Essays on Economic Growth and Industrial Development:

A comparative analysis between Brazil and South Korea

Igor Lopes Rocha

Sidney Sussex College

September 2015

This PhD dissertation is submitted for the degree of Doctor of Philosophy

DECLARATION
This dissertation is the result of my own work and includes nothing which is the outcome of
work done in collaboration except where specifically indicated in the text.

This dissertation does not exceed the word limit for the Degree Committee of Human, Social
and Political Sciences.

i
SUMMARY
This dissertation consists of four chapters/essays related to the role of the manufacturing
sector in economic growth. The concern with the manufacturing sector in the performance of
the economic dynamic derives from the longstanding debate surrounding the dichotomy
between East Asia and Latin America, particularly South Korea and Brazil, in terms of
productive structure. Therefore the thesis is structured as described below.

The first chapter, as the theoretical core of this research, systematises the literature on
economic growth and industrialisation in which the manufacturing sector plays an important
role as the main engine of growth. The main contribution relies on the attempt to systematise
the literature on sector specificity in a common channel that comprises the Structuralist,
Kaldorian and Neo-Schumpeterian contribution with regards to the role of industrialisation,
specifically manufacturing, in economic growth.

The second chapter seeks to analyse in a comparative perspective the remarkable contrasts
and parallels about the process of economic development undertaken by Brazil and South
Korea over the last 60 years. Circumscribed by factors which include the sectoral composition
of countries and the role of the state in guiding the developmental path, the main contribution
of the chapter is an identification of factors that determined different routes of economic
dynamism followed by Brazil and South Korea from the 1980s onwards.

The third chapter analyses the structural transition of Brazil and South Korea through an
input-output framework. This analysis investigates these economies regarding their patterns of
productive diversification, structural transition and density of industrial chains. The
contributions of this chapter to the existing literature are twofold. First, the chapter adopts the
methodology proposed by Imbs & Wacziargs (2003) in an inter- and intra-sectoral perspective
to identify the path of economic diversification undergone by Brazil and South Korea over the
decades. Second, in a complementary analysis, the input-output framework is used to identify
and quantify factors that contributed to this pattern of specialisation.

The fourth chapter, as in the previous one, uses principally input-output methodologies to
compare the Brazilian and South Korean productive structures in terms of productive
dynamisms and leading sectors. Moreover, aspects regarding the interdependence of the
manufacturing and services are also investigated. The contributions of this chapter are
twofold. First, this analysis provides a useful framework to understand specificities and
regularities of each economy and methodological referential for further empirical studies of
sectoral specificity of growth. Second, this chapter estimates the interdependence between
services and manufacturing through a new input-output methodology. Therefore, this study
advances some findings presented by the literature and opens an interesting window for future
research.

Supervisor: Dr Gabriel Palma

ii
To my parents, Armando and Fatima,

and my brother Felipe.

iii
 ACKNOWLEDMENTS

First and foremost, I would like to express my appreciation to my Supervisor, Gabriel Palma
for his suggestions on the development of this research and for the many lively discussions on
the subject. I am also extremely grateful for his considerate support and encouragement,
especially in my turbulent first year.

I also would like to thank other professors whose lectures I could attend in during my first
years and that contributed to my way of thinking. In particular, I would like to thank Dr. Ha-
Joon Chang, Peter Nolan and Shailaja Fennell. I am also very thankful to Ajit Singh to share
tirelessly his wide intellectual knowledge, before his sudden death.

I am grateful to Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (Capes) of

the Brazilian Ministry of Education for fully funding my studies at the University of
Cambridge. I must also acknowledge conference travel grants awarded to me by Sidney
Sussex College and The Centre of Development Studies.

I must also acknowledge the intellectual support and generous friendship of Prof. Mariano
Laplane, who personally checked my progress in Cambridge. I owe huge gratitude to Ricardo
Carneiro, Luiz Gonzaga Belluzzo, Guilherme Santos Mello, Marcos Vinicius Chiliatto Leite,
Celio Hiratuka and Fernando Sarti at the State University of Campinas (UNICAMP), Brazil,
Luiz Carlos Bresser-Pereira, Nelson Marconi, Paulo Gala and José Marcio Rego at Getúlio
Vargas Foundation (FGV-SP), Brazil, and Jose Roberto Afonso and Aloisio Campelo at the
Brazilian Institute of Economics (IBRE/FGV), Brazil. I also thank Joaquim Guilhoto, at the
University of São Paulo (USP), Brazil, for his always kind help and valuable suggestions with
the input-output methodology.

