, 42, 51n.68, 363, 649, 885n.6
backward and forward linkages 64, 71, 83–4
dynamics of the ‘Hiding Hand’ 8
exports 24
ISI 8
linkages 44, 64, 184–5, 520
Marxism 184, 185
Hobday, Michael 26, 35n.43
Hobson, John 524
Hoover, Herbert 536, 538
Page 5 of 14
Index of Names
Hume, David 520
Hwang, Jason 791, 797
Hymer, Stephen H. 25n.25, 252
I
Ibn Khaldun 538, 539
Imbs, Jean 243
Innis, Harrold 789
Itaman, Richard 851
J
Jacobs, Jane 587, 590
Jacobson, Mark 278n.17
(p. 898) Jacquemin, Alexis 159–60
Jalilian, Hossein 140n.28
Jeanne, Olivier 81
Jefferson, Thomas 573, 582, 583
Jenkins, Rhys 817
Jensen, Michael 562
Jiao, Hao 751
Jiao, Jie 751
Johnson, Chalmers 39, 325, 661, 683
Johnson, Harry 127
Johnson, Lyndon B. 576
Jomo, Kwame Sundaram 788
Jones, Jessie 566
Juhasz, Reka 113
Junankar, Pramod N. 458
Juncker, Jean-Claude 639
Jung, Moosup 669, 673
K
Kahn, Alfred 583
Kaldor, Nicolas 29, 30, 45, 511, 683
growth/growth laws 21–2, 23n.19, 45, 64, 65, 181
Kaldorian strategies in industrial policy 28
Kalecki, Michal 465–6, 468
Kalouptsidi, Myrto 113
Kanwar, Onkar 297
Kay, Cristobal 377
Kejin, Zhao 757, 770
Kelsall, Tim 885
Kessler, Martin 190
Kevles, Daniel 567
Keynes, John Maynard 480, 549, 550
Khan, Mushtaq H. 41, 117, 847
Khoman, Thanat 212–13n.13
Kim, Linsu 30n.31, 31, 35, 37, 664, 667, 683, 684
Kimenyi, Mwangi 771
King, Charles 534, 538
Kirchner, Cristina 296n.10
Page 6 of 14
Index of Names
Kiyota, Kozo 664n.1, 666

Kline, Stephen 571n.9
Klinger, Bailey 141–2
Knudsen, William 586
Kondratieff, Nikolai D. 486
Kowalski, Przemyslaw 251
Krugman, Paul 524–5, 574, 575
Kucera, David 433, 458
Kuddo, Arvo 476
Kumar, Subodh 119
Kuznets, Simon 29, 109, 564–5, 589n.30
L
Labory, Sandrine 611, 614, 624
Lagarde, Christine 415
Lall, Sanjaya 133, 163, 372
Lamoreaux, Nancy R. 575
Lancaster, Kelvin John 372
Landes, David 510
Lane, Nathan 113
Lange, Oscar 375
Lawrence, Mark 401
Lazonick, William 155, 584
Lazzarini, Sergio 352, 361–2
Lee, Jongwha 662
Lee, Joonkoo 491, 835
Lee, Keun 31, 32–3, 662, 663, 664, 666–7, 668, 669, 670, 672, 673, 721, 753, 791
Fourth Industrial Revolution 765
NIS5 835
Leibniz, Gottfried von 511
Lenin, Vladimir 193, 524, 723
Leontief, Wassily 191
Levinthal, Daniel 31, 36, 663
Lewis, Michael 4
Lewis, W. Arthur 67, 77, 126n.2, 465
Lewis, Zenia 771
Li Keqiang 298
Li Rongrong 724
Li, Yuefen 290
Lin, Justin Yifu 28n.28, 33, 134–5, 137, 138n.25, 537, 661
CAF/CAD industrial strategy 790, 791
leapfrogging 797
Lin, Li-Wen 732, 733
Lin, Nan 720, 733
Lincoln, Abraham 573
List, Friedrich 3, 112, 547, 600
infant industry theory 27
National System 27, 317
Page 7 of 14
Index of Names
Little, Ian 125, 128n.7, 140n.29

welfare theory 125–6n.1
Liu Shaoqi 723
(p. 899) Liu, Xielin 751, 759
López-Gómez, Carlos 769
Lopez-Gonzalez, Javier 650
Lösch, August 520
Lula da Silva, Luiz Inácio 822
M
Macri, Mauricio 824
Macron, Emmanuel 616
Madison, James 582
Magdoff, Harry 579
Malerba, Franco 31, 36, 46, 47n.64
Malm, Andreas 188
Mandelbaum, Kurt 464
Mann, Laura 856–7, 862
Mao Zedong 723
Marel, Erik 248
Marsh, Peter 769
Marshall, Alfred 47n.63, 127, 374, 587
industrial districts 520
Industry and Trade 373
law of increasing return 589–90
‘principle of continuity’ 9
Principles of Economics 520, 522, 536–7, 589
Marshall, George 535
Marshall Plan 511, 525, 535–6, 549, 601
Marx, Karl 29n.30, 181, 285, 370, 439n.15
labour theory of value 529
Marxism and industrial policy 183–90
Marxist approach to manufacturing and industrialization 182–3
‘modern machine industry’ 182
Mathews, Jessica 579n.15
Mathews, John A. 662, 670
Mathieu, Claude 616
Mazzucato, Mariana 39, 82, 316, 499, 683, 693, 800
Mbele, Sizwe 758, 771
McCombie, John 137n.24
Meachan, Carl 757, 770
Meade, James 127
Meinecke, Friedrich 516
Meredeth, David 869
Mesmer, Franz Anton 548
Meyer, David 576
Milberg, William 861
Milhaupt, Curtis J. 732, 733
Mill, James 509, [link]
Page 8 of 14
Index of Names
Mill, John Stuart 316, 509, [link] , 511, 521

Mirrlees, James 140n.29
Misselden, Edward 507, 508, 512, 516, 517
Mittelman, James H. 288
Modi, Narendra 290–1, 298, 303
Mokyr, Joel 349
Monga, Célestin 135n.18, 137
Mortimore, Michael 300n.12
Moslener, Ulf 343
Mourão, Hamilton 295, 297
Müller-Armack, Alfred 510n.2
Mun, Thomas 517
Musacchio, Aldo 352, 361–2
Myrdal, Gunnar 590, 788
N
Naastepad, C. W. M. 452, 453, 458
Nassif, Andre 828
Nataraj, Shanti 474
Naughton, Barry 725
Neale, Jonathan 189n.6
Necker, Jacques 548
Nelson, Richard 30, 31, 35, 133, 315, 317, 325
The Moon and the Ghetto 311–12, 323
upgrading 442n.21
Neto, Antonio S. M. 468
Nieburg, H. L. 567
Nixon, Richard 562, 583, 584n.23
Nolan, Peter 721
Noland, Marcus 828n.16
North, Douglass C. 161
Nübler, Irmgard 442n.21
Nurkse, Ragnar 464, 465
O
Ocampo, José Antonio 21, 76, 77, 86, 144, 418, 451, 795, 812n.1, 832
dynamic efficiency 19n.6, 63n.1
innovation 44, 683
NDBs 82
Ohno, Kenichi 150
Ohno, Taiichi 563, 585
Okazaki, Tetsuji 666
O’Neill, Jim 754
Oqubay, Arkebe 44, 84, 181, 851, 886
(p. 900) infant industry theory and manufacturing 27n.27
reciprocity 50n.67
Ricardian/Kaldorian strategies in industrial policy 28
Oresme, Nicolaus 523
Ortiz, Luis 522n.26
Ostrom, Elinor 315
Page 9 of 14
Index of Names
P
Page, John 788
Page, Scott 444
Palma, José Gabriel 67, 832
Pasinetti, Luigi Ludovico 21, 24, 379–80
technical knowledge 22n.15, 24n.21, 29
Patman, Wright 583
Paus, Eva 30n.33, 33, 52, 109
Pavitt, Keith 373–4
Pearson, Margaret 720
Pellegrin, Julie 634
Peña Nieto, Enrique 825
Penna, Caetano C. R. 82
Penrose, Edith 153, 587, 590
Peres, Wilson 827
Pérez, Carlota 79, 671
Perroux, François 520–1
Phelps, Edward 442n.21
Pieper, Ute 458
Pierce, C. S. 511
Pierrehumbert, Raymond 401n.8
Pigou, Arthur 127
Piketty, Thomas 157, 555–6
Pisano, Gary 155, 325–6
Polanyi, Karl 38, 42, 290, 315
Polhem, Christopher 516
Polterovich, Victor 795, 796–7, 801n.5
Popov, Vladimir 795, 801n.5
Porcile, Gabriel 144, 467, 468, 472
Porter, Michael 154, 499, 520
Postali, Fernando A. 352
Powell, Lewis 562, 584n.23
Prebisch, Raúl 21, 64, 76, 77, 78, 184, 464, 471
Prentiss, Mara 398, 402
Primi, Annalisa 827
Pritchett, Lant 66, 131n.11
Purushothaman, Roopa 778
Q
Qian, Yingyi 719
R
Rada, Cordina 44
Radosevic, Slavo 647
Raleigh, Sir Walter 516
Ramaphosa, Cyril 295n.8
Ramaswami, V. K. 127
Ramos, Narciso 213n.13
Rasiah, Rajah 683
Reagan, Ronald 568, 583
Page 10 of 14
Index of Names
Reddy, Sandjay G. 791, 793

Reinert, Erik S. 25n.23, 374, 683
Ricardo, David 27, 28, 112, 285, 316, 370
Cold War economics 508, 509, 535, 548
comparative advantage theory 507, 508, 529, 540
Principles of Economics 509, 529
‘Ricardian Vice’ 525, 541
trade theory 525, 535, 536, 541
Ricci, Matteo 516
Richardson, George 587
Rigo, Davide 253
Ritchie, Bryan 8
Robinson, James 539–40
Robinson, Joan 467, 471
Rodríguez, Francisco 68
Rodrik, Dani 65–6, 68, 115, 131–4, 144, 257, 622n.2, 646, 791, 800
de-industrialization 67
exports 797–8
hyper-globalization 290
‘industrial policy in the small’ 132, 135
poverty 793
product space 140
Roncolato, Leanne 458
Roosevelt, Franklin Delano 566, 573, 577, 583, 585
Arsenal of Democracy policy framework 563, 574, 585, 586
‘organized money’ 583n.21
Roosevelt, Theodore 582
Rose, Andrew K. 804
Rosenau, James 289
Rosenberg, Nathan 29–30, 111, 378–9
(p. 901) Rosenstein-Rodan, Paul 464, 649
Ruggiero, Renato 549
Russell, R. Robert 119
Rustomjee, Zavareh 185–6
S
Saez, Emmanuel 555–6
Samuelson, Paul 508–509, 535, 548, 549, 583
Sayers, Richard S. 342–3
Schmoller, Gustav 510, 547
Schneider, Ben Ross 357
Scholte, Jan Aart 485
Schonfield, Andrew 318, 319
Schroeder, Gerhard 608
Schumpeter, Joseph 64, 181, 271, 370, 520–1, 523, 524–5, 587
on Babbage 588–9
creative destruction 30, 64, 69
innovation 69, 516, 683
Page 11 of 14
Index of Names
Theory of Economic Development 521

Schwab, Klaus 298
Schwartz, Herman 28, 39
Scitovsky, Tibor 465
Seckendorff, Veit Ludwig von 514, 516
Seers, Dudley 281
Seguino, Stephanie 437, 439
Sekara, June 316
Seligman, Edwin 508
Selwyn, Benjamin 187
Senghor, Leopold 872
Serra, Antonio 370, 517, 521–3, 537, 538
Shi Zhengrong 275
Shin, Hochul 668, 673
Simachev, Yuri 752
Sing, Parminder Jeet 768
Singer, Hans W. 21, 76, 77, 78, 464, 471
Singh, Ajit 67
Sisulu, Lindiwe N. 298
Sklar, Sarah 338n.2
Skocpol, Theda 361n.27
Slater, Dan 8
Smith, Adam 3, 240–1, 285, 370, 539, 548, 683
free trade 316, 507–508
‘vent for surplus’ models of international trade 68
on wealth 240n.2, 598
The Wealth of Nations 508, 598
Soete, Luc 671
Solow, Robert M. 30n.32, 48, 65, 111
Somary, Felix 547–8
Sombart, Werner 539, 540n.54
Sorensen, Charles 586
Spratt, Stephen 418–19
Sraffa, Piero 371–2, 373
Stalin, Joseph 191, 192, 195, 720
Steinbeck, John 541
Stiglitz, Joseph 117, 342, 451, 661, 750, 788
Stoller, Matt 583, 584
Storm, Servaas 281, 452, 453, 458
Sturgeon, Timothy 228, 491
Subramanian, Arvind 290, 800
Suehiro, Akira 287–8
Suharto 214n.18, 686
Summers, Larry 794
Sundaram, Jomo K. 877
Sunkel, Osvaldo 300n.12
Sweezy, Paul 579
Symcox, Geoffrey 510n.3, 516
Page 12 of 14
Index of Names
Syrquin, Moshe 780
Szirmai, Adam 119
T
Tacchella, Andrea 112
Tarling, Roger J. 67n.10
Tassey, Gregory 379
Taylor, Lance 44, 468
Teaford, Jon 581
Teece, David J. 155, 316, 325–6
Tejani, Sheba 433
Thatcher, Margaret 604
Thirlwall, Anthony 4, 21–2, 25n.22
Thomas, Clive 876
Timmer, Marcel P. 119, 245–6
Tomlinson, Philip R. 613
Tourk, Khairy 769
Traub-Merz, Rudolf 230
Tregenna, Fiona 181, 182
Trippl, Michaela 647
Truman, Harry S. 536
Trump, Donald 536, 537
Tushman, Michael 671
U
(p. 902) Uy, Marilou 342
Uztáriz, Gerónimo de 516

V
Valdivia, Walter 7
Van Ark, Bart 119
Van der Linde, Claas 499
Vasilescu, Bogdan 444
Veblen, Thornstein 30n.31, 540, 590
Verdoorn, Petrus Johannes 74
Verdoorn’s Law 21, 181, 525
Verri, Pietro 370
Vico, Giambattista 511
Videgaray Caso, Luis 825
Voelker, Paul 568
Von Thünen, Johann Heinrich 517, 519–20
industrial city 517, 519 [link]
Von Tunzelmann, Nick 604
W
Wacziarg, Romain 243
Wade, Robert 43, 52, 52n.70, 181, 683
Wajcman, Judy 445
Wallis, John J. 575
Wang, Lili 119
Watanabe, Tsunehiko 191
Weber, Alfred 520
Page 13 of 14
Index of Names
Weber, Max 285, 325

Weinberg, Alvin 318–19
Weinstein, David E. 667–8
Weiss, John 17n.1, 140n.28
Westphal, Larry E. 664
Whitaker, Kenneth 563, 585
Whitefoot, Kate 7
Williams, Raymond 284
Williamson, John 130
Wilson, Dominic 778
Wilson, Woodrow 582
Winter, Sidney 31, 35, 325
Wolfe, Martin 373
Wolff, Christian 511, 851
Woo-Cummings, Meredith 361n.27
Wright, Gavin 577, 580n.18
Wu Bangguo 723–4
X
Xenophon 510
Xi Jinping 294–5, 297, 298
Y
Yajima, Toshiro 467
Yang, Maw Cheng 77
Yeung, Henry Wai-chung 862
Young, Allyn 22, 181, 587, 590, 683
Z
Zahra, Shaker 667
Zucman, Gabriel 555–6
Page 14 of 14
General Index
General Index

(p. 903) General Index

Note: Page numbers in bold refer to tables and boxes; page numbers in italics refer to fig
ures; ‘n.’ after a page number indicates the footnote number.
1997 Asian financial crisis 346

ASEAN 210n.7, 214
China 727
deregulation 214
Indonesia 214n.18, 690, 701
privatization 214
2008 global financial crisis 7, 53, 259
2007–9 North Atlantic financial crisis 67n.9, 80–1, 82
EU industrial policy 609, 621, 633, 639
globalization narratives 296
GVCs 485, 492
Indonesia 691
neo-liberalism 341
revival of industrial policy 661
state responses to 178
US Federal Reserve Bank 415, 523
A
abduction (phronesis) 511–12
ABRRs (asset-based reserve requirements) 446–7
absorptive capacity 26, 30n.33, 31, 34, 36n.45, 98, 258, 316, 662–3
building of 667
‘potential’/‘realized’ 667
South Korea 662, 663–7 see also capacity
advanced countries/economies 4, 13, 281, 340
energy consumption in 404
R&D spending 340
AEC (ASEAN Economic Community) 207, 215–16, 696, 703
2005 Blueprint 221–2
2020 Blueprint 209
Page 1 of 79
General Index
2025 Blueprint 221–2, 231 see also ASEAN

AEG (General Electricity Company, Germany) 345, 605
aerospace industry 487, 604, 718, 721, 828
Brazil 342, 345, 351, 353, 357, 358
China 735, 738
NDBs 362, 363
Africa 5, 14, 889
1980s debt crises 67n.9
African developmental state and effective industrial policy 883–9
African socialism 872
agricultural sector 868, 870, 876, 877, 881, 882, 886
BRICS and 297, 750, 771–2
capability building 843, 861
capitalism/colonialism identification 872, 874
China and 771–2, 846, 850, 860, 867, 883
commodity booms 469, 867
de-industrialization 67, 78
digital production technologies 383
diversification 844, 846, 873
exchange rates 878, 880
failure of industrial transformation 868, 875–6, 884, 889
gender/growth regime 442
GVCs 861, 862, 883, 886
industrial policy 5, 40–1, 867
industrial policy, colonial period 867–70, 889
industrialization 537–8, 788, 868, 878, 880, 881, 883
(p. 904) industrialization, introverted strategy of 876, 877, 882
infant industry protection 873–4

infrastructure 886
liberalization and industrial policy 874, 877–80, 889
macroeconomic policies 452, 880
manufacturing 868–870, 871, 875, 879, 883, 888
manufacturing value added 870, 871, 875, 880, 887, 888
mineral sector 858, 868, 870, 877, 881
political economy 867, 884–5, 889
post-colonial period 870–2
privatization 878, 880, 886
protectionism 873
resources-/needs-based basic industry policy 876–7
SMEs 882
SOEs 874, 880, 884, 886
structural adjustment programmes 844, 846, 876, 877–8, 879, 880, 889 see also Africa: ISI;
sub-Saharan Africa
Africa: ISI 846, 870–2, 889
industrial performance under ISI 871, 875–6
industrial policies tools for ISI 870, 873–4
political and economic objectives 872–3
renewed potential for second-/third-stage of ISI 881–3 see also Africa
Page 2 of 79
General Index
African Development Bank 412, 418

agriculture 794
1930s crisis 541
Africa 868, 870, 876, 877, 881, 882, 886
agrarian modernization 479n.19
agricultural employment 434n.10
agriculture/manufacturing linkage 7, 23, 45, 377, 465, 517, 520, 883
ASEAN 217
Cuba 196
industrialization 465, 542
‘industrialization of freshness’: agriculture/manufacturing linkage 7, 23, 45
manufacturing agrarian change 376–7
monopoly 541–2
nichification 541–5
AI (artificial intelligence) 53, 162, 165, 166, 298, 300, 323, 380, 486, 487, 497, 611, 617, 765, 769
AIIB (Asia Infrastructure Investment Bank) 80
Airbus 345, 358, 617, 622, 633, 889
Alibaba 166, 751
allocation
allocative efficiency 157, 326–7, 328
allocative optimality 125
credit allocation 145, 447
re-allocation 106
resource allocation 37, 75, 84, 134, 155, 789, 791, 813n.2
selectivity for resource allocation 136–8
Amazon 166–7
antitrust policy 316, 597
Italy 603
UK 604
US 541, 561, 562, 582–3
APEC (Asia-Pacific Economic Cooperation) 695
Apple 253–4, 329, 493
Argentina 812, 824–5, 830–1, 835, 836
1990s crisis 821
Convertibility Plan 821, 824
differential taxes 824, 825
protectionism 824
trade liberalization 819 see also Latin America
Armenia 783, 805
‘arms-length’ market relationship 151, 152, 164, 320
China 716, 718, 719
ASEAN (Association of Southeast Asian Nations) 12, 53
1997 Asian financial crisis 210n.7, 214
agricultural/agro-processing sectors 217
ASEAN-5 countries 212, 215n.23
ASEAN Summits 214n.22, 215
‘ASEAN Way’ 211n.11
Page 3 of 79
General Index
China and 209n.5, 210, 212, 216, 220–1

competitiveness 209, 221n.36
creation of 211–13
(p. 905) economic growth policies 208, 213
ELI 213–14
FDI 213, 218, 226
GDP 210
historical evolution of economic policies 211–16
industrial policy 209–10, 216–24
industrial policy, future of 230–1
industrial upgrading 208, 209, 210, 222, 223, 228–9, 230, 231
infant industry protection 214
innovation 209, 220, 222, 231
IPR 219–20, 224
ISI 213, 220n.33
manufacturing 213
member states 207n.1, 220
non-tariff barriers 208n.4, 215, 223, 224, 230
regional integration 207, 214–16, 218, 221
regional policy coordination 208, 212, 214, 223, 224–30
state intervention 211n.11, 217, 220n.33, 226 see also the entries below for ASEAN; AEC
ASEAN: agreements, projects
ACIA (ASEAN Comprehensive Investments Agreements) 218–19
AFTA (ASEAN Free Trade Agreement) 208–209n.4, 215, 692, 695, 703
AIA (ASEAN Investment Area) 218
AICA (ASEAN Industrial Complementation Arrangement) 218, 220
AICO (ASEAN Industrial Cooperation) 209–10, 230
AIJV (ASEAN Industrial Joint Venture) 209–10, 218, 230
AIP (ASEAN Industrial Project) 218, 220
ASEAN Complementarity Schemes 229–30
ASEAN IGA (ASEAN Investment Guarantee Agreement) 218
ASW (ASEAN Single Window) 215
ATIGA (ASEAN Trade in Goods Agreement) 208n.4, 215
BBC (Brand-to-Brand Complementation Scheme) 208, 230
CEPT (Common Effective Preferential Tariff) 215, 220
CSAP (strategic action plan) 221
FTAs (free trade agreements) 209
IAI (Initiative on ASEAN Integration) 215n.25 see also ASEAN
ASEAN: automotive industry 220, 224–8, 230
GVCs in 208, 218, 227–8
Indonesia 226, 228
Kia 224
regional policy coordination 229–30
Thailand 208, 226, 228
Toyota 208, 218, 220n.33, 224
wages 226 see also ASEAN; ASEAN: GVCs
ASEAN: firms
inclusive participation 209, 221–2, 230
Page 4 of 79
General Index
lead firms 209, 210, 223

local firms 207, 208, 209, 224, 231, 253, 258
SMEs 209, 221–2, 223, 230–1, 258
TNCs 213, 220
‘unicorn’ firms 219 see also ASEAN
ASEAN: GVCs 207, 215, 216–24, 227–9, 253
benefits from GVCs 208–209
GVC integration 211, 230
GVCs in automotive industry 208, 218, 227–8
policy assessment 222–4, 230–1 see also ASEAN
Asia
industrial policy 788 see also East Asia; South East Asia
Asian Development Bank 142n.35, 304
Asian Tigers 13, 213, 266, 325, 788
ASPAC (Asia and Pacific Council) 212
asymmetries 13, 66
‘asymmetric opening’ 668
information asymmetry 31, 72, 152, 216, 313
power asymmetry 224n.38, 231, 258
technological asymmetry 94, 104, 811, 813
(p. 906) austerity 549
debt and 524
de-industrialization and 469
fiscal austerity 460, 468, 479
Latin America 468–9, 818
authoritarianism/autocracies 192
ASEAN 220n.33
automation 6, 45, 53, 101, 164, 165, 382, 384
Germany 611
labour-intensive production and 487
automotive industry 225–6, [link] , 527
automotive demand 228
China 225, 739
Eastern and Central Europe 252
employment 225, 226
Germany 225
global North/global South relation 225
India 140n.28
Indonesia 226, 228
Italy 525, 601
Japan 138n.25, 225, 527
nation-based development paradigm 527
productivity explosion 527
South Africa 845, 858–9, 862
South East Asia 697, [link] , 699
South Korea 225, 527, 697, 699
Page 5 of 79
General Index
Spain 527
Thailand 208, 226, 228, 704, 709
US 225, 525 see also ASEAN: automotive industry
Azerbaijan 780, 781, 783, 785, 787, 790, 793, 805, 806
B
backwardness 468, 520, 720
advantages of 32, 797
backward economies 114–15, 340
backward/forward linkages 64, 71, 83–4, 184–5, 188, 191, 244, 247, 251, 377, 430, 835, 881
backward integration 762, 849–50
Bahrain 189
Baltic states 434, 781, 805
Bangladesh 137, 350
Bank Windhoek (Namibia) 272
bankruptcy 79, 275, 539, 545, 725
banks/banking see development banking; MDBs; NDBs
‘beggar-thy-neighbour’ policies 479, 795
Belarus 549–50, 805, 806
Berlin Wall, fall of (1989) 508, 510, 547, 781
economic growth since [link]
big-data analytics 164, 166, 167, 488, 617, 765, 768
BRICS 769
China 767–8
‘big push’ industrialization 345–6, 355n.22, 363, 463–9
‘big push’ models of ISI in BoP-constrained developing countries 464
BoP constraint 464–5
demand-side and 464
bioenergy 396, 408n.17
biotechnology 79, 196, 328, 341n.6, 345n.46, 693, 710, 711, 823, 824
R&D 707, 709
BIS (Bank for International Settlements) 81
BNDES (Brazil Development Bank) 82, 272, 345, 346, 819, 823
assessment of 356–8
BNDESPar 352, 357
corruption 359–60
emergence of innovative firms 353n.18
importance of broad political economy factors 354–6
‘insider/outsider’ problem 354, 359, 360
JBS case 360
limitations of 350–1, 354–6
long-run finance 339, 351, 352–3, 356
long-run finance, single/limited model 339, 348, 354, 358–9
‘national champions’ 345, 351, 352–3, 354, 356–7, 358
privatization programme 351, 352, 361, 819
SOEs 351, 352, 353n.18
structural transformation 353
subsidies 352n.16, 356, 357–8, 359, 360n.26
success 338, 351, 353 see also Brazil, development banking; development banking
Page 6 of 79
General Index
Boeing 544, 566, 724, 889

(p. 907) bonds 271
diaspora bonds 350
green bonds 271–2
long-run bonds 346, 362, 363
book-keeping: double-entry book-keeping 539
BoP (balance-of-payments) constraints 18, 24, 75, 197, 237, 432, 815, 863
export-led growth model with BoP constraint 460–3, 480
export-led growth model with cumulative causation and BoP constraint 463, 464, [link]
growth and 25n.22
macroeconomic policy and industrialization 452, 460–3, 464–7, [link] , 471, 473, 477, 480
South East Asia 688, 689, 692, 694, 697, 699, 704, 711
BoP crises 340, 345, 348, 355, 362, 815, 821, 827
BoP dominance 815
Botswana 781n.1, 801n.5
Brazil 118–19, 812
1990s crisis 821
clean energy investment project 419, 425
convergence/divergence 817–18, 831
de-industrialization 354, 818, 823, 832
digital production technologies 382–3
economic reforms (Washington Consensus) 752
fossil-fuel consumption 421, 426
globalization 292, 295, 297
hard-budget constraint legislation 356
ISI 751
PAC (Programme for the Acceleration of Growth) 823
PBM (Plano Brasil Maior) 823
PDP (Policy of Productive Development) 823
PITCE (Industrial, Technological and Trade Policy) 822–3
privatization 351, 352, 356n.23, 361, 752, 819
R&D 823, 834, 836
restructuring of Brazilian capitalism 352
solar industry 419–20
telecommunications 670
trade liberalization 819, 823
unemployment 360–1
wage-goods inflation 467 see also BRICS; Latin America
Brazil, development banking 350–61
politics of 358–61
positive side of 351–3
variations in manufacturing performance 351 see also BNDES
BRICS (Brazil, Russia, India, China, South Africa) 286, 609, 749, 772–3
catch-up 760, [link] , 773
challenge of heterogeneous industrial performances 757–65, 772
China 749, 757, 760, 770, 772
CIP Index (Competitive Industrial Performance Index) 760–1, [link]
collaboration 749, 755–7, 770, 772
Page 7 of 79
General Index
contribution to global economic activity 757–8, [link]

debates on 754, 757
e-commerce 755, 772
as economic block 755–6
economic modernization 759–60
FDI 751, 764–5, 764
from BRIC to BRICS 754–5
globalization narratives 284, 286–7, 292, 294–5, 296–7, 301, 303
GVCs 250–1, [link]
ICT 298, 301, 755, 760, 767, 769, 771
industrial policy 749, 750–3, 756–7
innovation capabilities 761–2
integration process 749, 754, 757, 764, 772
integration through trade and investment 762–5
IPRs 755, 759, 768, 772
manufacturing 756, 757, 758, 762–3, 763, 772
manufacturing value added 762
national champions 751
national governments 749, 751
productive capabilities 759–61
regionalism 292, 303–304
SMEs 756, 757, 767
(p. 908) STI collaboration 685, 751, 752, 755–6, 766, 768, 770, 772
structural change and capability accumulation 759–62, [link]

technology upgrading 761 see also the entries below for BRICS
BRICS and 4IR 298, 749–50, 765–70, 772
building basic framework conditions 767–8
fostering demand and adoption of 4IR technologies 768–9
strengthening skills and research capabilities 769–70
BRICS: international dimension 750, 770–2
Africa and 297, 750, 771–2
BRI (Belt and Road Initiative, China) 772
NDB (New Development Bank) 80, 770, 772
NEPAD (New Partnership for Africa’s Development) 297, 771
UN system 757, 770
UNIDO 757
BRICS: partnerships, programmes, strategies 769
2017 Action Plan 756
BRICS Consolidated Technology Platform 757
BRICS Economic Cooperation Strategy 755
BRICS ICT Development Agenda and Action Plan 767
BRICS Industry Ministers Meetings 756
BRICS Institute of Future Networks 769
BRICS IPRCM (Intellectual Property Rights Cooperation Mechanism) 768
Framework of BRICS Closer Economic Partnership 755
PartNIR (BRICS Partnership on New Industrial Revolution) 767
Strategy for the BRICS Economic Partnership 756
Page 8 of 79
General Index
bullionism 514, 521
bureaucracy 50, 64, 317, 339, 730, 882, 887
‘bureau-maximizing’ behaviour 313
China 317, 721–2, 727, 730
developmental state 325
elite bureaucracy 325, 720
industrial policy and 39, 51, 797
parastatal bureaucracy 884
Weberian bureaucracy 318
business models
modularity 168–9
open innovation 169
platforms 169
business-school perspective 154
buybacks 584
C
Cambodia 137
cameralism 514, 521
capabilities
Africa, capability gap 847–8
capabilities accumulation 117
capability-based view/Beijing–Seoul–Tokyo (BeST) consensus 662, 667
capability failure 661, 673, 731
China, government and enterprise capability and learning 730–2, 742
developing countries 661, 731
dynamic capabilities 325–6
firms, organizational capabilities 151, 153, 154, 170, 379, 739
importance of 661, 813
knowledge-based capabilities 442n.21
mayor contributions to the capability perspective 588–90
organizational capabilities 112, 150, 571, 572
organizational capabilities in public sector 322, 324–6 see also the entries below for
capability; production capabilities; technological capabilities
capability building 164, 717, 722, 730, 731, 742, 750, 752, 843
Africa 843, 861
firm level 845
GVCs 862
as path dependent 759
South Korea 662, 665, 666–7
capability development 11, 29, 30, 31, 35
firms 151, 153, 154, 170
(p. 909) capability triad 31, 35, 156, 562–3, 585–8
capacity
clean energy investment project: domestic resource capacities 419–22, 423, 425
public-sector capacities 325 see also absorptive capacity
capital 101, 337
capital accumulation 183, 184, 186, 196, 289, 299
capital-intensive industries 246
Page 9 of 79
General Index
‘patient capital’ 79, 82, 155, 363

capital controls 187, 214nn.19–20, 341, 850, 852
capital intensity 22, 102, 304, 433
capital-intensive sectors 192, 246, 253, 341, 665, 719, 817, 861, 862
capital market 81, 83, 464, 584, 603
capitalism 31, 41, 271
capitalist accumulation 179, 182, 183, 184
capitalist development 18, 186–7, 288
capitalist industrial development 184, 190, 198
competition 184
competitive capitalism 154
co-operative capitalism 154
digital capitalism 303
drivers of 29
end to 189
firms and characterization of different capitalisms 154, 157
global capitalism 527
hierarchical capitalism 257
industrial capitalism 30, 154, 197, 266, 285, 303, 304
industrial policy and 17–18
labour and 184
as mixed economy 38–9
modern capitalism 6–7
nation-based capitalism 527
origins in Italian city states 540n.54
radical industrial policy on 184–6, 190, 197–8, 199
rentier capitalism 558, 562
state intervention 38–9 see also state capitalism
CARs (capital account regulations) 81, 85–7, 459–60, 479
catch-up 31–4, 759, 773, 811
BRICS 760, [link] , 773
catch-up paradoxes 33
China 96, 98, 101, 104, 109, 111, 266, 720, 731, 735, 753, 773
developing countries 96
duty to catch up 317
East Asia 266
economic catch-up 32
from middle-income to high-income levels 33
Germany 317–18, 340
industrial policy 6, 18, 19, 28–9, 43–4, 118
industrial policy as driver of 10–11
innovation and 114, 683
investment 340
key aspects of 32
labour and 101, 187
late industrialization 32–3
learning and 32, 35
Page 10 of 79
General Index
private sector 150, 170

Russia 340
South Korea 662, 669, 671, 683, 762
state intervention 32, 38
structural transformation and 111–14
successful catching up 98
UNCTAD on 33–4
causality 75, 109, 538–40
CDB (China Development Bank) 82, 272, 275, 338, 346, 350, 363, 727, 729 see also China: fi
nance; development banking
Cecchini Report (1988) 546
Central Asia 805
Central Europe 252, 434, 602, 781, 783, 806
CEOs (chief executive officers) 584, 732
Chenery hypothesis 780, 789
Chile 187
CORFO (Chilean development agency) 82
China 5, 7, 25, 80, 191
11th Five-Year Plan 113
bureaucracy 317, 721–2, 727, 730
capital/output ratios compared with Italy and France 101, 102–103
(p. 910) catch-up 96, 98, 101, 104, 109, 111, 266, 720, 731, 735, 753, 773
digital technologies 382, 753, 765–6

ELI 761
exchange rate 451, 805
exports 797–8, 805
FDI 26
free trade 549, 803
‘frontier technologies’ 104
globalization 294–5, 296
industrial policies in a Sino-centric world 118–19
innovation 36, 52, 610, 761–2, 769–70
leapfrogging 276
macroeconomy 451
‘Made in China 2025’ 622, 623
manufacturing 758, 773, 783, 827n.15
manufacturing value added 105, 106, 609–10, 760–1, 762, 784
neo-liberalism 719
OBOR (One Belt, One Road) Initiative 622, 623
post-1978 approach to industrial policy 20
reforms 716, 718, 722, 724, 727, 730, 731, 734–5, 753, 782, 805
rise of 199, 209, 220, 718
shift-and-share decomposition methodology 106, 119–20
shipbuilding 113
SOEs 105, 256, 751
state 151, 163, 782–3, 827
structural change 105–106, 109
Page 11 of 79
General Index
subsidies 668, 733
successful latecomer industrialization 112, 451, 622
technological asymmetries 104
technology 98, 739, 753, 761–2
TNCs 213n.17, 722, 735
within-sector learning vs. structural change in productivity growth 106, 107–108
WTO 109n.3, 451, 609, 719, 727, 728, 735, 739, 803 see also the entries below for China;
BRICS
China: finance 348, 350, 363
Agricultural Development Bank of China 350, 727
banking reforms/restructuring 727–8, 731, 734
Big Five 727–9, 738
Big Four AMCs (asset management companies) 727–8, 729, 734
Central Huijin Investment Corporation 728, 730, 732
commercial banks 727, 728, 734
Export–Import Bank of China 350, 727, 729
Ministry of Finance 727, 728, 732, 734
policy banks 350, 727, 728, 729, 732, 742
rise of China’s national champions in finance 727–9 see also CDB; China: national champi
ons; SASAC
China: international dimension
Africa 771–2, 846, 850, 860, 867, 883
ASEAN 209n.5, 210, 212, 216, 220–1
BRICS 749, 757, 760, 770, 772
EU 609, 617, 622, 623, 630, 650
Germany and 612
Japan/South Korea models and 720–2
Latin America 816–17, 823, 836
US/China trade issues 546
China: labour
labour productivity and capital/labour ratios 101, [link]
labour productivity compared with Italy and France 98–9, [link] , 101
labour productivity growth 98, 99
labour productivity of exporters and non-exporters 109, [link]
labour productivity ratio 104–105, 105
skilled labour, rising costs of 300
wages 474n.16
China: national champions 36, 716–17, 739, 742
arm’s-length type state‐business relations 716, 718, 719
CCP (Chinese Communist Party) 720, 721, 722, 729, 730–1, 732–3, 734
centrally managed capitalism 720
criticism of national champions policy 719, 734
(p. 911) diversification 726
enterprise capability and learning 731–2, 742