I would like to thank my friends in Cambridge who made my days incredibly rich
intellectually and for their support. I will always remember them fondly. I must thank my
friends from the Centre of Development Studies, in particular Natalya Naqvi, Domna
Michailidou, Javier Gonzalez, Jostein Løhr Hauge, José Miguel Ahumada, João Silva, Ivanka
Mamic and Anne Tulpe. Thanks are also due to Guiherme Magacho, Rafael Ribeiro, Fabricio
Silveira, Fabio Terra, Roberto Borghi, João Prates and Paulo Camuri for the endless
discussions about economics and many extremely funny moments. I must also thank visiting
scholars in Cambridge, especially André Cunha, Fernando Ferrari, Marco Flávio Resende,

iv
Ricardo Araújo, Frederico Jayme Jr., and Alexandre Guimaraes. I would like to thank André
Cunha and Marco Flávio Resende for their valuable suggestions on this research.

Most importantly, I would like to express my gratitude to my family in Brazil, for their
constant support and unconditional love. Words cannot convey my gratitude to my parents,
Armando and Fatima, and my brother Felipe. I would like to thank my beloved grandmother
Roza Marques and my aunt Vera and my uncle Yoshimine. Although my grandfather (José
Lopes), my grandmother (Zoraide Ozimo), and my uncle (Nelson Rocha) are no longer with
us to read these words, I know they are happy for this moment.

v
 TABLE OF CONTENTS

INTRODUCTION ................................................................................................................ 1
STRUCTURE OF THE DISSERTATION AND METHODOLOGY .................................... 2

CHAPTER 1 – MANUFACTURING AS DRIVER OF ECONOMIC GROWTH .......... 5

Introduction .......................................................................................................................... 5
1. The structuralist view: An alternative approach to the neoclassical analytical framework
...................................................................................................................................... 8
1.1. The Early Structuralist approach to manufacturing: First insights ........................ 10
1.1.1. Industrialisation: Balanced and unbalanced growth theories........................ 11
1.2. Latin American Structuralism: Linking underdevelopment to the centre-periphery
paradigm ............................................................................................................................ 16
2. Kaldor’s stylised facts and the contemporaneous debate............................................... 21
2.1. Manufacturing as engine of growth: from industrialisation to de-industrialisation 22
2.2. Statistical illusions ................................................................................................ 26
2.3. Servicification ....................................................................................................... 27
2.4. Productive Fragmentation..................................................................................... 28
3. Manufacturing as the main source of productivity growth ............................................ 32
4. Manufacturing really does matter for the equilibrium in the balance of payments......... 34
5. Technological dynamics, innovations and economic growth ........................................ 36
5.1. The catching-up hypothesis: From imitation to innovation .................................... 37
5.2. Manufacturing shaping a hierarchy of national innovation systems: The case of
Machinery industries .......................................................................................................... 40
6. Concluding remarks ..................................................................................................... 43
Appendix 1.1 - The Kaldor-Dixon-Thirlwall cumulative causation model ............................ 46

CHAPTER 2 – FALLING BEHIND AND MOVING AHEAD: THE BRAZILIAN AND

SOUTH KOREAN PROCESS OF INDUSTRIALISATION.......................................... 48
Introduction ........................................................................................................................ 48
1. The Developmental State promoting a Big Push in economic growth ........................... 50
2. Mapping the Golden Age of the Industrialist Era.......................................................... 52
2.1. Brazil: 1930-1980 ................................................................................................. 52
2.2. South Korea: 1945-1980 ....................................................................................... 58

vi
3. The role of the State and private sector in the industrialisation process......................... 62
3.1. Creating National Champions: The South Korean Chaebols ................................. 62
3.2. Contrasting developmental patterns ...................................................................... 65
4. External Debt and Exogenous Shocks .......................................................................... 68
5. Technological Learning and Industrial Development ................................................... 71
6. The “fundamentalist” neoliberal shock ......................................................................... 75
7. Brazil falling behind and South Korea moving ahead ................................................... 78
7.1. Brazil: FDI flows and the loss of industrial dynamism........................................... 78
7.2. The loss of manufacturing and productive disarticulation ..................................... 81
7.3. Labour productivity............................................................................................... 85
7.4. Manufacturing Share and Income ‘Catching-up’ ................................................... 87
7.5. Export-trajectories: Two historical distinct patterns.............................................. 88
8. Concluding remarks ..................................................................................................... 92
Appendix 2.1 - Brazil’s Macroeconomic Indicators, 1950-2013 (%)..................................... 96
Appendix 2.2 - South Korea’s Macroeconomic Indicators, 1950-2013 (%)........................... 99
Appendix 2.3 – Classification of manufacturing industries by technology group ................ 102