Fortune Global 500: 716–17, 735, 736–7
government capability and learning 730–1, 742
industrial policy 717, 719
Page 12 of 79
General Index
institutional bridging 717, 732–3

international competitiveness 717, 723, 724, 726, 728, 734–9, 742
IPOs (initial public offerings) 725, 728
late industrialization and developmental state view 720–2
mainstream transition economics view 718–19
market-oriented environment 717, 727, 729, 742
mergers 726
‘national team’ 724–6, 738
networked hierarchy 717, 720, 732–3
privatization 719, 724
R&D intensity of China’s national champions 738–9, 740–1
reciprocal control mechanisms 717, 733–4
rise of China’s national champions 722–9, 739, 742
rise of China’s national champions in finance 727–9
rise of China’s national champions in industrial sectors 720, 723–6
SMEs 718
SOEs 716, 718–19
state‐business relations and 717, 729–34
state capitalism view 719–20
state-controlled corporations 163, 717, 722, 723, 724–5, 729, 731
tuolasi 723
zonggongsi 723, 724 see also China: finance; SASAC
China: renewable energy 267, 274–6, 279, 281, 652
LCRs (local content requirements) 273, 275
solar energy 275–6
wind power 273, 274–5, 281
WWS (water, wind, sun) 274, [link]
China Exim Bank 272
China Mobile 722, 725, 731, 735
circular economy 267, 269, 271, 277
China 267–8, 279
resource security 271 see also green industrial policy
city
administrative city 519
industrial city 517, 519–20, [link]
industrialization 598–9
manufacturing city 519–20, 537, 576, 577
sustained city growth 590
civil society 188, 286, 301, 315, 575, 766
classical political economy 17, 18, 27, 29, 184, 285
classical social theory 285
clean energy 398–407
clean renewable energy sources 396, 401, 405
CO2 emissions 188, 394–5, 398–9, 402, 404–405, 407, 414
geoengineering 400–401
global clean energy economy 396
natural gas 398–9
Page 13 of 79
General Index
nuclear energy 399–300 see also clean energy investment project; energy efficiency; renew
able energy
clean energy investment project 395
2050 global zero CO2 emissions 396, 397, 402, 405, 407–10, 422, 423
Brazil 419–20, 425, 426
clean renewable energy sources 407
costs 395, 407, 409–10, 410, 411, 422
developing countries 418–19
development banks 416–19
developmental impact on fossil-fuel exporters 398
domestic resource capacities 419–22, 423, 425
employment impacts of 397
financial policies 397, 413–19, 422, 425
fossil-fuel industry, contraction of 397, 421–2, 423
Germany 417, 419, 425
global model framework and calculations 408–10
increase in clean energy investments 395
Indonesia 419, 425
(p. 912) industrial policy 396–7, 411–13, 422–3
land use requirements for infrastructure 398

public/private investment 395, 418–19, 422, 425
South Africa 419, 420, 425
South Korea 419, 420, 425
thirty-year global GDP growth trajectory 407, 408–409, 409 see also clean energy; energy
efficiency; renewable energy
clientelism 690, 885
climate change 5, 7, 53, 311, 331
climate crisis 179, 188, 190, 199, 401n.8
CO2 emissions 188
fossil capital 188–9
IPCC’s climate stabilization goals 394, 422
radical industrial policy 188–90
radical state intervention 190 see also environmental issues
clusters 270, 520
Asia 46n.60
automotive industry 225
average wage per cluster category, Europe 532
benefits from 46–7
Botero and geographical clustering of economic activities 517
external linkages and 47n.64
GVCs 228–9
industrial city 517, 519–20, [link]
industrial district 47, 374, 492, 520, 587n.25, 603
industrial ecosystem and cluster dynamics 46–7
as post-Second World War phenomena 46
tech industry 229, 595
UNCTAD 47n.65
CNPC (China National Petroleum Corporation) 722, 725, 729, 731, 735
Page 14 of 79
General Index
CO2 emissions 188, 394–5, 398–9, 404–405, 414

2050 global zero CO2 emissions project 396, 397, 402, 405, 407–10, 422, 423
Colbertism 521, 601
Cold War 192, 212, 289, 300, 323, 568, 601, 602
Cold War economic theory 507–10
communism 508–10
free trade 507–509
neo-liberalism 507–10
Ricardo, David 508, 509, 535, 548
Colombia 812, 818, 821, 826–7, 831, 836
Bancóldex (foreign trade development bank) 819, 826
CONPES (National Council of Economic and Social Policy) 826
national development banks 819–20
trade liberalization 819–20, 826 see also Latin America
colonialism 158, 162, 213n.14, 267, 539–40
colonial economic policy 868
colonial industrial policy 868–9
neo-colonialism 535
Washington Consensus 535
commodities 76, 521
commodity dependence 76–9, 189
commodity dependence/development links 76–7
definition of 372
‘financialization’ of commodity markets 77
industrialization based on commodities 469
‘primary-commodity specialization trap’ 469, 471–3, [link] , 477
commodity boom 76, 78, 79, 85
Africa 469, 867
commodity-price boom 469, 470–1, [link]
Latin America 816, 823, 825, 827, 830, 836, 837
communism 212–13, 508–10, 535, 547, 601
China 719, 720
‘the irrational twins’ 510
USSR 720
comparative advantage 27, 216n.27, 237
CAD (comparative advantage defying) industrial strategy 28n.28, 790, 791
(p. 913) CAF (comparative advantage following) industrial strategy 28n.28, 790
‘comparative-advantage-defying’ measures 255, 469

dynamic comparative advantages 790
efficiency indicators 138
frequency of the term ‘comparative advantage’ [link]
harm from 241
industrial policy and 28, 135n.18, 216
Ricardo’s theory of 507, 508, 529, 540
specialization according to comparative advantage 240
Page 15 of 79
General Index
competition 135n.18, 313, 494, 595

capitalism and 184
classical competition 529, 530
collusive competition 529, 530
competition law 158, 160, 162
competitive capitalism 154
competitive discipline 160
domestic market 45n.57
EU, competition rules 622, 626–7, 628, 630–1, 644, 652
firms 157–9, 170
global competition 7, 52, 190
handicap competition 157
imperfect competition 159, 345, 374, 525, 529, [link] , 843
innovation-driven market competition 93
international markets 45
Japan, New Competition 570–1
LDCs 52
‘new competition’ 155–6
optimal competition 157, 159
perfect competition 157, 238, 374, 530, [link] , 533, 718, 803
performance competition 157
price-led competition 570, 589
product-led competition 570–1, 589
competitiveness 6, 19, 71, 311, 374
ASEAN 209, 221n.36
China: national champions 717, 723, 724, 726, 728, 734–9, 742
drivers of 29
EU 608, 630, 636, 651
firms 153, 157, 164, 595
industrial policy and 217
international competitiveness 35, 36n.47, 49, 75, 164
Latin America 821
complementarities 71–2, 101
dynamics of production structures 69
growth and 67
innovations/complementarities interplay 73–4
policy implications 83–4
consumer 525, 881
consumer choice 127–9, 167
consumption
FIR technologies and 487
innovation–production–consumption relation 488
core‐periphery relationship 21, 180, 227n.47, 242, 290
corporations
market power of 6
M-form corporation 96
US 154, 582–3, 584 see also MNCs; TNCs
corruption 131, 341, 885
Page 16 of 79
General Index
1907 US Tillman Act 582n.20

Africa 861, 862, 885
BNDES 359–60
China: national champions 734
corporations 582n.20
South East Asia 685
cost‐benefit analysis 129, 138–40, 145, 326, 327
Covid-19/Coronavirus pandemic 538, 586n.24
CPTPP (Comprehensive and Progressive Agreement for Trans-Pacific Partnership) 209, 219, 220
Crédit Mobilier 343
crises 190
1930s crisis 541
climate crisis 179, 188, 190, 199, 401n.8
Coronavirus crisis 538
debt crises 67n.9, 346, 846
financial crises 523–4 see also 1997 Asian financial crisis; 2008 global financial crisis
cronyism 130, 131, 136, 341, 349
cross-border flows 81, 289, 292–3
(p. 914) Cuba 191, 195, 196–7
cumulative causation 155, 182, 194, 452

export-led growth model with cumulative causation 456, [link]
currency see exchange rate; monetary policies
cyber-physical spaces/systems 487, 611, 631n.23, 648
D
3D-printing 166, 486, 489
debt 81–2
Africa 67n.9
austerity policies and 524
Brazil 355
debt crises 67n.9, 346, 846
debt-peonage system 563, 576
external debt 345n.9, 346, 465, 688–9, 691, 692
Latin America 67n.9, 346, 468, 816, 818–19, 836
public debt 330
unsustainable external debt 355, 461, 473, 818, 851
US 586n.24
de-globalization
de-globalization narrative 284, 286, 290–1, 293, 299
new technologies and 486, 490
signs of 486, 493–4, 499
de-industrialization 5–6, 178
Africa 67, 78
austerity and 469
EU 608, 621, 627–8
France 596
Indonesia 682, 701
Latin America 67, 78, 468–9, 632, 818, 825, 827, 832, 836
Page 17 of 79
General Index
neo-liberalism 558
post-communist countries 779, 783, 805
premature de-industrialization 5–6, 34, 52, 67, 243, 283, 437
Russia 779
South Africa 858, 861, 862
South East Asia 695–705
South East Asia, premature de-industrialization 681, 682, 705, 711
UK 596, 612, 613
US 556, 558, 568, 572, 581
Washington Institutions 549
Zimbabwe 538
demand/demand side 22, 180, 312, 329, 561, 574
‘big push’ industrialization 464
demand economics 579, 586
macroeconomic policy and industrialization 452, 453, 454, 456, 459–60, 464, 474–6, 479
wages and demand 474–6, 479
democracy 331
radical industrial policy 188, 190, 191, 193, 194, 197, 198, 199
social democracy 575
US 582
workplace democracy 193, 198
demographics 780
population density and industrialization 536–7
deregulation 126, 227, 285, 363
neo-liberalism 557–8
South East Asia 687, 694, 701, 703
South Korea 664
US 558, 561, 562, 583–4
developed countries see advanced countries
developing countries see emerging and developing countries
development 63, 314, 780–1
bottom-up/top-down approaches 791, 793
capitalist development 18, 186–7, 288
commodity dependence/development links 76–7
industrial policy and 5–6, 118, 811
technology and 22
as town/countryside synergy 520
trade/development links 237–8, 239–45, 254
Von Thünen’s model 520
(p. 915) development banking 272
bank-based industrial business groups 348–9

clean energy investments 416–19
development bank, definition 337n.1
Page 18 of 79
General Index
DFIs (development finance institutions) 337n.1

diversification of development financing 347–50, 361
economic rationale for 339–42
effectiveness of 338–9, 342–7, 349–50, 361–3
experimentation 349
facilitating risk-taking and learning 337, 338, 339
France 343, 347
gender equality and 447
Germany 347, 363, 605
government-led/government-regulated state development banking 341
green development banks 416, 417–18
Japan 347
‘life cycle’ of 338
macroeconomics 344, 345–7, 363
‘national champions’, financing of 362
neo-liberal criticism of 341
private development banking 345, 347–9, 354–6
reciprocal control mechanism 361
regional development banks 304, 344, 348
socializing risks 339, 343, 352
as SOEs 347
state ownership 347, 447
state-sponsored development banking 338
technological breakthroughs 341n.6
US 348 see also MDBs; NDBs; World Bank
developmental state 39, 40, 186–7, 191–2, 198, 257, 266, 317, 683
African developmental state and effective industrial policy 883–9
concept of 325
elite bureaucracy 325
late industrialization and developmental state view 720
public-sector capacities and capabilities 325
developmentalism 41, 287
dictatorship 509, 548, 549–50
diffusion-oriented policies 814
digital platforms 162, 166–8, 612
big data 166–7
competition enforcement 167–8
data ownership and control 165, 166, 168
EU 167, 168
industrial policy implications 168–70, 171
online search/social networks 162, 166
regulation of 171
technology giants 166–7
TNCs 165
digital production technologies 382–3, 765, 768
digital production technologies: challenges 384–6
infrastructure: basic and digital 385
production system: retrofitting and integration 385
Page 19 of 79
General Index
technology absorption, deployment, and technical change: ‘capability threshold’ 384

technology access and affordability: endogenous asymmetries 386
technology diffusion: 4IR islands and digital capability gap 371, 382–3, 385–6 see also digi
talization; Fourth Industrial Revolution
digitalization 45, 53, 299, 389, 768
China 382, 753, 765–6
concentration and 158, 165
digital capitalism 303
digital industrialization 386–9
digitalization of production 156, 162, 164–6, 171
economic power 167
EU 631, 648
global South 298, 299
globalization narratives and 297–9, 300–301, 304
industrial policy implications 168–70, 171
Russia 752
sectoral/cross-sectoral interventions 387–8
(p. 916) shortcomings 165 see also digital production technologies: challenges; Fourth In
dustrial Revolution
diversification 349, [link] , 544, 788–9
Africa 844, 846, 873
China: national champions 726
economic development and 243
importance of 243, 542
industrial policy and economic diversification 779–81
Latin America 816, 834
macroeconomic policy and industrialization 471, 473, 480
trade composition, structural transformation, and diversification 242–5, 260
US 546
Uzbekistan 793
domestic distortions theory 126–9
downsizing 178, 437
DRC (domestic resource cost) ratio 139–40, 143, 145
formula 139n.27
drones 486, 856
dualism 64, 72, 105
‘Dutch Disease’ 67, 79, 470–1, 792, 796, 806
South East Asia 686, 689, 711
E
East African Community 881
East Asia
catch-up 266
competitive discipline 160
ELI 41, 238, 788
export growth 68
Page 20 of 79
General Index
‘forced investment’ regime 465

high savings rates and coercive state power 346n.10
ICT 134
industrial policy success in 40–1, 257, 259, 270, 465, 788, 790
industrializing economies 4, 25, 34–5, 67
reciprocity 50
statist model of industrialization 191, 192, 193
East India Company 158
Eastern Europe 627, 628, 783, 805
ECLAC (United Nations Economic Commission for Latin America and the Caribbean) 21n.13, 64,
180, 818, 831
e-commerce 162, 166, 169, 215n.25, 219, 299, 751
BRICS 755, 772
economic activities
differences among 525, [link] , 527
economic harmony as man-made 548
‘good’/‘bad’ economic activities 532, 533
quality index of 530, [link] , 533, 538, 546
economic geography 517, 519, 520, 651
economic growth 4, 11, 17
1989 fall of Berlin Wall, economic growth since [link]
diversification and 243
emerging and developing countries 63, 78
exchange rate/economic growth link 68, 133
Indonesia 682, 700–701
industrial structure and 789
neoclassical approach to 125
staple theory of 789
state and 4, 885–6
structural transformation and 63, 66, 69, 83
technological change and 29–31, 93, 369
trade/economic growth links 241–2 see also growth
economic history 20, 39, 42, 43, 161
economic miracles 216, 563, 581, 585, 788, 789, 800, 806
economic viability 22, 49
economies of scale and scope 7, 26, 35, 67, 150, 152, 154, 166, 238, 602, 611, 724, 851
ECSC (European Coal and Steel Community) 602, 603, 624n.8
education 64, 134, 597
engineering education 573, 598, 756
Ethiopia 886
EU 636
gender educational gap 430, 432, 436, 443–4
Germany 600
‘horizontal’ education 117
(p. 917) industrial policy and 49, 595, 600
knowledge production 443
leapfrogging and 791
Sweden 600
Page 21 of 79
General Index
tertiary teaching 19, 708

universal basic education 600 see also training; universities
EEC (European Economic Community) 319, 527, 602, 603, 624, 626, 627
founding members of 624n.8 see also EU
efficiency 313
allocative efficiency 157, 326–7, 328
dynamic efficiency 11, 19, 23, 63, 69, 73, 83, 327–8, 476
Pareto efficiency 313, 314n.5
static efficiency 63
throughput efficiency 560, 570
efficiency indicators 138–40, 143
DRC (domestic resource cost) ratio 139–40, 139n.27, 143, 145
ERP (effective rate of protection) 139, 139n.26, 169n.8
EFRD (European Funds for Regional Development) 629, 633, 636n.34, 656
EIB (European Investment Bank) 338n.4, 346, 639, 640, 651
elastic factor supplies 67–8, 69
electric/electronics industry 37, 252, 382, 567
South East Asia 691, 692, 694, 696, 697, [link] , 699, [link] , 701, 702, 705, 709, 711
ELI (export-led industrialization) 25, 806, 843
ASEAN 213–14
China 761
East Asia 41, 238, 788
import substitution vs. export orientation 804–805
Indonesia 686, 690–1, 699
ISI/ELI relation 25, 42, 45
South East Asia 683, 690–6, 704, 705, 711 see also exports
Ellen Macarthur Foundation (UK) 278n.18
embeddedness
embedded autonomy 186, 318
embedded economies 315
embedded industries 486, 497, 498
embedded services 378
locally embedded firms 374
emerging and developing countries 5, 7–8, 13–14, 281
capabilities 661, 731
capability failure 661, 731
catch-up 96
clean energy investment project 418–19
constraints in 44
development banking 418
digital industrialization 299
diversification of production structures 83
economic growth in 63, 78
financing 81–2
green industrial policy 269
green investments in 418
GVCs 227–8, 245, 246–7, 251–2, 253, 258
innovation 85
Page 22 of 79
General Index
public, private, and gross investment in [link]

regional integration 85
RVCs and 259
trade 239, 255
employment 6
agriculture sector 434n.10
automotive industry 225, 226
clean energy investment project 397
gender and industrial policy 430, 431–8, 446
gender job segregation 430, 431–7, [link]
GVCs 229, 245
renewables-based industries 270
SMEs employment 222
structural transformation and 74
trade-off between labour productivity growth and employment growth 458–9, [link]
US manufacturing employment 555, 568, 574, 580, 581–2
emulation see imitation/emulation
energy
advanced economies, energy consumption in 404
financialization of 188–9
global energy consumption 402
(p. 918) green energy 270–1, 281 see also bioenergy; clean energy; clean energy investment
project; fossil fuels; nuclear energy; renewable energy

energy efficiency 401–405, 422
burning bonus 402
costs of efficiency gains 403–404, 407, 407, 424
costs reduction 401–402, 424
energy-efficiency standards 395, 401–402, 403, 407, 413, 422
energy intensity ratios 396, 402–403, 408–409, 423
governments and 411
rebound effects 404–405
US 402 see also clean energy; clean energy investment project
energy security 8, 266, 267
green industrial policy 268
engineering 558, 573, 581, 588n.27
engineering education 573–4, 598, 756
engineering industries 757, 833, [link]
production engineering 116, 558, 563, 570, 574, 589n.29, 665
reverse engineering 37
South Korea 665, 666
US 558, 563, 570, 573–4, 584, 586
England 512–13
import substitution 513
Tudor strategy 513, 522 see also UK
enterprises see firms; SMEs; SOEs
environmental issues 266, 549
China 266–7
environmental sustainability 20, 53–4, 388
Page 23 of 79
General Index
environmental taxes 127
EU 623, 631, 651–2
India 266, 267
pollution 266, 281 see also clean energy; clean energy investment project; climate change;
green industrial policy; Green New Deal; renewable energy
ERP (effective rate of protection) 139, 169n.8
formula 139n.26
ESIF (European Structural and Investment Funds) 622, 629–30, 635, 636, 638, 639–40, 642,
[link] , 644, 649, 651
assumption made for co-financing by member States of ESIF projects 656
categorization of thematic objectives in ESIF into industrial policies themes 656
CF (Cohesion Fund) 629n.19, 636n.34, 642
EAFRD (European Agricultural Fund for Rural Development) 629n.19, 645n.50
EFRD (European Fund for Regional Development) 629, 633, 636n.34, 642, 644, 645, 651,
656
EMFF (European Maritime & Fisheries Fund) 629n.19
ESF (European Social Fund) 629n.19, 636n.34, 656
EUInvest 639–40
‘Juncker Plan’ 639
mapping of thematic objectives in the ESIF to the industrial policy themes 656 see also EU
industrial policy
ESPRIT (European Strategic Programme on Research in Information Technology) 602, 629n.15
Ethiopia 135, 362, 881, 887, 888
Addis Ababa 10
apparel export industry 845, 851–5, 861
cement industry 845, 850–1, 862, 887
Development Bank of Ethiopia 852
diaspora bonds and Renaissance Dam project 350
gender job segregation 434
green industrial policy 276–7
GVCs 853, 855
industrial parks 854, 855
infrastructure and education 886
leapfrogging 277
learning process 855
TIDI (Textile Industry Development Institute) 852, 853, 855 see also sub-Saharan Africa
EU (European Union)
1970 Colonna Report 602
1977 Davignon Plan 603
euro 546, 548
globalization 292, 293, 608, 609, 617
(p. 919) member states and country groupings 654 see also EEC; EU industrial policy; EU
treaties
EU industrial policy 602–603, 608–609, 616–17, 621–2, 652–4
2008 global financial crisis 609, 621, 633, 639
2030 EU Industrial Policy 498
bilateral liberalization 632
CEF (Connecting Europe Facility) 639, 640, 656
Page 24 of 79
General Index
challenges 623, 647–52
China and 609, 617, 622, 623, 630, 650
competition rules 622, 626–7, 628, 630–1, 644, 652
competitiveness 608, 630, 636, 651
convergence/cohesion policy 623, 628, 650–1, 653–4
de-industrialization 608, 621, 627–8
digital industry policy 631, 648
digital platforms: competition enforcement 167, 168
education and training 636
emerging markets and 623, 649–50
environmental issues 623, 631, 651–2
European Common Market 626–7
European integration process and 602, 608
‘European paradox’ 608
Fourth Industrial Revolution 611–12, 617
green industrial policies 630n.21, 631, 636, 637, 640, 644, 653n.58
GVCs 211, 492, 632, 650
Horizon Europe/Horizon 2020 programme 311, 631, 637, 644–5, [link] , 648
industrial policy structure [link] , 630, 633
‘Industry 4.0’ 631, 648
infrastructure 636, 639–40, 644
innovation 602, 608, 623, 630, 631, 636, 637, 638, 642, 644, 647–9, 651, 653
Lisbon Agenda 623, 629
manufacturing value added 609, [link]
market-failure approach 622, 630
‘mission-oriented’ policies 320, 621, 631, 633, 645, 648, 651, 652, 653–4
mixed approach 621, 633
multilevel governance 617
national industrial policies 608, 622, 635
new industrial policy at the turn of the century 609–16
R&D 602, 608, 610, 628–9, 631, 633, 636, 637, 638, 642, 644, 648, 653
Single Market 546, 602, 622, 628, 631
SMEs 602, 636
SOEs 595, 601–602, 625
structural development issues 649–50
subsidies/state aid 622, 624n.4, 626n.13, 627, 628, 635–6, 637
technology 623, 626, 628, 629, 630, 636, 637, 638, 642, 644, 647–9, 653
TEN (Trans-European Network) policy 638–9, 644
TNCs regulation 163
US and 617, 626, 629, 632 see also the entries below for EU industrial policy; EU; European
Commission; Finland; France; Germany; Ireland; Italy; smart specialization; UK
EU industrial policy, evolution of 625
1950–80 interventionist phase 594–5, 596, 601–602, 624–7
1980–2005 liberal phase 594, 595, 608, 624, 627–30
2005–present: pragmatic phase 594, 595, 624, 630–3
EU and industrial policy funding 655–6 see also ESIF; EU industrial policy priorities; MFF
EU industrial policy priorities 633–4, 653
country dimension: differences in industrial policy expenditure 641–2, [link]
Page 25 of 79
General Index
country dimension: different priorities 642–7, [link]