CHAPTER 3 – PRODUCTIVE DIVERSIFICATION, STRUCTURAL TRANSITION

AND DENSITY OF INDUSTRIAL CHAINS: ARE BRAZIL AND SOUTH KOREA
VERY DIFFERENT? ......................................................................................................103
Introduction .......................................................................................................................103
1. Structural change and economic dynamic: Specialisation vs. Diversification...............105
1.1. The U-shaped curve for Brazil and South Korea ..................................................106
1.2. Sectoral composition and economic dynamism .....................................................111
2. The structural decomposition analysis .........................................................................114
3. Interpreting results in light of distinct economic policies .............................................117
4. The density of industrial chains ...................................................................................121
5. Concluding remarks ....................................................................................................124
Appendix 3.1 – Technical Details ....................................................................................... 127
Appendix 3.2 – Map of sectoral aggregation (UNIDO) ...................................................... 131
Appendix 3.3 – Map of sectoral aggregation (ECLAC-PADI) ............................................ 132
Appendix 3.4 – Classification of manufacturing subsectors by technology group ............... 133

vii
CHAPTER 4 – ENGINES OF GROWTH AND SECTORAL INTERDEPENDENCE IN
BRAZIL AND SOUTH KOREA .....................................................................................134
Introduction .......................................................................................................................134
1. The relevance of industrial linkages: a Hirschmanian perspective ...............................136
2. Output Multipliers and Hirschman-Rasmussen Linkages ............................................138
2.1. Output Multiplier .................................................................................................138
2.2. Hirschman-Rasmussen Linkages ..........................................................................140
2.3. Pure Normalised Backward and Forward Linkages .............................................143
3. The interdependence between manufacturing and services ..........................................146
3.1. Opening intersectoral linkages .............................................................................147
3.1.1. Forward Linkage decomposition .................................................................148
3.1.2. Backward Linkage decomposition ...............................................................149
4. Concluding Remarks...................................................................................................151
Appendix 4.1 – Technical Details ....................................................................................... 154
Appendix 4.1 – Map of sectoral abbreviation and aggregation........................................... 159
Appendix 4.2 - Hirschman-Rasmussen Backward Linkage index (Brazil) ........................... 160
Appendix 4.3 - Hirschman-Rasmussen Forward Linkage index (Brazil) ............................. 161
Appendix 4.4 - Hirschman-Rasmussen Backward Linkage index (South Korea) ................. 162
Appendix 4.5 - Hirschman-Rasmussen Forward Linkage index (South Korea) ................... 163
Appendix 4.6 - Backward and Forward Linkages Decomposition (Brazil, %) ..................... 164
Appendix 4.6 - Backward and Forward Linkages Decomposition (South Korea, %) ........... 165

CONCLUSIONS ..............................................................................................................166
REFERENCES ................................................................................................................173

viii
LIST OF FIGURES

Figure 1.1 - Annual GDP growth and change in the share of manufacturing value added in the
global manufacturing GDP, 1970-2010 ................................................................................ 24
Figure 1.2 - The ‘inverted-U’ relationship between manufacturing and income per capita .... 25
Figure 1.3 - Value added in the Global Value Chain............................................................. 29
Figure 2.1 - Number of state-owned enterprises created in each year (1950-1990)................ 66
Figure 2.2 - External Debt, total (% GDP) and Total external debt/exports (ratio), 1961 - 2013
............................................................................................................................................ 69
Figure 2.3 - Brazil: Foreign Direct Investment, 1990-2008 ................................................... 79
Figure 2.4 - Brazil: Manufacturing Share in GDP and Value Added by Technology Intensity
(%) ...................................................................................................................................... 83
Figure 2.5 - South Korea: Manufacturing Share in GDP and Value Added by Technology
Intensity (%) ........................................................................................................................ 84
Figure 2.6 - Labour Productivity and ‘Catching-up’ with the US, 1960-2010 ....................... 86
Figure 2.7 - Share of manufacturing in GDP and relative income gap with the US, 1950-2010
............................................................................................................................................ 87
Figure 2.8 - Share of Manufacturing and Exports in GDP (%) (1960-2010).......................... 89
Figure 3.1 - The inter-sectoral U-shaped curve and sectoral share for South Korea............. 108
Figure 3.2 - The inter-sectoral U-shaped curve and sectoral share for Brazil ...................... 110
Figure 3.3 - The intra-sectoral U-shaped curve for South Korea ......................................... 111
Figure 3.4 - Shares in global manufacturing and GDP ........................................................ 114
Figure 4.1 - Hirschman-Rasmussen BL and FL indices (Average 1995-2009) .................... 141
Figure 4.2 - Intermediate consumption of services in manufacturing value-added (%)........ 147
Figure 4.3 - Forward Linkage decomposition (%) .............................................................. 149
Figure 4.4 - Backward Linkage decomposition (%) ............................................................ 150