industrial policy spending at EU and member-state levels 634–6, [link] , 652
SAS (State Aid Scoreboard) 634
thematic dimension 636–41, 637, [link] , [link] see also ESIF; MFF
EU treaties 629
Maastricht Treaty 602, 608, 610, 629
Rome Treaty 602
TFEU (Treaty on the Functioning of the European Union) 626n.13, 629n.16
(p. 920) EUREKA (European Research Coordination Programme) 602, 629n.15
Eurocentrism 515–16
Europe
1945–twenty-first-century industrial policy 601–609
boomeranged universalist projects of 547–8
war industries 601 see also EU
European Central Bank 523
Green Bond programmes 414, 415–16, 422
European Commission 608, 629, 633, 645, 652
2006 Global Europe strategy 632
2012 policy strategy 633
2014 European Industrial Renaissance 632
2017 policy strategy 633–4
2020 Industrial Policy Communication 630n.21
competition rules 622, 626–7
criteria for selecting missions 323
Siemens-Alstom merger 622n.3, 630 see also EU industrial policy
evolutionary economics 29, 64, 70, 93–120, 312, 843, 845
systemic approach 814
technical change 813
exchange rate 794, 814
Africa 878, 880
China 451, 805
commodity dependence and 77–8, 79
competitive exchange rate 68, 85–6
devaluation/undervaluation 432, 451n.1, 459, 467, 468, 479, 687, 688, 694, 792, 793, 794–5,
805, 806, 815, 878
euro 546
exchange rate/economic growth link 68, 133
exchange rate protection 128, 132–3, 794, 795
foreign exchange management 86–7
Latin America 812, 816, 817, 820, 821, 822, [link] , 824, 825, 836–7
overvaluation 12, 460n.6, 559, 568, 796, 815, 861
premature de-industrialization and 451–2
real exchange rate 68, 77, 85, 86, 87, 128, 814n.3
reserve accumulation 86, 794–6 see also monetary policies
exports 260, 270
benefits from exporting 237, 238
Page 26 of 79
General Index
China 797–8, 805
‘close’/‘distant’ goods 141–2
East Asia 68
export basket 141–2, 228, 244, 463
export discipline 50, 84
export growth 68
export-led growth model of economic growth 453–63, 480, 793–4
export performance 210, 214, 223, 344–5, 669, 758, 828
export sophistication/EXPY 141, 244, 251, 257–8, 260, 463, 780–1, 789, 797
export targets 141, 142n.34, 143
externalities 780
FDI and 25
growth and 24, 780
high-tech goods 781
income elasticity of exports and imports 461, 462, 463
Latin America 513–14, 816
learning 18, 25, 35, 109
post-communist countries 779–80, 781, 785–6, [link]
Russia 780
strategic role of 23–6
subsidies 127, 132, 142, 224, 353
Taiwan 697
Washington Consensus 314 see also ELI
external economies 64, 74, 156, 376, 589
externalities 313, 789–90
exports externalities 780, 791
(p. 921) industrial policy and 135n.18
informational externalities 133n.16, 134

F
Facebook 167
‘factory economies’/‘headquarters economies’ 252, 632, 650
FDI (foreign direct investment) 6
ASEAN 213, 218, 226
BRICS 751, 764–5, 764
China 26
domestic firms/FDI relationship 46
exports and 25
financialization of 80–1n.23
growth and 25, 800–801
GVCs and 6, 73, 229, 245, 246
Russia 752
Singapore 26
South East Asia 683, 711
Page 27 of 79
General Index
South Korea 25
feed-in tariffs 412
feminism 285, 438
finance 271, 361n.27
capitalism and 271
clean energy investment project 397, 413–19, 422, 425
financial globalization 81, 292
financial sector/real economy separation 522–3
green finance 270, 271–2
macroeconomics/finance/structural change links 79–83 see also China: finance; develop
ment banking
financialization 52, 156, 187, 189, 199, 346
of commodity markets 77
energy 188–9
of FDI 80–1n.23
Finland 606–607, 617
firm, theories of 150, 151–6
capabilities, development of 151, 153, 154, 170
capitalism and 154, 157
competitiveness of firms 153
firm’s ‘strategic dimensions’ 155
Germany 154
innovation 153
innovation and finance 155
innovative enterprise theory 155
institutional-based theories 152
learning process 153
market failures in 152
neoclassical theories 150, 157
organizational capabilities 151, 153, 154, 170
performance of firms 155–6
resource-based theory 151, 152, 153–6, 163–4, 170, 316
transaction-cost theories 150–1, 152, 163, 170
US 154
firms
common ownership 160
competition 157–8, 170
competition and productivity 158–9
competitiveness 153, 157, 164, 595
concentration of 156–7, 158, 170
development finance and 345
disciplining mechanisms 157
domestic firms/FDI relationship 46
domestic industrial policies and regulations 151
economic power of 159–60, 170
economic power, rents, and political settlement 151 160, 161–2
Page 28 of 79
General Index
innovation 153, 155, 158, 159, 575

internationalization of production 156, 162–4, 171, 485
knowledge expertise 6–7
large firms 150, 157, 160, 161, 170
large firms and technological capability 36n.47
large firms and wealth 154
large international firms 6, 162, 242, 245, 255
lead firms 151, 160, 162, 164, 228
lead firms in GVCs 245, 246, 252, 253
learning trap 847–8
lobbying 151, 158
(p. 922) market power, competition, and enterprise strategies 156–62
migration to lower-wage locations 135

monopoly 151, 158, 166
Pavitt’s taxonomy of industrial firms 373–4
performance of large firms 349–50
state and 157
value addition 162, 164 see also ASEAN: firms; digital platforms; MNCs; SMEs; SOEs; TNCs
First Industrial Revolution 29, 53, 97, 182, 298, 369, 383–4, 389, 599
British Industrial Revolution 486
policy instruments 598
productivity explosion 527, [link]
UK 599–600
fiscal policy 479, 814
fiscal austerity 460, 468, 479 see also taxation
‘flying geese’ paradigm 135, 259, 299–300, 650, 790, 827
Japan 666
Latin America 300n.12
food and beverage industry 137, 375–6, 686, 688, 875, 881
Ford Motor Company 218n.28, 586, 665
Fordism 374, 542, 544, 545
forerunners 32, 33, 52, 287
Fortune Global 500: 340
China 716–17, 735, 736
fossil fuels 188, 190, 266–7, 280–1, 398
carbon cap 412–13
carbon taxes 270n.5, 404, 412–13, 414, 416, 418–19n.31, 422
China 267, 274, 279
coal 188, 190, 267, 279, 398
consumption of 421–2, 426
contraction of industry 397, 421–2, 423
disadvantages 277–8
fracking 398
oil wars 278
reduction in consumption of 395–6, 412–13
subsidies, elimination of 416, 422
zero consumption of 395–6 see also energy
Fourth Industrial Revolution (4IR/FIR, digital economy) 4, 6, 53, 298, 384, 611
Page 29 of 79
General Index
4IR business model 304

4IR islands 385–6
EU 611–12, 617
FIR disruptive change 487, 611–12
global South and 297–8
innovation/production co-location 487, 490
innovation/production/consumption continuum 488
innovation/production re-coupling 487, 490
neo-liberal vision of 299
technological change and 486–90
WEF 298 see also BRICS and 4IR; digital production technologies: challenges; digitalization;
‘Industry 4.0’
France
AII (Agency for Industrial Innovation) 614–15
BPI France 615
colonial industrial policy 868
development banking 343, 347
FUI (Single Innovation Fund) 615–16
industrial policy 601, 602, 603–604, 614–16, 626
industrialization 601
nationalization 603–604
OSEO 615
planning culture 318
R&D 602, 604, 616
SMEs 615, 616
state intervention 601, 615, 644
subsidies 604
Fraunhofer Institutes 605, 613
free market 38, 217, 227n.45, 316, 495, 558, 566, 612 see also liberalization
free trade 53, 84, 508
automatic harmony derived from 549
benefits of 803–804
(p. 923) China 549, 803
criticism 240
free-trade imperialism 536, 538
neoclassical economics 125, 126
Singapore 549
Smith, Adam 316, 507–508
UK 112 see also laissez-faire; trade liberalization
French Revolution 548
Fürstenspiegel 514
G
garments and textiles 143n.36, 229, 376, 431
England 512–13
Ethiopia 845, 851–5, 861
Multi Fibre Arrangement 136
Page 30 of 79
General Index
South Africa 845, 858, 859–61, 862

South East Asia 691, 692, 696, 701, 703, 704, 711
sub-Saharan Africa 851–2
US 576, 578, 580, 582
Gates Foundation 320
GATT (General Agreement on Tariffs and Trade) 270n.5, 804, 870
gender/growth regimes 439–43, 441
BRICS 440–2, 441
exploitation regime 440, 442
mutual regime 440–1, 442n.20
time squeeze regime 440, 442, 445
wage squeeze regime 440, 442 see also social reproduction
gender and industrial policy 429–30, 446
employment 430, 431–8, 446
feminization of foreign exchange earnings 431
gender, diversity, and technical design 444–5
gender equality 430, 431, 440
gender equality feedback loops and industrial policy success 438–45
gender equality: impact of industrial policy and structural change on 430–8
gender-inclusive industrial policies 430, 438, 446
global feminization of labour 431
industrial upgrading 438, 442, 443–4, 446
innovation and technological change 430, 446
knowledge production 430, 443–4
labour productivity growth and gender 438–43
radical industrial policy 178, 179 see also gender/growth regimes; gender inequality; social
reproduction
gender inequality 179, 431, 436–8, 439, 443
de-feminization of labour 433, 434n.9, 436, 437
educational gap 430, 432, 436, 443–4
exclusion 436, 437, 443, 446
exploitation 436, 437, 442
gender division of labour 179, 431
gender stereotypes 432, 436, 437, 444n.24, 446
gender wage gap 433, 446
high-tech jobs 430, 443, 444, 446–7
low-wage female labour 179, 430, 431–2, 446
lower-quality occupations 431
macroeconomics 437
neoclassical theory 432–3
STEM programmes 443–4
stratification theory 436, 444n.24 see also gender and industrial policy; women
General Motors 586, 724
genomics 611, 707
geoengineering 400–401
CCS (carbon capture and sequestration) 400
SAI (stratospheric aerosol injections) 400–401
Page 31 of 79
General Index
Germany 154
automotive industry 225
banking system 605
catch-up 317–18, 340
China and 612
clean energy investment project 417, 419, 425
development banking 347, 363, 605
education 600
Energiewende (energy transition) 320, 323, 637, 642
(p. 924) fossil-fuel consumption 421, 426
industrial policy 550, 600, 604–606, 611–12, 613–14, 642, 788

‘Industry 4.0’ 297n.11, 605, 611, 612, 631n.23
innovation 605, 612
leapfrogging 114
manufacturing 609, 611
market-centric policymaking 557
Marshall Plan 536, 605
multilevel governance 605, 613
post-war economic miracle 587
R&D 112–13, 605
reunification 605
SMEs/Mittlestand 596, 605, 612
successful latecomer industrialization 112–13
System Althoff 318
universities 112, 318
US and 557, 559–60, 569 see also KfW Group
Ghana 887, 888
global governance 6, 289, 380, 770
global North 178, 193, 291, 300, 301, 445n.25
shift to the global South 225, 259, 288
global production 52, 246, 255
ASEAN 211n.11, 229–30
changing patterns of 52–3
‘de-territorialized’ production choices 485, 490
global production networks 12, 52, 162, 208, 209, 380
globalization of production 485, 487, 490 see also GVCs
global South 193, 285, 793–4
digital industrialization 298, 299
Fourth Industrial Revolution 297–8
knowledge production/industry linkages 301
shift from the global North to 225, 259, 288
globalization 162, 245, 595
Brazil 292, 295, 297
China 294–5, 296
critical approach to 485, 527, 759
dynamic global markets 84
EU and 292, 293, 608, 609, 617
Page 32 of 79
General Index
financial globalization 81, 292

global capitalism 527
global economy 6, 239
multi-polar global power structure 53
regionalism and 292
South Africa 295, 296–7, 303 see also de-globalization
globalization narratives 284
2008 global financial crisis 296
BRICS 284, 286–7, 292, 294–5, 296–7, 301, 303
critical political-economy approach 285
data 285, 286–7, 293
de-globalization narrative 284, 286, 290–1, 293, 299
DHL Global Connectedness Index 291, 292–3
digital industrialization 297–9, 300–301, 304
elite globalization narratives and industrial strategies 284, 291, 304
GDLP (global division of labour and power) 286, 288, 289, 291, 300, 302, 303, 304
globalization indices 284, 291–4
globalization as inexorable yet malleable 294–6, 303
hyper-globalization narrative 284, 286, 289–90, 291, 293, 299
keywords 284n.1, 288
KOF Globalization Index 291–2
language 285
leveraging storylines as policy space 295, 303–305
methodology 286–7
narrative traps 288, 294, 303
neo-liberalism 293, 295, 300
populism 286, 290–1, 295
protectionism 295, 299, 303
regionalization and global neo-liberalism 296–7, 303
storylines 284, 294–9, 303–305
theoretical perspectives 287–9
words, spoken and written 284, 285
goods 372
classification of 375
consumer goods 185, 668–9, 828, 870, 872, 874, 882, 886
(p. 925) public/private goods 315 see also commodities; product
Google 167–8, 329, 493, 566

governance
global governance 6, 289, 380, 770
good governance 39, 83, 521–2
multilevel governance 495, 605, 613, 617
government
BRICS, national governments 749, 751
capability triad and 587–8
entrepreneurial role 499, 800
good governance 39, 83, 521–2
government failure 313–14, 781–2
Page 33 of 79
General Index
government regulation 126, 575

industrial policy and 556, 597
information and 134
neoclassical economics: government interventions 131, 132, 134–5, 144
role in mixed economy 315, 317
technological upgrading and 134
US 559, 562, 565–8, 574 see also state
grand challenges 311–13, 315, 321, 323, 326, 328, 388
challenge-driven innovation policy 312, 317
challenge-driven policy frameworks 311, 326
challenge-led growth 312
directionality of growth 312, 314
‘heterodox’ policy approaches 314–15
market co-creating 312, 316, 326
market failures 312, 313–15, 322
market shaping 312, 313, 316, 317, 322, 327–8, 331
SDGs 311, 312, 313, 321
socio-economic challenges 312, 318
technical challenges 312 see also ‘mission-oriented’ policies; public value
Great Depression 341, 555n.1, 579
Greater Boston (US) 567, 568, 574, 579
Greece 523
green industrial policy 12, 20, 53–4, 430–1
China 267–8, 274–6, 281, 652
climate perspective/green growth perspective distinction 279, 280
decarbonization 268, 273, 279, 331
differential principle 279–80
energy security 268
Ethiopia 276–7
EU 630n.21, 631, 636, 637, 640, 644, 653n.58
green development banks 416, 417–18
green energy 270–1, 281
green finance 270, 271–2
green growth 267, 269, 272–3, 279, 280
green resources 270, 271
imitation and emulation 281
India 268, 272–4, 281
instruments of 270n.5
late latecomers and 276–7
low-carbon economy 268
resource security 268, 281
waste/wasting 267, 271, 279
zero growth 269, 277, 279, 280 see also clean energy investment project; circular economy;
environmental issues; Green New Deal; renewable energy
Green New Deal 189, 190, 396, 397, 398, 422, 431 see also clean energy investment project
growth
Page 34 of 79
General Index
BoP constraints and 25n.22

challenge-led growth 312
complementarities and 67
directionality of 312, 314
dynamic efficiency of economic structures 63, 83
exports and 24, 780
FDI and 25, 800–801
green growth 267, 269, 272–3, 279, 280
growth collapses 67, 345n.9, 354
growth patterns 65–8
human capital and 65
industrialization and 788
investment and 65
Kaldor’s growth laws 21–2, 45, 64, 181
Kaldor-Verdoorn function/Law 74–6, [link] , 456, 473, 813, 815
manufacturing and 21–3, 180–1, 789–90
outperformers 349–50
productivity/growth links 74–6
(p. 926) protection and 68, 804
rapid growth 585–6, 587

successful industrial policy and 457–8, [link] , 460, [link]
sustainable growth 312
sustained city growth 590
technology and 30n.31
trade liberalization/growth relation 68, 76, 240, 241, 243
Verdoorn’s law 21, 181, 525
World Bank on 662
zero growth 269, 277, 279, 280 see also development; economic growth; productivity
growth
GSP (Generalized System of Preferences) 690, 694, 703, 804
GVCs (global value chains) 6, 37, 162, 164, 238, 245, 380, 485, 487, 490
2008 global financial crisis 485, 492
advanced economies 247, 252, 253
benefits from 227–9, 245–6, 251
‘buyer-led’/‘producer-led’ GVCs 223–4
clusters 228–9
concept of 227
coupling, de-coupling, re-coupling 844, 862
developing countries 227–8, 245, 246–7, 251–2, 253, 258
employment 229, 245
FDI and 6, 73, 229, 245, 246
GDP 246
governance structure of 258, 491
ICT and 245
impact of 251–4
importance of 163
inequality and 163, 247
integration 246, 247, 252
Page 35 of 79
General Index
iPod value chain, uneven returns 253–4, 254

lead firms in 245, 246, 252, 253
lower-income countries 6
manufacturing hollowing out and 490–3
mapping GVCs 246–51
market liberalization and 245, 843
MNCs/MNEs and 52n.70, 238, 252–3, 256, 258, 490, 492, 494, 495
off-shoring 485, 490, 491–2
participation in 247–9, [link] , 252
rise of 52
RVCs and 259
shortcomings 228, 229, 238, 251–2, 258
structural transformation and 251, 253
system integrators 735, 858–9
TNCs 251
upgrading 246, 251–3, 257–8, 260, 491
upstreaming/downstreaming 247–50, [link] , [link]
value addition 164, 490–1
World System Theory 227n.47 see also ASEAN: GVCs; value chain
GVCs: specific countries/regions/regional organizations
Africa 861, 862, 883, 886
BRIC economies 250–1, [link]
China 251
Ethiopia 853, 855
EU 211, 492, 632, 650
H
Harvard Business School 512, 520
Heckscher-Ohlin-Samuelson theorems 111
human capacities 438–40, 447
women, social reproduction and development of 439, 440, 442, 443, 446
human capital 70n.13, 438
concept of 439n.15
growth and 65
industrial policy and 19, 49, 54, 600
South Korea 665
upgrading and 442n.21
Hyundai Motors (South Korea) 345, 527, 665, 668
I
ICT (information and communications technology) 23, 53, 114–15, 328, 375, 378
BRICS 298, 301, 755, 760, 767, 769, 771
East Asia 134
GVCs and 245
ICT-based economy 112, 116
ICT industry 301
India 752–3
(p. 927) internationalization of production 162
Kenya 857
women and 445n.25
Page 36 of 79
General Index
ideology 4, 18, 161