LIST OF TABLES

Table 2.1 - Brazilian Economic Development Plans and Economic Indicators ..................... 58
Table 2.2 - South Korean Economic Development Plans and Economic Indicators .............. 62
Table 2.3 - The 10 largest Chaebol groups in South Korea from late 1950s to 2000 ............. 64
Table 2.4 - Major indicators of innovation activities............................................................. 73
Table 2.5 - Stocks and inflows of foreign direct investment in Brazil, by sector ................... 81
Table 3.1 - Sectoral composition of the manufacturing sector (%), 1990-2008 ................... 112

ix
Table 3.2 - Inter and Intra-sectoral decomposition of output growth (%), 1995-2002.......... 118
Table 3.3 - Inter and Intra-sectoral decomposition of output growth (%), 2003-2008.......... 120
Table 3.4 - Import input coefficients for Korea, 1995-2008 ................................................ 122
Table 3.5 - Import input coefficients for Brazil, 1995-2008 ................................................ 123
Table 4.1 - Output Multipliers ............................................................................................ 139
Table 4.2 - Pure normalised backward linkages .................................................................. 144
Table 4.3 - Pure normalised forward linkages..................................................................... 145

x
 “After some four decades of concerned attention of development,
we might ask how much economics can explain. Economic theory can
determine the necessary, though not the sufficient, conditions of growth.
The so-called noneconomic factors account for the gap between the
necessary and the sufficient. Any evaluation of development can only
state that the necessary conditions for growth exist or are being created;
it cannot predict with certainty that growth will actually take place. One
can learn a lot from past performance, but the criteria of evaluation are
ex ante concepts. They yield a probability judgment and have, therefore,
to be continually checked" (Rosenstein-Rodan, 1984, p. 207)

“Economic development has much to do with human

endowments, social attitudes, political conditions – and historical
accidents. Capital is a necessary but not a sufficient condition of
progress.” (Nurkse 1953, p.1)

“To explain why certain regions have become highly

industrialised, while others have not we must introduce quite different
kinds of considerations – what Myrdal (1957) called the principle of
‘circular cumulative causation’. This is nothing but the existence of
increasing returns to scale – using the term in the broadest sense – in
processing activities. (…) As Allyn Young (1928) pointed out in a famous
paper, Adam Smith’s principle of ‘division of labour’ operates through
the constant sub-division of industries, the emergence of new kinds of
specialised firms, of steadily increasing differentiation – more than
through the expansion in the size of the individual plant or the individual
firm.” Nicholas Kaldor, “The Case for Regional Policies”, 1970

“The most important Agenda of the state relates not to those

activities which private individuals are already fulfilling, but to those
functions which fall outside the sphere of the individual, to those
decisions which are made by no one if the state does not make them. The
important thing for government is not to do things which individuals are
doing already, and to do them a little better or a little worse; but to do
those things which at present are not done at all.” John Maynard
Keynes, “The end of Laissez-faire”, 1926

xi
INTRODUCTION

This dissertation consists of four chapters/essays related to the role of the manufacturing
sector in economic growth. The concern with the manufacturing sector in the performance of
the economic dynamic derives from the longstanding debate about the dichotomy between
East Asia and Latin America, particularly South Korea and Brazil, in terms of productive
structure. During the developmental era, despite superficial similarities marked by dictatorial
regimes and economic development plans, both regions had employed very different models
of industrialisation. With the outbreak of the debt crisis during the 1980s, these distinct
productive structures engendered a clear polarisation of the emerging world expressed in
different trajectories and performances. Furthermore, it prompted a wide debate about the
industrial pattern and economic policies that contrast the South Korean success with the
Brazilian failure in terms of economic development.

Moreover, it raised a controversial debate between mainstream and heterodox

economists. While the conventional view neglects the sectoral-specific approach of the
economic growth, the developmental school states that the specialisation pattern matters
for the pace and scope of the development process. Consequently, the literature has explored
the dilemma between the capacity of the state in coordinating policies in favour of the
manufacturing sector and the so-called market-friendly policies of the neoliberal approach.
Thus, the main aim of this dissertation is to analyse and compare economic growth and
industrial development in Brazil and South Korea, focusing on the importance of the
manufacturing sector for economic development. Moreover, this study seeks to summarise
important aspects related to the manufacturing sector in economic growth within a theoretical,
historical and empirical approach. Thus, in the light of these propositions, the main questions
for which this thesis aims to provide answers are:

• Is manufacturing still the main engine of growth?