ASEAN 213
gender stereotypes 432
ideological hegemony 287
Latin America 822
Marshall Plan ideology 535
myths surrounding industrial policy 42–3
neo-liberalism 287
IEA (International Energy Agency) 396, 402, 407, 408n.17, 409, 410
IFC (International Finance Corporation) 350
IG Farben (Germany) 605
IMF (International Monetary Fund) 290, 476, 689, 690, 692, 694
Africa 880
capital market liberalization 81
industrial policy, importance of 788
structural adjustment programmes 468, 690, 692
‘The Return of the Policy That Shall Not Be Named’ 5, 17
Washington Institutions 522, 527, 535, 549–50
imitation/emulation 30, 114–15, 843
adaptive imitation 33, 34, 37
catch-up 33
China, from imitation to innovation 276, 281
England 513
imperialism 213n.13, 267, 524
free-trade imperialism 536, 538
imports
benefits from importing 237
England 512–13
import duties 112, 208, 215, 304, 686, 689, 794, 803, 887
import protection 127, 132
import tariffs 117, 128, 130, 132, 135n.20
income elasticity of exports and imports 461, 462, 463 see also ISI
income
developed/developing countries’ divergence of incomes 66
income linkages 244
per capita incomes 65–6, 93, 111
increasing returns 44, 153, 182, 271, 327, 345, 371, 510, 549, 589–90
clusters 46
dynamic increasing returns 22, 156, 181, 590, 683, 843
energy 190
increasing/diminishing returns dichotomy 523–5, 536, 549
industrial city and 517, 520
law of increasing return 589n.31
manufacturing 18, 22, 522, 537, 546
neoclassical economics 64, 181
Page 37 of 79
General Index
protection of 537
static increasing returns 22, 181 see also returns
India 168, 194, 216, 266, 267, 800
automobile sector 140n.28
development 272n.11
green industrial policy 268, 272–4, 281
ICT technologies 752–3
ISI 752
leapfrogging 753
manufacturing 752–3, 758
renewable energy 272–3, 418–19n.31
wage-goods inflation 467
WTO 273–4 see also BRICS
Indonesia 213n.14, 463, 681
1997 Asian financial crisis 214n.18, 690, 701
clean energy investment project 419, 425
economic growth 682, 700–701
ELI 686, 690–1, 699
(p. 928) fossil-fuel consumption 421, 426
IMF structural adjustment package 690

ISI 685, 686–7, 695
manufacturing 700–701
technological upgrading/STI infrastructure 695, 705–706 see also South East Asia
industrial district 47, 374, 376, 492, 520, 587n.25, 603
industrial ecosystem framework 371, 380–2, [link] , 388, 389
industrial planning 319
industrial policy
adaptability 18, 19, 43, 51
aim of 430
context and connections 11–12
core of 269–70, 787
criticism of 4
debates on 5, 6, 17, 28, 255
defensive/proactive industrial policies 597, 603, 616
effectiveness of 429, 673
failures/‘cabinet of errors’ 8–9
government and 556, 597
multiple policy domains of 750
myths surrounding industrial policy 42–3
pragmatism 20, 34, 480
‘production/productive transformation policies’ 9
Page 38 of 79
General Index
relevance of 383–4, 529, 556, 788

role of 114–19, 217, 813
shifting terrain of 6–7, 19
successful industrial policy 145, 156, 216–17, 257, 388, 512, 563, 780, 805–806
taxonomy 597–8
unevenness 18, 51–2
vertical/horizontal industrial policies 255, 338, 362, 595, 602, 621–2, 811–12, 819 see the
entries below for industrial policy; radical industrial policy
industrial policy: challenges 5–6
competitiveness/discipline tension 50–1
de-industrialization 5–6
policy space restriction 6, 18, 19–20
prohibition of industrial policy by Washington Institutions 549–50
strategic compatibility/coherence tension 47–8
synergy and connections to other policies 48–50
industrial policy: history 3–5, 17, 494–5, 513–50
1980s–1990s industrial policy fatigue 621
historical roots of industrial policy theory 510–13
return/revival of 621, 661
industrial policy: phases since 1945
1950–80 interventionist phase 594–5, 596, 601–602, 624–7
1980–90s/2005 liberal phase 594, 595, 608, 624, 627–30
2000s/2005–present: pragmatic phase 594, 595, 624, 630–3
industrial policy: practice 43–51, 513
industrial ecosystem and cluster dynamics 46–7
instruments/tools of industrial policy 43–4, 47, 51, 598, 599, 673, 794–7, 806, 863, 870, 873–
4
linkage dynamics and interdependence 44–6
selecting what industries to support 788–94
selective/non-selective tools 794–5, 806
tensions in industrial policy 47–51
industrial policy: theoretical perspectives 3, 11, 19–28
concept of industrial policy 19–20, 216, 217, 429, 595, 596–7, 661, 786, 811, 867
conceptual perspectives on 17n.1
‘market-based’ approach 17n.1
pragmatic approach 20
‘promotional’ approach 17n.1
trial-and-error approach 20
industrial upgrading 33, 256, 429, 433, 436, 437
ASEAN 208, 209, 210, 222, 223, 228–9, 230, 231
gender and 438, 442, 443–4, 446 see also upgrading
industrialization 118, 266, 369, 600
agriculture 465, 542
BRICS 749, 755
(p. 929) China 268
growth and 788
importance of 429
India 268, 752
Page 39 of 79
General Index
industrial policy and 594, 598–601

industrial policy implemented by followers 600
‘industrialization of freshness’ 7, 23
low-income countries 5
Marxist approach to 182–3
population density and 536–7
stalled industrialization 34, 52, 437
thin industrialization 252
weak industrialization 243
wealth and 537, 598, 843 see also ‘big push’ industrialization; de-industrialization; ELI; First
Industrial Revolution; Fourth Industrial Revolution; ISI; late industrialization; Second Indus
trial Revolution; Third Industrial Revolution
‘Industry 4.0’ 487, 611–12, 766
EU 631, 648
Germany 297n.11, 605, 611, 612, 631n.23
Industry 4.0+ 487, [link] , 497
Italy 614
structural transformations 605
transformations brought about by 487, [link] , 496, 497
UK 613
inequality 157–8, 289, 331
deregulation 584
economic globalization 304
GVCs 163, 247
neo-liberalism and 548–9
TNCs 163 see also gender inequality
infant industry 112, 113, 216, 511
Africa 873–4
ASEAN 214
infant industry theory 17, 18, 27–8
necessity of nurturing infant industries 116–18
neoclassical economics 128–9, 128
South Korea 662, 667–9
subsidies for 350, 511
inflation 355, 814
hyperinflation 345, 347, 355–6
inflation tax 345, 355, 356
inflationary pressure 345, 355
wage-goods inflation 465–7, 473
information
government and 134
information asymmetry 31, 72, 152, 216
information flows 130, 292
informational externalities 133n.16, 134
infrastructure 49, 600
Africa 886
capability triad 586–7
Page 40 of 79
General Index
dynamic development process and 64

EU 636, 639–40, 644
Germany 587
infrastructural bottlenecks 385
innovation systems and 35–6
investment in 126, 499
new technologies and 499
innovation 18, 30, 69–71, 83, 595
catch-up and 114, 683
challenge-driven innovation policy 312, 317
complementarities/innovations interplay 73–4
definition 30, 30nn.32–3, 69, 132, 761
diffusion of 71
directionality of 321, 813
as driver of economic development 11
dynamics of production structures 69
firms 153, 155, 158, 159, 575
funding 80, 155, 328–9
gender and 430, 446
incentive to innovate 70
incremental and radical innovations 683, 684, 694
incremental/true innovations distinction 30n.32
innovation-driven economy 37, 49, 52
innovation-driven industrial policy 684, [link]
innovation-driven market competition 93
innovations/complementarities interplay 73–4
innovations for defence purpose and industrial applications 616
(p. 930) investment and 70
learning and 70
market failure perspective 31, 314, 320
mayor contributions to the innovation perspective 588–90
‘mission-oriented’ policies 321, 499
neoclassical economics 31, 132–4, 136, 144, 145
new technologies and innovation/production co-location 487, 490
new technologies and innovation/production/consumption continuum 488
new technologies and innovation/production re-coupling 487, 490
organizational innovation 30n.33, 114
path dependence 29, 67, 328
post-communist countries 797–8
private sector 314, 319
renewable energy 270–1
reward for 328–9
science and 29
state 132nn.12–13, 134, 314, 329–30
subsidies 132–3
taxation and 328
technological change 369, 381–2
Page 41 of 79
General Index
waves of innovation 64, 67 see also the entries below for innovation; STI
innovation: specific countries/regional organizations
ASEAN 209, 220, 222, 231
BRICS 761–2
China 36, 52, 610, 761–2, 769–70
EU 602, 608, 623, 630, 631, 636, 637, 638, 642, 644, 647–9, 651, 653
Finland 607
Germany 605, 612
Japan 113, 569
Singapore 708
South Korea 36, 669, 684
Taiwan 684
UK 604
US 564, 565, 572, 573
innovation systems 31, 35, 70, 113, 321, 327
collaboration among key economic players 37
infrastructure and 35–6
Japan 113
NISs/national innovation systems 35–6, 70
institutional economics 94, 154, 539–40
Intel 670, 701
Inter-American Development Bank 144
interest groups 313, 316
international trade see trade
Internet 411–12, 566
investment 313
‘big push’ investments 346
catch-up 340
fall in private investment 467
innovation 70
investment climate 130, 355
investment linkages 244
middle-income trap and 37
productive capacity, investment in 244, 439
technical change/investment interaction 22
technological learning and 35 see also clean energy investment project
IoT (Internet of things) technologies 165, 166, 298, 382, 384, 486, 487, 489, 497, 500, 611, 765,
769
IPCC (Intergovernmental Panel on Climate Change) 394, 401n.8, 422
IPRs (intellectual property rights) 70, 71, 238, 750, 843
ASEAN 219–20, 224
BRICS 755, 759, 768, 772
intellectual monopoly 220
Latin America 822
South East Asia 709, 710, [link] see also patents
Ireland 495, 607
IRENA (International Renewable Energy Agency) 405
ISDS (investor–state dispute settlement) 219, 220
Page 42 of 79
General Index
ISI (import-substitution industrialization) 3–4, 8, 25, 42, 213, 843

ASEAN 213, 220n.33
India 752
Indonesia 685, 686–7, 695
(p. 931) ISI/ELI relation 25, 42, 45
Latin America 257, 464, 812n.1, 816, 818

Nigeria 848–51
protection of increasing returns and 537
Russia 752
South East Asia 683, 685–90, 695, 710–11
structural transformation and 21 see also Africa: ISI; imports
Italy 542, 601, 603, 614
Industria 4.0 614
manufacturing 603
origins of capitalism in Italian city states 540n.54
populism 614
regional inequalities 614
SMEs 603
state intervention 603
J
Japan 17, 25, 66n.7, 843
automotive industry 138n.25, 225, 527
entry control 669
‘flying geese’ paradigm 666
Fukushima Daiichi power plant meltdown 400
indigenous innovation capabilities 113
industrial policy 138, 560, 569–70, 720, 721, 788
innovation system 113, 569
kaizen 560, 570
kereitsus 721, 722
manufacturing advantage 569–71
MITI (Ministry of International Trade and Industry) 318, 325
New Competition 570–1
technology 571, 572
US/Japan rivalry 557, 559–60, 568, 569–72
yen 568
Jesuits 515–16
just-in-time production 178, 376, 564, 570
K
Kaldor-Verdoorn function/Law 74–6, [link] , 456, 473, 813, 815
Kazakhstan 780, 781, 783, 785, 787, 790, 793, 805, 806
Kenya
Page 43 of 79
General Index
BPO (business-process outsourcing) sector 855, 857–8, 861, 862

knowledge-based service sector 845, 857–8
trade liberalization 878 see also sub-Saharan Africa
KfW Group (Kreditanstalt für Wiederaufbau, German Development Bank) 338, 343–4, 345, 346,
363, 605
clean energy economy 82, 417, 418
Kia 224
Timor–KIA aircraft manufacturing joint venture 687, 690
KLEMS database 136n.23
know-how 245, 300
commercial know-how 71
Germany 612
know-how transfer 26, 30
South Korea 663, 664–5, 666–7
technical know-how 70, 240
knowledge
abduction (phronesis) 511–12
gender and knowledge production 430, 443–4
knowledge absorption 599
knowledge accumulation 94, 111, 340
knowledge-based capabilities 442n.21
knowledge capital 70n.13
knowledge intensity 22, 23, 27, 461
knowledge monopoly 52
knowledge production/industry linkages 301
knowledge spillovers 73, 74, 328
knowledge transfer 24, 133, 213n.16, 302, 305, 604, 645n.50, 761
surveys and fact-finding 514
tacit knowledge 30, 133, 374, 667, 684, 691, 844, 857, 861, 862
technological knowledge 22n.15, 23n.19, 24n.21, 29–30, 71, 115, 117, 571, 761
Krupp (Germany) 345, 605
Kyrgyzstan 786, 805
L
(p. 932) labour
capitalism and 184
catch-up 101, 187
climate jobs 190
de-feminization of labour 433, 434n.9, 436, 437
developing countries and labour regulation 474
international labour migration 72
labour laws as industrial policy instruments 453, 473–8, 479–80
labour productivity 438
labour productivity growth and gender 438–43
‘perversity thesis’ 474, 478
radical industrial policy as labour-centred 179, 187–8, 191, 193–5, 197, 198
reallocation from low- to high-productivity sectors 66–7
Page 44 of 79
General Index
reallocation from traditional to modern sectors 67–8

socialization of labour 182
trade-off between labour productivity growth and employment growth 458–9, [link] see also
the entries below for labour; China: labour; wages; workers
labour, division of 179, 182, 242, 247, 370, 485, 598, 650
GDLP (global division of labour and power) 286, 288, 289, 291, 300, 302, 303, 304
gender division of labour 179, 431
IDL (international division of labour) 288, 300
NIDL (new international division of labour) 288
labour-intensive production 442n.20, 493, 561, 673, 719, 817
Africa 845, 846, 883
automation and 487
gender and 431, 432, 433, 434, 446
GVCs 225, 227, 231, 490
renewables industries 270
South East Asia 690, 691, 692, 694, 695, 701
US 578, 580
laissez-faire 31n.35, 38, 125, 126, 768
naive optimism of 510, 547
negative effects of 549
Physiocratic laissez-faire doctrine 548 see also free trade
late development 31–2, 187, 339–40, 343, 720
learning and 32, 34
NDBs and 337, 338, 340, 341
late industrialization
catch-up 32–3
challenges for 469, 733
disadvantages for 240
green industrial policy 276–7
late industrialization and developmental state view 720–2
latecomer industrialization 18, 20, 97, 112, 452, 594, 599, 681, 683
learning by doing 30
macroeconomic policy and late industrialization 453, 454, 464, 479–80
risks of late industrialization 337, 339–40
South East Asia 681, 682–3
Latin America 13–14, 116, 816, 836–7, 885n.6
1980s debt crisis 67n.9, 346, 468, 816, 818–19, 836
1998–2002 external crises 821–2
2000s timid return of industrial policy 816, 822–7, 836
2004–14 commodity boom 816, 823, 825, 827, 830, 836, 837
austerity 468–9, 818
China and 816–17, 823, 836
competitiveness 821
convergence and divergence, phases of 829–31
de-industrialization 67, 78, 468–9, 632, 818, 825, 827, 832, 836
Development Bank of Latin America 418
diversification 816, 834
Page 45 of 79
General Index
economic integration 820, 822

exports 513–14, 816
‘flying geese’ paradigm 300n.12
(p. 933) from state-led industrialization to debt crisis 817–19
from state-led industrialization to Washington Consensus 815–17

hierarchical capitalism 257
ideology 822
industrial policy weak performance 40–1, 257
investment ration and capital account openness 820, [link]
IPRs 822
ISI 257, 464, 812n.1, 816, 818
lagging behind East Asia and the world economy 812
local industrial policies 540n.53
macroeconomic policies 452, 816, 818, 823–4, 827, 836
manufacturing 827, 831–4, [link] , [link] , [link]
middle-income trap 109
neoclassical economics 812
NIS5 (National System of Innovation) 835, [link]
privatization 819
R&D 822, 823, [link] , 836
real exchange rate 812, 816, 817, 820, 821, 822, [link] , 824, 825, 836–7
reciprocal control mechanisms 836
relative productivity vs. United States 829–31, [link]
reprimarization of export structure 816–17, 836
South Korea as benchmark 823, 827–9, [link] , 831, 833–6, [link]
state failure 817
state-led industrialization 812, 816, 817, 836
structural adjustment programmes 468
structural change and convergence 831–6
structuralism 64, 66, 72, 812
trade liberalization 819–20, 825, 826
Washington Consensus years 819–22, 836
WTO 822 see also Argentina; Brazil; Colombia; Mexico
latitude for performance 41, 44, 50
LDCs (least developed countries) 339, 343, 354, 358, 362
competition 52
constraints and 19n.7, 20
NDBs 338, 348
leapfrogging 33, 37, 114–15, 277, 281, 287, 320, 759
China 276
education and 791
Germany 114
India 753
public–private joint R&D 670–1
South Korea 671–2, 683
learning 34n.41
Page 46 of 79
General Index
catch-up 32, 35
China, government and enterprise capability and learning 730–2, 742
cross-sectoral learning 312
as driver of economic development 11
economic catch-up and late development 32
exports 18, 25, 35, 109
innovation 70
international learning 18, 24, 25, 26, 35, 45, 50
late development 32, 34
learning capabilities 36n.45
learning curves [link] , 529
‘learning by exporting’ 109
as local and cumulative 95, 813
policy learning 20, 34
sources of 37
sub-Saharan Africa, learning costs/investment 843–4, 853–4, 855
technological learning 18, 26, 28–9, 34–5, 36–7, 96–7, 115–16
learning by doing 20, 22, 33, 34, 48, 71, 113, 182, 430, 436, 813
advanced countries 340
latecomer economies 30
neoclassical economics 128n.6
liberalization 285, 494, 716
1980s ‘business environment reform’ 126, 130
1980s–1990s structural adjustment era 130
Africa 874, 877–80, 889
(p. 934) ‘asymmetric opening’ 668
capital market liberalization 81

Washington Consensus 314 see also free market; free trade; trade liberalization
licensing 26, 37, 388
linear economy 271, 277, 278
linkages 44, 185
agriculture/manufacturing linkage 7, 23, 45, 377, 465, 517, 520
backward/forward linkages 64, 71, 83–4, 184–5, 188, 191, 244, 247, 251, 377, 430, 835, 881
income linkages 244
industry‐university linkages 37, 301
intersectoral linkages 23, 133, 239, 244, 377
investment linkages 244
knowledge production/industry linkages 301–302
linkage dynamics and interdependence 27, 44–6
local economies 25, 213, 220, 253
location theory 517, 520
M
McKinsey & Company 402, 403, 404
McKinsey Global Institute 349–50
macroeconomy 11, 19, 49
‘big push’ industrialization 463–9
BoP constraint 452, 460–3, 464–7, [link] , [link] , 471, 473, 477, 480
Page 47 of 79
General Index
CARs (capital account regulations) 459–60, 479

commodity dependence 77–8, 79
commodity-price boom 469, 470–1, [link]
cumulative causation 452
demand-side 452, 453, 454, 456, 459–60, 464, 474–6, 479
development banking 344, 345–7, 363
diversification and upgrading 471, 473, 480
drop in domestic expenditure growth 467
‘Dutch Disease’ 470–1
ELI strategies 469–73
export-led growth model of economic growth 453–63
forced savings, restrictive monetary policy, destabilizing austerity 468–9
gender inequality 437
growth and successful industrial policy 457–8, [link] , 460, [link]
Kaldorian open-economy growth model 452
labour laws as industrial policy instruments 453, 473–8, 479–80
late industrializing economy 453, 454, 464, 479–80
macroeconomic crises 24, 353
macroeconomic instability 64
macroeconomic mismanagement 452, 460
macroeconomic policies 19, 49, 66, 85–6, 345, 439, 441, 451–2, 814–15
macroeconomic stability 312
macroeconomics/finance/structural change links 79–83
macro-policy lessons for industrialization 478–80
stabilization policies 557, 560, 561–2, 565, 587, 597, 609, 818, 880
structural change and 69
supply-side 452, 453, 456
‘technology-forcing’ labour-market regulation 476–7, 479–80
trade-off between labour productivity growth and employment growth 458–9, [link]
‘wage-goods inflation’ barrier 465–7
wages: impact on economic growth of a one percentage point increase in real wage growth
474–6, 475
wages: share in the gross output price 474–5, 475
(p. 935) wages and demand 474–6, 479
wages and profit rate 477–8, [link]

macroeconomy: specific countries/regions
Africa 452, 880
China 451
Latin America 452, 816, 818, 823–4, 827, 836
US 560, 568
Madagascar 259, 852, 875
Malaysia 214, 681
Page 48 of 79
General Index

economic growth 682, 701–702
ELI 690, 691–2, 702
ethnically coloured policies 691–2, 711
FTZs (free-trade zones) 687, 691
industrial digitalization 300
ISI 685, 687–8, 691, 695
technological upgrading/STI infrastructure 687–8, 691, 695, 705, 706–707, 711 see also
South East Asia
manufacturing 5–6, 18, 270, 843
agriculture/manufacturing linkage 7, 23, 45, 377, 465, 517, 520, 883
decline of manufacturing relative to service industries 303, 789
growth and 21–3, 180–1, 789–90
GVCs, manufacturing hollowing out and 490–3
increasing returns 18, 22, 522, 537, 546
intersectoral linkages 23
manufacturing city 519–20, 537, 576, 577
Marxist approach to 182–3
post-communist countries 783, [link]
renewed interest in 492–3
‘servicification’ 7, 23, 377–8, 882
structuralism 21, 180–1 see also the entries below for manufacturing
manufacturing: specific countries/regions/regional organizations
Africa 868–870, 871, 875, 879, 883, 888
BRICS 756, 757, 758, 762–3, 763, 772
China 758, 773, 783, 827n.15
Germany 609, 611
India 752–3, 758
Indonesia 700–701
Italy 603
Latin America 827, 831–4, [link] , [link] , [link]
Malaysia 701–702
the Philippines 702–703
Russia 779, 783
Singapore 695–6, 703
South East Asia 695–6, [link] , 710
Thailand 704
UK 512–13, 612–13
US 534, 562, 564, 573–4, 576, 578, 584
Vietnam 783
manufacturing value added
Africa 870, 871, 875, 880, 887, 888
BRICS 762
Page 49 of 79
General Index
China 105, 106, 609–10, 760–1, 762, 784

EU 609, [link]
post-communist countries 783–4, [link]
South East Asia [link] , 701, 704, 711
market
frontier markets 81
market-centric policymaking 556, 557, 586
market co-creating 312, 316, 326, 341
market shaping 312, 313, 316, 317, 322, 327–8, 331
market socialism 195
market unification 600
neoclassical economics 125
self-regulating markets 38
social market economy 510 see also free market
market failure 11, 38, 313, 556n.3, 780, 799
EU, market-failure approach 622, 630
firm, theories of 152
‘government failure’ 313–14
grand challenges 312, 313–15, 322
innovation and 31, 314, 320
key aspects of 126, 313–14
market failure paradigm 31, 661
neoclassical economics and 125–6, 127, 131, 134, 135, 136, 138, 144, 145
policy intervention 314n.5
structural economic change 314
(p. 936) market fundamentalism 31n.35, 38, 572, 575, 583–4, 880
market imperfections see market failure

Marshall Plan 511, 525, 535–6, 549, 601
Marxism 39
approach to manufacturing and industrialization 182–3
‘micro-Marxism’ 185
state and 509
mass production 545, 601, 611
US 541–2, 544, 560, 573
materialism 285
MDBs (multilateral development banks) 80, 82, 83, 85, 344 see also development banking
MENA (Middle East and North Africa) 434
mercantilism 514, 521, 795
MERCOSUR (South American trade bloc) 211, 820, 822
mergers
China 726
digital platforms 167–8
meritocracy 317–18, 325
Mesmerism, science of 548
meso-economic economies/processes 69, 71, 72, 73, 75, 84
Mexico 252, 539–40, 650, 812, 825–6, 831, 835, 836
Page 50 of 79
General Index
1994 Tequila crisis 821

convergence 817–18
trade liberalization 819, 825 see also Latin America
MFF (Multi-Annual Financial Framework, EU) 636, 639, 645, 656
EU industrial policy priorities 634, 636
industrial policy item within the MFF 2014–20 655
mapping of budget items in the MFF 2014–20 to the industrial policy themes 655 see also
EU industrial policy
microeconomic theory 313
Microsoft 544, 607, 671
middle-income trap 29, 33–4, 52, 66, 109, 111, 674, 721
4IR 765
innovation capabilities and 759
investment and 37
Latin America 109
South East Asia 682
World Bank on 33n.38
Middle Way 510
migration 373, 522, 536–7, 538, 549
‘flying geese’ pattern of migration, East Asia 135
international labour migration 72
US Great Migration 577
military issues
global military spending 415
innovations for defence purposes and industrial applications 616
military industrial policy 4, 318
US military spending 415, 559, 567, 578–9, 580
‘mission-oriented’ policies 312–13, 316, 331, 341, 388, 389, 683, 814
assessment and evaluation 322, 326–8
directionality of 321
EU 320, 621, 631, 633, 645, 648, 651, 652, 653–4
European Commission, criteria for selecting missions 323
evolution of 317–21
experimentalism 320
innovation 321, 499
instrumentalizing missions as industrial policy 321–2
organizational capabilities in public sector 322, 324–6
research 318–20
risks and rewards 322, 328–31
ROAR 322
routes and directionality 322–4
Singapore 693, 705, 711
technological missions 321, 328 see also grand challenges; public value
mixed economy 38–9, 583
government’s role in 315, 317
(p. 937) MNCs/MNEs (multinational corporations/enterprises) 52n.70, 53, 227
Page 51 of 79
General Index
GVCs 52n.70, 238, 252–3, 256, 258, 490, 492, 494, 495

location choices of 490
‘multinational’ term 162
modernization 198, 311, 317, 319, 826
agrarian modernization 479n.19
Brazil 351
BRICS 759–60
China 730
economic modernization 212, 759
Mondragon Group (Spain) 193–4
monetary policy 214n.19, 415, 446, 523, 814
countercyclical monetary policy 78, 87
monetary policy autonomy 479, 796
restrictive monetary policy 12, 468–9, 824, 829
South East Asia 694
South Korea 683–4, 812
Taiwan 683–4
Uzbekistan 792, 793 see also exchange rate
monopolies 313, 494, 668, 669
agriculture 541–2
firms 151, 158, 166
global monopolization 288
intellectual monopoly 220
regulation of 126
taxation and 316
Morocco 889
multilateralism 53
multilateral trade 19–20, 52, 219, 240, 804
N
NAFINSA (Mexico) 83
NAFTA (North American Free Trade Agreement) 211, 219, 650n.57, 819, 825, 835
NAICS (North American Industry Classification System) 7
nanotechnology 79, 321, 328, 341n.6, 611, 707, 709, 769, 823, 824
NAS (US Academy of Sciences) 402, 403, 404
NASA 329, 578, 580
nation
industrialization and 594
nation-based capitalism 527
national champions 494–5, 596
BNDES, Brazil 345, 351, 352–3, 354, 356–7, 358
BRICS 751
development banking and 362
EU 602, 626
South Korea 36 see also China: national champions; ‘picking winners’
national interests 212, 219, 317–18
nationalism 43, 286, 290, 721
nationalization of industries 187
Page 52 of 79
General Index
France 603–604
natural resources 68, 72, 79
industrial policy in resource rich countries 796–7, 806
‘resource curse’ 398, 471, 796
resource-ialization 783, 805
resource security 266–7, 268, 271, 281
water security 277
NDBs (national development banks) 79, 83, 84, 85, 337
assets of 338n.2, 338n.4, 346
co-financing 339, 341, 344, 348, 361
countercyclical financing 82, 344
diversified/multiple financing model 339, 348–50
enhancing firm capacities and state bureaucracy capacities 339
late development 340
LDCs 338, 348
loans 83, 338
long-term finance 337, 338, 340–1, 342, 344
main functions 82, 85, 337, 338n.5
relevance of 338
risks 337, 340, 342, 362
sectoral technical expertise 339, 342–3, 361
single/limited long-run financing model 339, 348, 354
‘targeted ambition’ 363
targeting sectors/projects that generate net foreign exchange earnings 362
as venture capitalists and ‘lenders of first resort’ 362 see also Brazil, development banking;
China: finance; development banking; EIB; KfW Group
(p. 938) neoclassical economics 64, 70n.13, 312, 313
1980s ‘business environment reform’ 126, 130, 133

1980s–1990s structural adjustment era 129–31, 145
2000s neoclassical revival of industrial policy 131–4, 144
Cold War economic theory 507–10
criticism of 133–4, 145
domestic distortions theory 126–9
efficiency indicators 138–40, 143
equilibrium 125
firm, theories of 150, 157
free trade 125, 126
gender inequality 432–3
government interventions 131, 132, 134–5, 144
increasing returns 64, 181
industrial policy 4, 43, 125–45, 811–12
infant industry policy and 128–9, 128
innovation 31, 132–4, 136, 144, 145
interventionist industrial policy 126, 127, 131, 132, 134–5
interventionist vertical/horizontal measures 136
Latin America 812
market failure and 125–6, 127, 131, 134, 135, 136, 138, 144, 145
‘modern industrial policy’ 131
Page 53 of 79
General Index
monopoly of 550
policy hierarchy 11, 127–9, 128, 130, 131, 132, 138, 144
price system 125, 134, 145, 508–509
product discovery 131, 133
product space 140–4, 144n.39
production subsidy 127, 128
selectivity 136–44, 145
state and 509
neo-liberalism 227, 287, 584–5
China 719
Cold War 507–10
criticism of 341–2, 510, 584–5
de-industrialization and 558
deregulation 557–8
exports 23
Fourth Industrial Revolution 299
free market economy 227n.45
globalization narratives 293, 295, 300
inequality and 548–9
‘the irrational twins’ 510, 547
state and 509
tax reductions on corporations and the wealthy 557
trade liberalization 23
UK 4, 612, 627
US 4, 557–8, 562, 572, 582–5, 586 see also Washington Consensus
neo-patrimonialism 885, 889
the Netherlands 292, 293
New Deal 561, 562, 568, 577, 581
failure of 586
infrastructural investments in the South 577–8 see also US
New Economics 579
New Public Management reforms 320
New Structural Economics 134–5, 452
n-gram technology 508, 509, [link]
nichification 541–5
Nigeria 869, 881, 886
backward integration 849–50
cement industry 845, 848–51, 862, 887
ISI 848–51
privatization 849 see also sub-Saharan Africa
NISs (national innovation systems) 35–6, 644, 649, 835, [link]
Nokia 607, 617, 791
non-tariff barriers 128, 129, 130, 216, 688, 697, 804, 818, 819
ASEAN 208n.4, 215, 223, 224, 230
non-tradable inputs/activities 72, 84, 246n.3, 420, 423, 814–15, 844, 845, 857, 861, 862
Page 54 of 79
General Index
nuclear energy 395, 396, 399–400, 401, 407, 601, 824

O
OBM (own brand manufacturer) 37
ODM (own design and manufacturer) 37
(p. 939) OECD (Organisation for Economic Co-operation and Development) countries 340, 346,
417–18
OEM (original equipment manufacturer) 37, 386, 858–9
off-shoring 178, 227, 485, 491, 492, 596
GVCs 485, 490, 491–2
receding of 493, 494
OLI (ownership, location and internalization) factors 163
Oman 189, 781n.1, 789
‘one-size-fits-all’ approach 229, 479
opportunity gain 142n.34
outsourcing 23, 164, 178, 246, 249, 377, 843
P
Pacific Alliance 822
paradigm 97
technological paradigms and trajectories 94–5
theory of production 98
patents 36, 328
China 610
EU 610
foreign patents 666
generation of 667
patent applications 349, 610, 666–7, 761, 765
patent system 97, 219, 224, 664
South Korea 664–5, 666–7
triadic patent family 610n.6
US 610 see also IPRs
path dependency 41, 155
technological changes 30
Peru 539–40
Petrobras 352n.15, 353n.19, 357, 359, 360
PetroChina 725
the Philippines 681, 688
economic growth 682, 703, 707
ELI 692
IMF structural adjustment package 692
ISI 685, 688–9, 692, 695
overseas remittances 689
technological upgrading/STI infrastructure 688, 692, 695, 707–708 see also South East Asia
Physiocrats 548
‘picking winners’ 237, 314, 321, 325, 661–2
BNDES 357
Page 55 of 79
General Index
as common misconception of industrial policy 42

interventionist governments and 495, 595
vertical industrial policy 812 see also national champions
place-based approach 495–6
place-based policy 496, 497–8, 633, 647
place-based S3: 647 see also transformative industrial policy
policies
‘capital policy sins’ 118, 119
challenge-driven policy frameworks 311
‘heterodox’ policy approaches 314–15
horizontal policies 84, 255, 312
policy capture 136
policy targets 256, 314
selective policies 84, 85
vertical policies 169, 255, 312
policy hierarchy 11, 127–9, 130, 131–2, 138, 144
industry support by alternative measures 128
limitations 129
policy learning 20, 34
policy space 6, 84
globalization narratives: leveraging storylines as policy space 295, 303–305
national policy space 295
neo-liberalism and 300
policymakers 115, 231
policymaking 41
inaction 314
interest groups and 313
production-centric policymaking 556
political economy 6, 588
Africa 867, 884–5, 889
classical political economy 17, 18, 27, 29, 184, 285
industrial policy and 7–8, 40, 41
(p. 940) political economy feasibility of interventions 388
politics and 39–40
political settlements 8, 186, 388–9, 885
Africa 849, 885, 889
firms 151, 160, 161–2
politics
conflict and political stability 41–2
corporate elites and 557
industrial policy and 40, 41, 734
political economy and 39–40
populism 286, 290–1, 295
POSCO (Pohang Steel Co., South Korea) 345, 665, 666
positivism 293
post-communist countries 13, 805–806
changes in structure of economies, overview of 781–6, [link]
Page 56 of 79
General Index
de-industrialization 779, 783, 805

economic liberalization 781
exports 779–80, 781, 785–6, [link]
FSU (former Soviet Union) countries 783, [link] , [link] , 806
government failure 781–2
industrial policy: tools 794–7
industrial policy in resource rich countries 796–7, 806
industry value added 785, [link]
innovation, research, and development 797–8
international trade and trade policy 803–805
manufacturing 783, [link]
manufacturing value added 783–4, [link]
market-oriented reforms 781, 805
private sector, share in GDP 800, [link]
privatization 781
R&D expenditure 798, 799
resource-ialization 783, 805
resource rich countries 781, 785
service sector 779, 783
service value added [link]
state vs. private sector 799–802, [link] , [link]
structural economic transformation and industrial policy 786–8, 805
Uzbekistan, industrial policy 792–3
WTO 805
postcolonial theories 285, 287
poststructuralism 285
poverty 521, 540, 549
poverty reduction 240, 242, 289, 296, 793–4, 800
‘poverty trap’ 66
vicious circles of national poverty 529
Powell Memorandum 562, 584n.23
pragmatism 20, 34–5, 196
Prebisch–Singer hypothesis 21, 76–7
price/price system
factor price equalization 508–509, 535, 541, 548
factor price polarization 548
neoclassical economics 125, 134, 145, 508–509
price-led competition 570, 589
shadow prices 126, 138
private sector
barriers to private-sector expansion 130
catch-up 150, 170
de-risking by the state 325, 331
disciplining the 40–1
fall in private investment 467
innovation 314, 319
Page 57 of 79
General Index
post-communist countries 800, [link]

private development banking 345, 347–9, 354–6
renewable energy 412
state/private division 315
state/private sector relation 131, 217
state vs. private sector 799–802 see also firms
privatization 38, 257, 285, 342n.7, 494, 716
Africa 878, 880, 886
Brazil 351, 352, 356n.23, 361, 752, 819
China 719, 724
Latin America 819
New Public Management 320
Nigeria 849
post-communist countries 781
(p. 941) public laboratories 320
of research 320
UK 604, 627
product
product complexity/sophistication 141–2, 244
product discovery 131, 133
product-led competition 570–1, 589
product space 140–4, 144n.39
single/multi-product flow 560
taxonomies of 372
production
‘de-territorialized’ production choices 485, 490
digitalization of production 156, 162, 164–6, 171
dynamics of production structures 69–76
fragmentation of 227, 246, 252, 253, 843
internationalization of production 156, 162–4, 171, 182, 199, 255
new technologies and scaling-down of production 488–9
production engineering 116, 558, 563, 570, 574, 589n.29, 665
production process as a value chain 490
production subsidy 127, 128
production theory 97, 98
production-centric policymaking 556
spatial hierarchy of production sites 843
technological innovation and 588
tertiary production 372 see also global production
production capabilities 40, 137, 138, 141, 260, 662, 673, 756
BRICS 759–61
technological change and 382, 383, 384, 385
US 557, 558, 560, 562, 564, 573, 575, 584, 585, 586
productive capacity 35, 258, 259, 397, 419
definition 20n.11
firms: competition and productivity 158–9
Page 58 of 79
General Index