• Why did these countries develop different industrial structures from the 1980s
onwards?
• How deep is the dichotomy between the Brazilian and the South Korean development
patterns?
• What are the sectoral engines of growth in Brazil and South Korea?

1
 • Are manufacturing and services becoming more interdependent in these countries over
the years?

STRUCTURE OF THE DISSERTATION AND METHODOLOGY

To fulfil the questions summarised in the last section, this dissertation presents four
chapters which diverge widely in scope and methodology. Each chapter/essay is related to the
theme of economic growth and industrial development with current references to important
issues neglected by the neoclassical view synthesised in the first chapter. In this way, all of
them are structured through a heterodox background where three strands can be highlighted:
the Anglo-Saxon and Latin American Structuralism, the Kaldorian and the Neo-
Schumpeterian approach.

The first chapter, entitled “Manufacturing as driver of economic growth”, is a study

which makes an effort to combine different theoretical strands on development regarding the
importance of the manufacturing industry to the economic growth. Through a confluence of
the Keynesian-Kaldorian, Structuralist and Neo-Schumpeterian frameworks, the essay argues
that the manufacturing industry presents some special properties which are not found in other
sectors. The discussion constitutes the theoretical core of this dissertation. The first section
describes the Anglo-Saxon Structuralism focused on structural change dynamics and the Latin
American Structuralist view of underdevelopment, according to which the economic
development results from technical progress induced or enabled by capital accumulation. The
second section seeks to describe the Kaldorian approach to growth, understood as “laws”
where Kaldor explains the differences in international growth rates recovering important
elements in the contemporaneous debate. The third section investigates the Neo-
Schumpeterian route to development, exploring relations between innovation, economic
dynamics and catching-up in a sectoral specific approach.

The second chapter, entitled “Falling behind and moving ahead: the Brazilian and
South Korean Process of industrialisation”, seeks to analyse the remarkable contrasts and
parallels among the development process in these countries over the last 60 years. The first
section elucidates briefly the whole of the developmental state for economic growth. The
second section, based on a historical retrospection, describes the industrialisation process
from the 1950s to the debt crisis in the 1980s. The third section makes an attempt to evaluate
the role of public and private sectors in the formation of these late industrialised countries.

2
Thus, seeking to answer why the decoupling in growth rates after the 1980s was much more
drastic in Latin American countries vis-à-vis Asian economies, the fourth section explains the
genesis of the debt crisis. The fifth section analyses the reason behind distinct technological
dynamisms between these two economies and the trajectory of technological learning
undertaken by South Korea. The rise of the neoliberal era is described in the sixth section to
give the necessary support to understand the economic policy change from the 1980s
onwards. Therefore, the outcomes in terms of industrialisation and sustained economic growth
in a long-term perspective are approached in the seventh section highlighting the dichotomy
between Brazil and South Korea in terms of labour productivity, catching-up and exports. The
last section presents some concluding remarks.

The third and fourth chapters compose the empirical part of this dissertation with a
focus on Brazilian and South Korean industrial structures. The methodologies used in both
chapters are based on the input-output analysis. Therefore, data available at the World Input-
Output Database (WIOD) was used when applied to the input-output methodology in a
comparative perspective. This database covers 40 countries for the period 1995-2009, and the
data is available in both current and previous years’ prices. Additionally, the methodology
encompassed secondary sources. This included among others The Brazilian Institute of
Geography and Statistics (IBGE), The Institute for Applied Economic Research (Ipea), The
Bank of Korea (BOK), The Brazilian Central Bank (BCB), The Conference Board and
Groningen Growth and Development Centre, The United Nations Conference on Trade and
Development (UNCTAD), The United Nations Industrial Development
Organization (UNIDO) and The Organisation for Economic Co-operation and Development
(OECD).