investment in 244, 439
production capacity/technological capabilities distinction 116
public value theory and 316
technological learning/production capacity complementarity 115–16
productive interests 170, 387, 388–9
productivity
productivity explosion 527, [link]
productivity/growth links 74–6
productivity growth 22, 33, 34n.41, 842
equilibrium productivity growth 456, 473
sources of 791
profit 313
as only purpose of business 562
profit sharing 329
rent/profits distinction 316
protection/protectionism 295, 493, 508, 806
1930s protectionist turn 3
Africa 873
exchange rate protection 128, 132–3, 794, 795
globalization narratives 295, 299, 303
growth and 68, 804
import protection 127, 132
import substitution and 804
nineteenth-century 525
selective protectionism 295, 299
shrinking protection 53
trade protection 43, 113, 127, 129, 132
UK 508
US 112
Uzbekistan 793
public-choice theory 8, 313–14
public‐private partnership 132, 144
clean energy investment project 395, 418–19, 422, 425
grand challenges and dynamic public‐private partnerships 326
South Korea, public–private joint R&D 662, 669–73
public sector
organizational capabilities 322, 324–6
public debt 330
public value 313, 317, 318, 326, 327, 329
collective theory of value creation 315–16
public good 315
public value theory 315–16
Q
(p. 942) Qatar 189
quotas 129, 145
Page 59 of 79
General Index
R
R&D 4, 30–1, 169, 314, 318–19
advanced countries 340
Brazil 823, 834, 836
China 738–9, 740–1, 762
EU 602, 608, 610, 628–9, 631, 633, 636, 637, 638, 642, 644, 648, 653
France 602, 604, 616
Germany 112–13, 605
global South 301
industrial application of research 605, 606, 613
Japan 113
Latin America 822, 823, [link] , 836
market failure approach and 315
mission-driven research laboratories 319
OECD countries 340
post-communist countries 797–8, 799
R&D expenditure 738, 797, [link]
research 318–20
research, privatization of 320
‘servicification’ 7
South Korea 662, 663, 666–7, 674, 828, 834–5
South Korea, public–private joint R&D 662, 669–73
spending on 301, 340
UK 602, 612
US 567, 579, 580 see also innovation
radical industrial policy 178–9
analytic level 179, 186–7, 190, 198
benefits 198–9
capitalism 184–6, 190, 197–8, 199
class and 178–9, 197–8, 199
climate change 188–90
democracy and 188, 190, 191, 193, 194, 197, 198–9
experiences of 191–7
gender issues 178, 179
Kaldorian and structuralist approaches 180–1, 183, 197
labour-centred 179, 187–8, 191, 193–5, 197, 198
Marxism and industrial policy 183–90
Marxist approach to manufacturing and industrialization 182–3, 197
participatory planning models 195–7
prescriptive level 179, 187–8, 190, 198
racial issues 179
radical approaches to manufacturing and industrialization 180–3
state 187, 190, 198
statist models 191–3, 198
value, exploitation, and power in capitalism 184–6
worker co-operatives 193–5, 196, 198
rational choice theory 39
reciprocal control mechanisms 26, 84, 256–7, 817, 848, 855, 859
Page 60 of 79
General Index
China 717, 733–4
gender inclusion 446
Latin America 836
reciprocity 41, 50, 84
regional development 304, 578, 587
regional development banks 304, 344, 348
RDAs (Regional Development Agencies) 614
regional industrial policy 12, 43, 495
transformative regional industrial policy 497–8, 500
regional integration 231
ASEAN 207, 214–16, 218, 221
BRICS 749, 754, 757, 762–5, 772
emerging and developing countries 85
EU 602, 608
inter-regional integration 297, 303
Latin America 820, 822
regional markets 53
Regional Studies Association 496
regional trade 220, 240, 260
regionalism 292, 820
BRICS 292, 303–304
globalization narratives and regionalization 296–7, 303
renewable energy 4, 267, 270, 395, 405–406
benefits 270
capital costs 406
(p. 943) clean renewable energy sources 396, 401, 405
costs of 395, 405–407, 424

costs reduction 267, [link] , 276
employment 270
energy storage systems 405–406
energy subsidies 270n.5
feed-in tariffs 412
geothermal energy 395, 396, 405, 406
Germany: Energiewende 320, 323, 637, 642
governments and 411
hydro-energy 395, 396, 406
India 272–3
innovation 270–1
KfW and clean energy 82, 417, 418
levellized costs of 405, 424
private sector and 412
rise in renewables 270n.6
shift to renewables 267, 269, 274
SMEs and 270
targeting 272 see also China: renewable energy; clean energy; clean energy investment
project; green industrial policy; solar energy/solar PV; wind-power industry
rents 161
Page 61 of 79
General Index
Africa, rent seeking 844, 885

rent/profits distinction 316
rent seeking 4, 83, 156, 238, 348–9, 640, 684, 719, 811–12
state and management of 41
reshoring 6, 486, 493–4
resource allocation 37, 75, 84, 134, 155
resource-based theory of the firm 151, 152, 153–6, 170, 316
shortcomings 153
resources see natural resources
returns 521
diminishing returns 520, 521, 522, 524, 536–7, 549
Law of Diminishing Returns 537 see also increasing returns
reward system 224
innovation 328–9
‘mission-oriented’ policies 322, 328–30
Ricardian trap 465
risks
de-risking 312, 325, 331
development banking, socializing risks 339, 343, 352
late industrialization 337, 339–40
‘mission-oriented’ policies 330–1
NDBs 337, 340, 342, 362
risk capital 79, 132
state 132n.12, 330–1
robotics 53, 164, 300, 382, 384, 487, 611
Russia 752, 769
catch-up 340
digital economy programme 752
exports 780
FDI policy 752
ISI 752
manufacturing 779, 783
resource-rich country 781, 785, 796
SOEs 751 see also BRICS
Rwanda 142–3, 301, 881, 887, 888
BPO (business-process outsourcing) sector 856–7, 861, 862
knowledge-based service sector 845, 855–7 see also sub-Saharan Africa
S
Samsung Electronics (South Korea) 665, 671, 683, 738
SAP (Systems, Applications & Products in Data Processing, Germany) 617, 648n.53
SASAC (State-owned Assets Supervision and Administration Commission, China) 724–6, 730,
733–4
national champions 717, 723, 724, 732, 738 see also China: national champions
Saudi Arabia 189
saving
Page 62 of 79
General Index
domestic savings 49–50, 80, 452, 453, 468, 469, 793, 872

East Asia: high savings rates and coercive state power 346n.10
forced savings 346n.10, 351, 468–9
SCBA (social cost‐benefit analysis) 327
(p. 944) science 318–19, 512
innovation and 29
social sciences 514
technology and 571
US 567 see also STI
SDGs (Sustainable Development Goals) 311, 312, 313, 321
SDG 14, mission-oriented approach 323–4, [link] see also sustainable development
Second Industrial Revolution 97, 298, 384, 386, 486, 599
sector, conceptualizations of 370–9, 389
commodity/product-based concept 372, 375, 380
location-based concept: industrial district 374, 375, 376
production/technology-based concept 372, 373, 375
‘what’/‘how’/‘where’ of production 375–8 see also technological changes
sectoral value chains 371, 380–1, 382, 383–4, 388, 389
sectors
cross-sectoral investment 323
cross-sectoral learning 312
industrial mutations 371, 376, 382
industrial sector 371, 379
intersectoral linkages 23, 133, 239, 244, 377
Marxism 182
Pavitt’s taxonomy of industrial sectors 373–4
primary/secondary/tertiary sectors 370, 371, 372, 373, 377
sector specificity of growth 180, 183, 197
sectoral productivity 370
technical change and classification of sectors 375
technological ranking of 372
US differences among industrial sectors 525–7, [link] , 529
vertically integrated sector 379–80, 381 see also technological changes
security issues
nuclear energy 399
resource security 266–7, 268, 271, 281
water security 277 see also energy security
self-discovery 144, 145n.41, 312, 323
sensoring 486, 489
service sector 162
decline of manufacturing relative to service industries 303, 789
job segregation in 434n.10
Kenya, knowledge-based service sector 845, 857–8
post-communist countries 779, 783
Rwanda, knowledge-based service sector 845, 855–7
‘servicification’ 7, 23, 377–8, 488, 882
Page 63 of 79
General Index
shift-and-share decomposition methodology 106, 119–20

Siemens (Germany) 345, 605, 718, 726, 739
Siemens-Alstom merger 622, 630
Silicon Valley (US) 229, 363, 567, 579
Singapore 213n.15, 681
de-industrialization 682, 703–704
economic growth and structural change 682
EDB (Economic Development Board) 692, 693, 695, 708
EE (electrical and electronics) exports 697
ELI 683, 685–6, 690, 692–3, 697, 703
FDI 26
free trade 549
growth 681, 711
innovation 708
ISI 685
manufacturing 695–6, 703
‘mission-oriented state’ 693, 705, 711
technological upgrading/STI infrastructure 682, 683, 693, 695, 704, 705, 708, 710, 711
TNCs 213, 693 see also South East Asia
Sinopec (China) 717, 722, 723, 725, 729, 731, 735
smart regulation 499
smart specialization (EU) 496, 609, 621, 633, 651, 652, 653–4
challenges 647
entrepreneurial discovery 646
(p. 945) ‘good’/‘bad’ economic activities 532, 533
place-based innovation-oriented industrial policy 633, 647

R&D and innovation policies 647
Strategy for Smart Specialization 645–6
‘unsmart’ specialization 533 see also EU industrial policy
SMEs (small and medium-sized enterprises) 82, 83, 154
Africa 882
ASEAN 209, 221–2, 223, 230–1, 258
BRICS 756, 757, 767
China 718
employment 222
EU 602, 636
France 615, 616
Germany/Mittlestand 596, 605, 612
Italy 603
UK 613
social market economy 510
social reproduction 186, 430, 431, 432, 439–40, 442, 445, 446
concept of 439n.15
women, unpaid social reproduction 179, 439 see also gender/growth regimes; gender and
industrial policy
Page 64 of 79
General Index
social sciences 514
socialism 193, 196, 199, 716
African socialism 872
market socialism 195
municipal socialism 195
SOEs (state-owned enterprises) 342n.7, 347, 362
Africa 874, 880, 884, 886
BNDES, Brazil 351, 352, 353n.18
China 105, 256, 716, 718–19, 751
EU 595, 601–602, 625
Russia 751
solar energy/solar PV 4, 395, 405, 406
Africa 412
Brazil 419–20
China 275–6
as clean renewable energy 396
costs of 267, [link] , 405, 406
financing 412
India, National Solar Mission 272–4, [link]
South Africa 420
South Korea 420 see also renewable energy
South Africa 132n.14, 185–6, 753
apartheid industrial policies 753, 858, 860, 869
apparel sector 845, 858, 859–61, 862
automotive sector 845, 858–9, 862
corruption 861, 862
digital production technologies 382–3
diversifying manufacturing beyond heavy industries 845
globalization 295, 296–7, 303
Joule Electric Vehicle 672
manufacturing 869
modernization programme 753
privatization 753
solar industry 420
WTO 753 see also BRICS; sub-Saharan Africa
South East Asia 13, 682n.2, 710–11
automotive industry 697, [link] , 699
BoP problems 688, 689, 692, 694, 697, 699, 704, 711
corruption 685
deregulation 687, 694, 701, 703
‘Dutch Disease’ 686, 689, 711
electrical/electronics 691, 692, 694, 696, 697, [link] , 699, [link] , 701, 702, 705, 709, 711
ELI 683, 690–6, 704, 705, 711
Page 65 of 79
General Index
EPZs (export-processing zones) 687, 688, 692, 694

‘fallacy of composition’ syndrome 683, 686, 688, 696, 711
FDI 683, 711
GERD (gross expenditure on R&D) [link] , 708, 709
IPR 709, 710, [link]
ISI 683, 685–90, 695, 710–11
(p. 946) latecomer industrialization 681, 682–3
macroeconomic mismanagement 452
manufacturing 695–6, [link] , 710
manufacturing value added [link] , 701, 704, 711
middle-income trap 682
monetary policies 694
motor vehicle exports 697, [link] , 699, [link]
premature de-industrialization 681, 682, 705, 711
smuggling 685, 692
STI infrastructure 687, 691–2, 694, 703, 705, 707–10, 711
technological upgrading 690, 695, 705–10
TNCs 682, 688, 690, 691, 693, 694, 695, 701, 704, 706, 709, 710, 711 see also Indonesia;
Malaysia; the Philippines; Singapore; Thailand
1960 Foreign Exchange Control Act 663
1997 crisis 829, 837
absorptive capacity 662, 663–7
automotive industry 225, 527, 697, 699
capability building 662, 665, 666–7
catch-up 662, 669, 671, 683, 762
chaebols 159, 224, 684, 721, 722
competition regulation 159
deregulation 664
engineering 665, 666
entry control 662, 668–9, 673
exports 697, 828
FDI 25
foreign technology acquisition 663–6, 669
green industrial policies 652, 828
growth 673
HCI (Heavy Chemical and Industry) policy 113
human capital 665, 828
industrial policy 138, 217, 720, 721
infant industry protection 662, 667–9
in-house R&D 663, 666–7, 669, 674
investment capabilities 662, 673
Japan and 666, 671
KFTC (Korea Fair Trade Commission) 159, 160
KOITA (Korea Industrial Technology Association) 664
Page 66 of 79
General Index
Korea Development Bank 665, 666

Latin America, comparison with 823, 827–9, [link] , 831, 833–6, [link]
leapfrogging 671–2, 683
macroeconomic policies 828–9, 837
manufacturing 109
monetary policies 683–4, 812
patents 664–5, 666–7
production capabilities 662, 673
public–private joint R&D 662, 669–73
R&D 662, 663, 666–7, 674, 828, 834–5
Schumpeterian and capability-based view of industrial policy 661–74
solar industry 420
state intervention 665, 673–4
successful latecomer industrialization 112, 217, 469, 663, 720, 812, 814
tariffs 662, 667–9, 673
technology licensing 662–7, 673
technology transfer 666
training 665
Southern Rhodesia 869, 870
Soviet Union 196, 318
statist model of industrialization 191–3, 198
space technology 486, 489
Spain 523, 527
specialization 471, [link] , 544,
bad specialization 542
comparative advantage, specialization according to 240
evolution of technological capabilities and production specializations 111–14
inertia in 813
patterns of 73, 84, 471–2, 473, 633, 811, 823
trade specialization 141, 803 see also smart specialization
stage theories 539
(p. 947) stagflation 466, 473, 479, 561–2
standardization 382, 541, 542

state 38–43
2008 financial crisis, responses to 178
China 151, 163, 782–3, 827
economic development and 4, 885–6
entrepreneurial state 800
as facilitator for the private sector 131
firms and 157
industrial development and 217
innovation and 132nn.12–13, 134, 314, 329–30
neoclassical economics and 509
as organizer of production 35, 38
as pioneer and catalyst of industrial development 608
Page 67 of 79
General Index
private sector/state relation 131, 217, 330, 331

radical industrial policy 187, 190, 198
rents, management of 41
risk-taking 132n.12, 330–1
state failure 8, 799, 817
state–market interaction 37, 38–9, 42
state/private division 315
trade and 238
von Thünen’s modern state 519–20, [link]
welfare state 7, 323, 359 see also developmental state; state intervention
state capitalism 632, 717, 718
China 719–20
reinvention/re-emergence of 352, 362
US 562
state intervention 11, 134, 313
1945–1980 interventionist approach to industrial policy 594–5, 596, 601–602
ASEAN 211n.11, 217, 220n.33, 226
capitalism 38–9
catch-up 32, 38
climate change 190
contradictions 187
dynamic efficiency and 63
France 601, 615, 644
industrial policy and 42, 217, 473, 494–5
Italy 603
market failures and 313–14, 494
‘picking winners’ 495, 595
state activism 38–9
technological capability 37, 38
STI (science, technology, and innovation)
BRICS 685, 751, 752, 755–6, 766, 768, 770, 772
South East Asia 687, 691–2, 694, 703, 705, 707–10, 711
structural adjustment (1980s–1990s) 126, 129–31, 145, 468
Africa 844, 846, 876, 877–8, 879, 880, 889
Indonesia 690
Latin America 468
the Philippines 692
structural heterogeneity 64, 72, 73, 74, 85, 375
structural transformation 20–1, 242–4, 759, 811, 842
catch-up and 111–14
China 105–106, 109
‘deep’/‘shallow’ transformation 73–4
distributive effects of structural transformation strategies 65
economic development and 63, 66, 69, 83
GVCs and 251, 253
industrial policy and 7, 18, 19, 26, 43–4, 83–7
Page 68 of 79
General Index
industrial policy as driver of 10–11, 26

industry-oriented structural transformation 794
ISI and 21
macroeconomics/finance/structural change links 79–83
manufacturing and 21–3
‘regressive’ structural change 811
state and 11
successful structural transformation 112
technological change and 369–70
technological learning and 35
structuralism 63–4, 125, 180, 812, 813–15
exports 24
Latin American structuralism 64, 66, 72
(p. 948) manufacturing 21, 180–1
neo-structuralism 64
sub-Saharan Africa 256, 259, 434
1990s–2010s growth revival 842, 846
apparel sector 851–2
capability gap 847–8
cement industry 848–9, 870
challenges for late-late industrializing countries 843–4
domestic politics 848
external debt crises 846
firms 847–8, 861
industrial policy 842, 844–5, 863
learning costs/investment 843–4, 853–4, 855
learning rents 848, 849, 861
limited progress in developing manufacturing sectors 845–6
technological capabilities 846
trade liberalization 846 see also Africa; Ethiopia; Kenya; Nigeria; Rwanda
subsidies 339–40, 344–5, 794
BNDES 352n.16, 356, 357–8, 359, 360n.26
China 668, 733
energy subsidies 270n.5
EU, state aid 622, 624n.4, 626n.13, 627, 628, 635–6, 637
exports 127, 132, 142, 224, 353
fossil fuels subsidies 416, 422
France 604
infant industries 350, 511
innovation 132–3
production subsidies 127, 128
tax-subsidies 127, 145
supply side 180, 217, 237, 312, 314, 429, 437
macroeconomic mismanagement 452
macroeconomic policy and industrialization 452, 453, 456
supply-side biases of industrial policy 452
Page 69 of 79
General Index
supply-side interventions 133
sustainable development 54, 178, 311
sustainable growth 312 see also SDGs
Sweden 606
Swiss Federal Technical Cooperation 513
Switzerland 292
T
Taiwan 116, 670, 683, 762
exports 697
innovation 684
monetary policies 683–4
TNC-led industrialization 213n.17
Tajikistan 781, 783, 805, 806
Tanzania 881–2, 883, 885, 887, 888–9
targeting 26, 47, 138, 216, 256, 389, 473
China 256
NDBs 362, 363
targeting failure 672
upgrading 260
tariff barriers 128, 215, 242, 697, 804, 818
tariff protection 128n.5, 519, 599, 667–8, 870
taxation
avoidance/evasion of 188, 329, 362
carbon taxes 270n.5, 404, 412, 414, 416, 418–19n.31, 422
corporate taxation 329
diversification and 546
England 512
environmental taxes 127
favouring labour saving over resource saving 331
inflation tax 345, 355, 356
innovation and 328
monopolies and 316
neo-liberalism 557
pre/post-tax subsidies 416n.28
primary role of 316
tax concessions 465, 794, 804
tax holidays 132n.13, 688, 690
tax reduction 558, 561, 581
tax subsidy 127, 145
UK 600
US 558, 561, 576–7, 581 see also fiscal policy
technical progress function 74–6, [link]
technological capabilities 29, 31, 35, 42, 49, 52, 134, 153, 842–4
digital capability gap 371, 382–3, 385–6
(p. 949) evolution of technological capabilities and production specializations 111–14
large firms 36n.47
production capacity/technological capabilities distinction 116
Page 70 of 79
General Index
state intervention 37
sub-Saharan Africa 846
tacitness of 813
upgrading of 813–14 see also capabilities
technological change 53, 70–1, 94, 165, 595
classification of sectors and 375
cross-sectoral technical change 381, 384
definition 29
digitalization/digital technical change 382–3, 389
economic development and 29–31, 93, 369
evolutionary theory 813
Fourth Industrial Revolution and 486–90
industrial ecosystem framework 371, 380–2, [link] , 388, 389
industrial mutations 371, 376, 382
industrial policy and 6, 7, 383–9
innovation 369, 381–2
interdependency of 379
localized technical change 813
path dependence 30
relevance of 369
as sector-specific process 381, 384
sectoral shifts 370
shifting terrain of the industrial 382, 389
sources of 376
structural change as non-linear process 377
technical change/investment interaction 22 see also digital production technologies: chal
lenges; sector, conceptualizations of
technological progress 20–1, 78, 95, 97, 373
importance of 29
TFP (total factor productivity) 136
technological upgrading 135n.19, 150, 161, 258, 446
BRICS 761
government and 134
South East Asia 690, 695, 705–10 see also upgrading
technology 97
borrowing technology 97
development and 22
disruptive technologies 566, 611–12
frontier technologies 4, 104, 114
fundamental properties of 94–7
GPTs (general purpose technologies) 379
growth and 30n.31
interdependency 379
new technologies 3
paradigms, routines, organizations 96–7
science and 571
tacit nature of 70–1
Page 71 of 79
General Index
technological asymmetries 94, 104, 811, 813

technological dominance, micro-heterogeneity, and non-substitution 97–110
technological paradigms and trajectories 94–5
technology diffusion 30, 385–6
technology gaps 98, 101, 185, 813
transfer of 97, 258, 666, 739, 759 see also automation; catch-up; knowledge; STI
technology: specific countries
China 98, 739, 753, 761–2
EU 623, 626, 628, 629, 630, 636, 637, 638, 642, 644, 647–9, 653
Japan 571, 572
South Korea, foreign technology acquisition 663–6, 669
South Korea, technology licensing 662–7, 673
US 564, 566, 567–8, 571–2, 573, 580, 616
technology-intensive sectors 104, 246, 357, 472, 473, 614, 691, 692, 719, 760, 767, 832
telecommunications 351, 569, 583, 669–70, 726
TFP (total factor productivity) 101, 111, 136–7, 667–8
Thailand 253, 463, 681
automotive industry 208, 226, 228, 704, 709
digital technical change 382, 383
(p. 950) economic growth and structural change 682
ELI 689, 690, 693–4, 704

growth 681, 694
IMF/World Bank assistance 689, 694
ISI 685, 689, 695
manufacturing 704
technological upgrading/STI infrastructure 694, 695, 709
universities 709 see also South East Asia
Third Industrial Revolution 298, 383, 386, 486–7, 495, 601, 611
Third Way 510n.2
Thirlwall’s Law 461
TNCs (transnational corporations) 26, 152, 162–3, 213, 251, 288, 300n.12
ASEAN 213, 220
China 213n.17, 722, 735
digital platforms 165
global concentration in 158
inequality and 163
regulation of 163
Singapore 213, 693
South East Asia 682, 688, 690, 691, 693, 694, 695, 701, 704, 706, 709, 710, 711
Taiwan 213n.17
TNC-led industrialization 213
‘transnational’ term 162
Toyota 345, 738, 739
ASEAN and 208, 218, 220n.33, 224
TPP (Trans-Pacific Partnership) 53, 211, 219, 220
trade
bilateral trade 53, 84, 112, 804
Page 72 of 79
General Index
developing countries 239, 255

gains from 239–42
Heckscher-Ohlin model of international trade 227n.46
large firms and international trade 238
losses from 240, 242
multilateral trade 19–20, 52, 219, 240, 804
post-communist countries 803–805
regional trade 220, 240, 260
state and 238
symmetrical trade 527, 538
trade/development links 237–45, 254
trade/economic growth links 241–2
trade/industry connection 245, 288
trade law 301
trade models 238, 240, 241, 251
trade protection 43, 113, 127, 129, 132
trade specialization 141, 803
trade wars 85, 210, 493
‘vent for surplus’ models of international trade 68 see also the entries below related to
trade; free trade
trade integration 238, 292
active vs. passive policies 255–7
GVCs 245–54
policy challenges 254–60
RVCs (regional value chains) 259–60
trade integration/development links 239–45, 254
upgrading and sophistication 257–8 see also GVCs
trade liberalization 23, 130, 668, 803–804
Brazil 819, 823
DRC ratio 139
impact of 243, 296
Kenya 878
Latin America 819–20, 825, 826
shortcomings 240–1
South Africa 858
sub-Saharan Africa 846
trade liberalization/growth relation 68, 76, 240, 241, 243 see also free trade
trade theory 524, 525, 529, 539
Ricardian trade theory 525, 535, 536, 541
trade union 187, 190, 194
anti-union legislation 178, 558, 561, 562
traditional industries
decline of 3, 165
upgrading of 298
(p. 951) training 5, 37, 115, 116, 117, 258, 595
EU 636
Page 73 of 79
General Index
globalization and 300, 301, 302, 305

on-the-job training 436, 438
South Korea 665
vocational training 222, 231 see also education; learning
transformative industrial policy 486, 496, 497–500
transformative regional industrial policy 497–8, 500 see also place-based approach
Trinidad and Tobago 189
TRIPS Agreement (Trade-Related Aspects of Intellectual Property Rights) 52
Turkmenistan 780, 781, 785, 787, 793, 805, 806
U
UAE (United Arab Emirates) 189, 293
UK (United Kingdom) 17, 27, 594
2017 Industrial Strategy White Paper 311
Calico Act 600
catapult centres 613
colonial industrial policy 869
digital platforms: competition enforcement 167–8
First Industrial Revolution 599–600, 616
free trade 112
industrial policy 599–600, 604, 612–14, 626
‘Industry 4.0’ 613
innovation 604
‘municipal socialism’ 195
neo-liberalism 4, 612, 627
privatization 604, 627
protectionism 508
R&D 602, 612
regional inequalities 614
SMEs 613
taxation 600 see also England
Ukraine 549
UN (United Nations) 188, 271, 311, 535, 794–5
BRICS and 757, 770
UN ESCAP (UN Economic and Social Commission for Asia and the Pacific) 788–9, 794
UNCTAD (UN Conference on Trade and Development) 189, 290, 513, 535, 621
2018 World Investment Report 5
catch-up 33–4
clusters 47n.65
symmetrical trade 538
underemployment 72, 73, 75, 126, 331
unemployment 126, 214, 298, 331, 360–1
UNEP (UN Environment Programme) 269
UNFCCC (UN Framework Convention on Climate Change) 274
UNIDO (UN Industrial Development Organization) 137, 253, 258, 286, 296, 535, 685
CIP Index (Competitive Industrial Performance Index) 760–1, [link]
digitalization/4IR technologies 298, 766, 767, 769
Page 74 of 79
General Index
universities 302
Carnegie Mellon University 856
European University Institute, Florence 649
Germany 112, 318
industry‐university linkages 37, 49, 301
IUCs (industry‐university collaborations) 301
monopoly of neoclassical economics at university level 550
Nanyang Technical University 708
National University of Singapore 708
research universities 579, 580, 612
STEM programmes 443–4
Thailand 709
UK, catapult centres 613
University of the Philippines Diliman 708
upgrading 243, 260, 288, 759
Europe 511
gender and industrial policy 438, 442, 443–4, 446
GVCs 246, 251–3, 257–8, 260, 491
labour’s skills/industrial upgrading complementarity 442n.21
learning and 444
(p. 952) macroeconomic policy and industrialization 471, 473, 480 see also industrial up
grading; technological upgrading

US (United States) 17, 154, 548, 550, 556, 566, 788
1930s Depression 577
1970s 555, 556n.2, 561–2, 568, 580–1, 582
1980s 557, 558
American North/South divide 557–9, 560–2, 572–82, 587
anti-union legislation 558, 561, 562
capability triad 562–3, 585–8
corporations 154, 582–3, 584
dark money 562
debt 586n.24
de-industrialization 556, 558, 568, 572, 581
democracy and wealth 582
deregulation 558, 561, 562, 583–4
differences among industries 525–7, [link] , 529
engineering 558, 563, 570, 573–4, 584, 586
garments and textiles 576, 578, 580, 582
government 559, 562, 565–8, 574
innovation 564, 565, 572, 573
leapfrogging 114
macroeconomic policy 560, 568
Manhattan Project 318, 566
manufacturing 534, 562, 564, 573–4, 576, 578, 584
manufacturing employment 555, 568, 574, 580, 581–2
Page 75 of 79
General Index
mass production 541–2, 544, 560, 573
military/defence spending 415, 559, 567, 578–9, 580
‘municipal socialism’ 195
neo-liberalism 4, 557–8, 562, 572, 582–5, 586
Pax America 559, 568
plantation economy 557, 560, 561, 572, 575
post-war policy 559–60, 566–8, 578, 580, 616
productivity growth 556
protectionism 112
R&D 567, 579, 580
slavery 557, 561, 575–7
structural change 112
successful latecomer industrialization 112, 242
taxation 558, 561, 576–7, 581
technology 564, 566, 567–8, 571–2, 573, 580, 616
wages, ‘great reversal’ 555
weapons industry 559, 567, 579, 580
World War II 556, 579n.16
World War II experience/policy 560, 561, 563–6, 570, 587 see also the entries below for US;
New Deal
US: agencies and departments 566
DARPA (Defense Advanced Research Projects Agency) 566
Department of Commerce 565, 566
Department of Defense 566
Department of Energy 566
DPC (Defense Plant Corporation) 564, 565, 566, 586
MDC (Manpower Development Commission) 564
OSRD (Office of Scientific Research and Development) 564, 565, 567
PWA (Public Works Administration) 577
RFC (Reconstruction Finance Corporation) 565–6, 586
TVA (Tennessee Valley Authority) 563, 577, 578, 586–7
WPB (War Production Board) 564, 565, 578
US: ‘belts’ 572
‘Cotton Belt’ 561
manufacturing belt 558, 560, 562, 568, 572, 573–5, 576, 577, 578, 580, 581
Rust Belt 558, 568, 580, 581
‘Sun Belt’ 561, 575–9
US: Federal Reserve Bank 523, 562, 565, 568, 584
2007–09 global financial crisis 415, 523
Federal Reserve Act 582
macro stabilization 565
(p. 953) US: international dimension
China/US trade issues 546

EU 617, 626, 629, 632
Germany 557, 559–60, 569
Page 76 of 79
General Index
Japan/US rivalry 557, 559–60, 568, 569–72

USSR (Union of Soviet Socialist Republics) 718, 780, 781, 792, 793
late industrialization 720 see also post-communist countries; Russia
Uzbekistan 780, 781, 785, 787, 791, 792–3, 797, 805, 806
V
value 315
NPV (net present value) 327 see also the entries below for value; public value
value added 162, 164
‘cult of added value’ 513, [link]
domestic value added 84, 210–11, 246, 247
foreign value added 210–11
GVCs 164, 490–1
increasing value added 228, 259, 544, 825
industrial policy theory 513, 514, 517, 546
low value-added manufacturing/trap 223n.38, 231, 252, 463
post-communist countries, industry value added 785, [link]
post-communist countries, service value added [link]
wealth and 514 see also manufacturing value added
value chain
new technologies and 489–90
production process as a value chain 490
RVCs (regional value chains) 259–60
sad curve 489, [link]
sectoral value chains 371, 380–1, 382, 383–4, 388, 389
‘short’/‘long’ value chains 250
smile curve 247, [link] , 252, 489, 491 see also GVCs
value creation 151, 156, 165, 168, 217, 380, 558
collective theory of 315–16
GVCs 499
public value creation 316, 329
value chain 489–90, 494
value extraction 184, 316, 329, 330
Venezuela 822, 827
Verdoorn’s Law 21, 181, 525
Vietnam 191, 212, 253, 782–3, 805
digital technical change 383
manufacturing 783
Vinnväxt programme (Sweden) 498–9
virtual reality 445, 611
W
wages
ASEAN: automotive industry 226
average wage per cluster category, Europe 532
China 474n.16
demand and 474–6, 479
Page 77 of 79
General Index
gender wage gap 433, 446

high wages 525
impact on economic growth of a one percentage point increase in real wage growth 474–6,
475
low-wage female labour 179, 430, 431–2, 446
lower wages 178, 431
share in the gross output price 474–5, 475
wage and profit rate 477–8, [link]
wage differentials 74, 135
wage earners 468, 525, [link]
wage-goods inflation 465–7, 473
wage rates 127, 226
Washington Consensus 7, 126, 227, 718, 751
colonial policy 535
criticism of 131
deregulation 314
exports 314
free market 38
Latin America 819–22, 836 see also neo-liberalism
Washington Institutions see IMF; World Bank
wealth 549
industrialization and 537, 598, 843
large firms and 154
technical knowledge and 22n.15, 29
(p. 954) value added and 514
wealth creation 18
WEF (World Economic Forum) 286, 295, 298, 304
welfare economics 125–6n.1, 145, 313
welfare/welfare state 7, 187, 323, 359
wind-power industry 395, 406
China 273, 274–5, 281
as clean renewable energy 396
costs of 405, 406
onshore wind 405 see also renewable energy
women
human capacities and 439, 440, 442, 443, 446
ICT and 445n.25
‘marriage bar’ 432n.4
unpaid work 430, 432, 437, 439, 445n.25, 446
weaker bargaining power 433 see also gender and industrial policy; gender inequality; so
cial reproduction
workers
austerity for 178
co-operatives 193–5, 196, 198 see also labour; wages
World Bank 290, 338n.4, 403, 404, 476, 689, 781
development banking and diversity of funding 350
on growth 662
Page 78 of 79
General Index
on middle-income trap 33n.38
policy hierarchy approach 129n.10
renewable energy investments 418
structural adjustment programmes 130, 468
The East Asian Miracle 805
Washington Institutions 522, 527, 535, 549–50
World System Theory 227n.47
World War II 4, 557, 685, 868
clusters as post-Second World War phenomena 46
post-Second World War ‘golden age’ 66, 67n.10
US 556, 579n.16
US World War II policy 560, 561, 563–6, 570, 587
WTO (World Trade Organization) 53, 84, 290, 804
2005 Agreement on Textiles and Clothing 860
China 109n.3, 451, 609, 719, 727, 728, 735, 739, 803
constraints posed by 19
e-commerce moratorium 299, 303
India 273–4
ISI policies 695
Latin America 822
multilateral trade rules 19–20, 52
Y
Yugoslavia
BOALs (Basic Organizations of Associate Labor) 195–6
radical industrial policies 191, 195–7
Z
Zimbabwe
apartheid boycott 537
de-industrialization 538 (p. 955) (p. 956) (p. 957) (p. 958) (p. 959) (p. 960) (p. 961) (p. 962)
Page 79 of 79

Arkebe Oqubay, Christopher Cramer, Ha-Joon Chang, and Richard Kozul-Wright - The Oxford Handbook of Industrial Policy-Oxford University Press (2020)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Arkebe Oqubay, Christopher Cramer, Ha-Joon Chang, and Richard Kozul-Wright - The Oxford Handbook of Industrial Policy-Oxford University Press (2020)

Uploaded by

Copyright:

Available Formats

Copyright Page

Print Publication Date: Oct 2020 Subject: Economics and Finance

(p. iv) Copyright Page

Great Clarendon Street, Oxford, OX2 6DP,

Oxford University Press is a department of the University of Oxford.

© Oxford University Press 2020

The moral rights of the authors have been asserted

First Edition published in 2020

All rights reserved. No part of this publication may be reproduced, stored in

You must not circulate this work in any other form

Published in the United States of America by Oxford University Press

198 Madison Avenue, New York, NY 10016, United States of America

British Library Cataloguing in Publication Data

Library of Congress Control Number: 2020937433

Printed and bound by

Print Publication Date: Oct 2020 Subject: Economics and Finance

Print Publication Date: Oct 2020 Subject: Economics and Finance

(p. xi) List of Figures

17.4 An export-led growth model with cumulative causation and a balance-of-pay­

BoP-constrained growth rate 466

Print Publication Date: Oct 2020 Subject: Economics and Finance

(p. xv) List of Tables

Print Publication Date: Oct 2020 Subject: Economics and Finance

(p. xvii) List of Contributors

is professor of the political economy of development at SOAS, University of London.

(p. xviii) Richard Kozul-Wright

is director of the Globalisation and Development Strategies Division in UNCTAD. He

is associate professor in industrial economics at University College London (UCL),

is associate professor in the School of Economics at the University of Johannesburg

is professor of business economics at the University of Birmingham and senior fellow

has been actively engaged in industrial restructuring policy in London, Massachu­

is an assistant professor at the School of International Studies, Peking University. Her

(p. xx) Lisa De Propris

is professor of regional economic development, University of Birmingham. She has

is a senior lecturer in the political economy of development at SOAS, University of

is professor of economics at Sant’Anna School of Advanced Studies, Pisa, Italy, and

is an associate professor in the American University’s School of International Service

processes forging perceptions of legitimacy where democratic institutions are frac­

is an economist at the United Nations Conference for Trade and Development. He

is an Estonian academic and science administrator. He is deputy director and profes­

is emeritus professor of economics at Keele University, United Kingdom. He holds a

Analysis of Economic Catch-up (Cambridge University Press, 2013), as well as the

is professor emeritus in the Faculty of Business and Economics at Macquarie Univer­

is professor of the economics of innovation and public value at University College

is distinguished research professor and university professor emeritus at the American

José Antonio Ocampo

is professor at Columbia University’s School of International and Public Affairs, and

is distinguished university professor of economics and co-director of the Political

is a principal researcher in the Central Economics and Mathematics Institute of the

is an economic affairs officer at ECLAC and professor of economics at the Federal

currently holds the position of distinguished professor of economics at the University

a Norwegian citizen, is a professor at Tallinn University of Technology and guest re­

is a professor of economics at the University of Johannesburg and an economics direc­

is an industrial policy officer at the United Nations Industrial Development Organiza­

consulted with numerous international organizations, including the UNDP, UNRISD,

is a senior lecturer at Delft University of Technology. He obtained a PhD in economics

(p. xxvi) Fiona Tregenna

is emeritus professor of development economics at the University of Bradford, United

is an assistant professor in the Department of Entrepreneurship and Innovation at

Introduction to Industrial Policy and Development

Abstract and Keywords

1.1 An Overview of Industrial Policy and Devel­

1.1.2 Evolving Challenges to Industrial Policy

17.4 An export-led growth model with cumulative causation and a balance-of-pay

has been actively engaged in industrial restructuring policy in London, Massachu

processes forging perceptions of legitimacy where democratic institutions are frac

is an Estonian academic and science administrator. He is deputy director and profes

is professor emeritus in the Faculty of Business and Economics at Macquarie Univer

a Norwegian citizen, is a professor at Tallinn University of Technology and guest re

is a professor of economics at the University of Johannesburg and an economics direc

is an industrial policy officer at the United Nations Industrial Development Organiza

1.1 An Overview of Industrial Policy and Devel

and in analyses of particular countries’ positions in international alliances and hierar

Hirschman, Albert O. (1968) ‘The Political Economy of Import-substituting Industrializa

Richard Kozul-Wright is director of the Globalisation and Development Strategies Di

From a long-term, strategic perspective, industrial policy underpins structural transfor

2.2 Theoretical Perspectives on Industrial Poli

An active industrial policy underpinning structural transformation and catch-up compris

terms of investment concentration, learning, and innovation…This is what climb

A commitment to structural transformation and developing technological capability re

2.4 The State and the Politics of Industrial Poli