Hence, the third chapter, entitled “Productive Diversification, Structural Transition

and density of Industrial Chains: Are Brazil and South Korea very Different?”, seeks to
analyse the structural transition undertaken by Brazil and South Korea over the decades. The
first section discuss the long-lasting debate on specialisation or diversification and applies the
method proposed by Imbs & Wacziarg (2003) to investigate the stages of sectoral
diversification presented by Brazil and South Korea in a inter- and intra-sectoral perspective
over the decades. In this part the methodology is also described, including database sources
and level of sectoral aggregation. The second section focuses on the sectoral composition of
manufacturing in these economies and their economic dynamism in a global perspective. The
third section, inspired particularly by Chenery’s studies, adopts the so-called structural

3
decomposition analysis (SDA) to investigate distinct factors, i.e. demand, trade and
technological change, which contributed to the movement of specialisation observed in these
economies. In the fourth section, the chapter consolidates the understanding of the structural
economic dynamics via movements related to the density of industrial chains. This section
advances the input-output methodology employed by Rocha (2011) to estimate imported input
coefficients for Brazil and South Korea through the World Input-Output Database (WIOD).
The fifth section concludes the chapter.

The final chapter is entitled “Engines of Growth and Sectoral Interdependences in

Brazil and South Korea”. This last chapter uses input-output methodology to study certain
growth dimensions of the sectoral structure in Brazil and South Korea. The first section
provides a brief discussion on the relevance of industrial linkages for economic growth, based
mainly on Hirschman’s theory. The second section measures the output multipliers and
Hirschman-Rasmussen backward and forward linkages comprising all productive subsectors
of manufacturing, services and the primary sector. The output multiplier indicates how much
is produced for each monetary unit spent on final demand. The Backward Linkage (BL) index
indicates the extent to which a given sector demands from other sectors while the Forward
Linkage (FL) index, in turn, measures the extension to which a given sector is demanded by
other sectors. Additionally, in the same section, the Pure Normalised linkages (PNL) – also
known as Guilhoto, Hewing & Sonis (GHS) model – are calculated to evaluate the productive
structure of Brazil and South Korea, considering the size of sectors in the economy each
country. In the face of important findings related to the services sector, the third section
analyses the interdependence between manufacturing and services. Finally concluding
remarks are provided in the fourth section.

4
CHAPTER 1 – MANUFACTURING AS DRIVER
OF ECONOMIC GROWTH

Introduction

Theoretical, historical and empirical evidences have shown the importance of manufacturing
to sustained economic growth. However, over the years, neoclassical economics has neglected
such evidences, and advocated the idea that economic growth is sector-indifferent and in
some models also activity-indifferent1. A remarkable example of models in which state that
growth is activity- and sector-indifferent are the early neoclassical growth models, i.e. Solow-
Swan type models and the early endogenous growth models, namely “AK” models (Palma,
2005)2. Solow-Swan type models are a result of classical contributions of Solow (1956, 1957)
and Swan (1956) and became the dominant approach for the analysis of economic growth
from mid-1950s until 1970s. However, over the years this model started to be considered by
many economists an unrealistic description of the process leading to economic growth. In this
approach, since investments have diminishing returns to capital, only continuous
technological changes explain long-term economic growth. In spite of being a central
explanatory factor, technological progress is not sector- or activity-specific3.

The dissatisfaction with the assumptions of the Solow-Swan model and its capacity to
explain the non-convergence of living standards in the world economy stimulated further
developments in the neoclassical growth theories. Frankel (1962), in an early version of the
so-called “AK” growth models, endogenised the main factors, such as technological progress,

1
There is an important distinction between “activity” and “sector’. Examples of the former are research and
development (R&D) and education; examples of the latter are manufacturing, agriculture and services (Palma,
2005).
2
See also Tregenna (2009).
3
Although economists have long recognized the crucial importance of technological change as a major source of
dynamism in capitalist economies (especially Karl Marx and Joseph Schumpeter), it was Solow’s work that
brought technological progress to prominence in the mainstream economics as a central explanatory factor in the
analysis of long-term rate of economic growth. However, somewhat paradoxically, in Solow’s theory
technological progress is exogenous, i.e, it is not explained by the model (Snowdon & Vane, 2005). While the
Solow model is widely used as a baseline model of economic growth, it is still considered by many to be
unsatisfactory as a description of the process leading to economic growth. This is because the model views
improvements in total factor productivity (technological progress) to be the ultimate source of growth in output
per worker, but does not provide an explanation as to where these improvements come from. In the language of
economists, long-term growth is determined by something that is exogenous in the model.

5
that drive economic growth in the long run4. As pointed out by Palma (2005), recent
endogenous growth theories are also included in this class of models, in which changes in the
rate of growth are the result of the cumulative effect of market imperfections arising in the
process of technological progress that operates in an obscure cumulative effect, creating
increasing returns5. Despite relevant developments in the neoclassical framework, sectoral
and activity specificity were still absent in these theoretical approaches.

In the years that followed, important references regarding activity specificity arose in
the neoclassical theory. The most recent wave of endogenous models, the so-called New
6
Growth Theory, initiated by the research of Paul Romer (1986, 1990) and Robert Lucas
(1988), emerged in response to perceived theoretical and empirical deficiencies associated
with the neoclassical growth model. In this approach, economic growth is explicitly attributed
to the type of activity presented in the economy, particularly Research and Development
(R&D) and education. Thus, the most important mechanisms concern the creation of new
technical knowledge in R&D departments of firms (Romer, 1990) and the formation of
human capital in education processes (Lucas, 1988).

In this type of growth model, neoclassical economics (and also early neo-
Schumpeterian models – such as in Aghion & Howitt (1998) and Grossman & Helpman,
(1991), which focus on technological spillovers) approached the activity specificity as the
main source of endogenous growth. Although this approach incorporates the endogenous
characteristics of economic development (in contrast with Solow-Swan models, in which
economic development is a result of exogenous shocks), the mechanisms employed continued
to assume a self-regulating nature of the growth process, where technological change occurs
automatically and is affected only by macroeconomic aggregates, that is, physical or human
capital stocks.

Therefore, although recognising endogenous elements of economic growth, the

importance of sector specificity remained overlooked in the neoclassical approach. Moreover,

4
Aghion & Howitt (1998) note that Frankel (1962) presented an early AK model that went largely unnoticed by
the profession. In this model, the ‘endogenous’ technological progress offsets the growth-cushioning effects of
diminishing returns to capital accumulation that characterise the Solow model. However, it is important to
highlight that in the heterodox strand, Robinson (1956) was the first to endogenise productivity growth.
5
See Snowdon & Vane (2005), Barro & Sala-i-Martin (2004), and Blankenburg (2000, 2004) for overviews of
new growth theories.
6
Romer (1986) explains technological progress as an unintentional by-product of capital accumulation by
individual firms. However, a few years later, Romer (1990), dissatisfied with his initial approach, proceeded to
develop a second strand of new growth theory which embraces a neo-Schumpeterian framework of endogenous
technological change (Snowdon & Vane, 2005).

6
neoclassical economics presumes that different types of economic sectors are structurally
similar enough to be aggregated in a single representative sector. In such a way, both the
mechanisms which trigger the process of economic growth and the structural economic
dynamics are ignored. Additionally, inserted in a framework of general equilibrium and self-
adjustment of economic variables, this strand of economic thought does not consider
industrial policy as an effective way to promote economic development. In contrast to the
structuralist approach, including the Latin American one and the Keynesian-Kaldorian one,
traditional neoclassical models, as well as some of the later endogenous growth models,
increasing returns, though generated by research-intensive activities, are explicitly not
associated with the size, depth or strength of the manufacturing sector as such or with the
process of capital accumulation within the manufacturing sector. Nor do they allow for
specific effects from the manufacturing sector on the R&D activities (Palma, 2005).

In contrast to the traditional neoclassical view, the heterodox literature has emphasised
sectoral specificity as a central feature of the economic growth process. This implies that, as
opposed to the neoclassical view, a value-added unit is not necessarily equivalent across
sectors, especially in terms of inducing and enhancing economic growth (Tregenna 2009).
Commonly, the fundamental role of manufacturing in economic growth is only approached by
the Structuralist and the Keynesian-Kaldorian views. However, a careful analysis of the
heterodox literature reveals that manufacturing is also the main engine of technological
dynamism and specifically a locus of innovation in the Schumpeterian sense. The
convergence of all these theories through a common channel, i.e. the special properties of the
manufacturing sector, include a complex linking between theoretical approaches at different
levels of economic theory, i.e. micro (firm), meso (sector and sub-sectors) and macro
(economy). Therefore, as a corollary to this analysis, the heterodox triad composed by the
Structuralist, Keynesian-Kaldorian and Neo-Schumpeterian views constitutes the mainspring
of this study.

In this context, the aim of this chapter is to systematise the literature on economic
growth and industrialisation in which the manufacturing sector plays an important role as the
main engine of growth. Thus, the main contribution relies on the attempt to systematise the
literature on sector specificity in a common channel that comprises the Structuralist,
Kaldorian and Neo-Schumpeterian contribution to the role of industrialisation, specifically
manufacturing, in economic growth. In this way, the first section reviews the main aspects
related to the Anglo-Saxon and the Latin American approaches which, in turn, shed light on

7
the industrialisation process as a mechanism to overcome underdevelopment. The second
section explores how manufacturing is the main engine of growth in the Kaldorian
framework, understood as “laws” where Kaldor explains the differences in international
growth rates and the approach developed for these “laws”, and the importance of
manufacturing for the balance of payments. In the third section, another heterodox strand is
considered, i.e. the Neo-Schumpeterian approach, exploring relationships between innovation
and economic growth. Particularly, it analyses the idea that manufacturing is also the main
engine of technological dynamism and specifically a locus of innovation. The catching-up
hypothesis is emphasised, highlighting the connection between national systems of innovation
and productive structure. Moreover, the sectoral specificity of the concept of systems of
innovation is emphasised. The last section presents some concluding remarks.

1. The structuralist view: An alternative approach to the neoclassical analytical

framework

Although structuralism is a popular term in the economic field, it is important to

recognise that many concepts are derived from neighbouring sciences such as Levi-Strauss,
Godelier (anthropology), Lacan, Piaget (psychology), Althusser, Derrida, Foucault
(philosophy) (Palma, 1987; Gibson, 2003; Blankenburg, Palma & Tregenna, 2008).
Therefore, it is possible to conceive structuralist economics, as it emerged in the 1940s, as an
outgrowth or extension of earlier work in these and other fields. Within this multidisciplinary
background, structuralist economics is fundamentally a theoretical approach that confronts the
neoclassical methods of empiricism and positivism (Palma, 1987). Structuralism uses a
method of inference analogous to that of abduction or retroduction. It begins with an
observation of determinate phenomenon, “what is out there”, and then works backward to a
theory. Its focus is not on prediction but description and explanation (Baghirathan, Rada &
Taylor, 2004). Consequently, structuralism can be understood as an alternative way to
theorise in economics, since the mainstream theory represented mainly by the neoclassical
approach is deductive and expressed in terms of ‘uniformities’ interpreted as (actual or
hypothetical) correlations or event regularities.

Furthermore, as pointed out by Blankenburg, Palma & Tregenna (2008),

structuralism takes place in conception of “an integrated system of distinguishable yet
mutually constitutive elements”. In other words, the relationships that constitute structures are
more important than individual elements. This assertion is a central feature of the structuralist

8
view and distinguishes this theory from the neoclassical approach7. According to the latter,
the analysis of human action can be performed in a micro approach from the perspective of
individual agents (methodological individualism). However, in the former, structural analysis
emphasises that interdependent elements of the economic system form a complex whole in a
macro perspective and incorporates systemic properties that cannot be reduced to the analyses
of individual elements. Thereby, the hallmark of the structuralist approach is its reliance on
internal relations among parts making up a whole, which is closely related to the
methodological holism8 (Jackson, 2003).

In economics, structuralism is principally associated with the so-called Anglo-Saxon

or Early Structuralism and the Latin American strand9. Both strands base their analyses on the
concept of complementarities and poverty traps, linkages, and dualism (Ancochea, 2007). The
structuralist view usually stresses that economic development is strongly linked to a radical
transformation in the structure of production to suppress obstacles, bottlenecks and other
rigidities of underdevelopment. Based on the hypothesis that the industrial structure affects
both the rhythm and the direction of economic development, structuralist literature highlights
the importance of the industrialisation as a process of structural change where the
manufacturing sector plays a central role.

The structuralist strand states that without a dynamic industrialisation, it is not feasible
to increase employment, productivity and income per capita and, consequently, to reduce
poverty. The main argument stresses that the development process involves a production
reallocation from low productivity to high productivity sectors where increasing returns to
scale prevail. Inserted in this theoretical background, economic structuralism has provided

7
Moreover, as stressed by Street & James (1982), structuralism assumes two basic conceptions against the
conventional neoclassical view. The first conception regards the economic system as an evolving process rather
than an equilibrating mechanism of stable economic relations centring on market activities. The second
conceives of human behaviour as characterised by habitual patterns resulting from cultural conditioning but
capable of intelligent response to changing realities. It is thus distinguished from the neoclassical economic view
that human behaviour is primarily devoted to utilitarian motivation and pecuniary calculation in a static system
of markets. In a convergent analysis, Chenery, writing in the American Economic Review in 1975, exposes the
origin of economic structuralism as a general view against the neoclassical approach based on the free market,
emphasising the importance of economic planning principally in late-development countries. In other words,
economic interventionism is seen as a central variable to overcome various inhibiting factors in economic
growth. Consequently, the structuralist approach is an attempt to “identify specific rigidities, lags, and other
characteristics of the structure of developing economies that affect economic adjustments and the choice
of development policy” (Chenery, 1975, p. 310).
8
For a detailed explanation about the methodological individualism versus the methodological holism, see
Kincaid (2008).
9
It is important to emphasise that the French structuralism, represented particularly by François Perroux, was
very influential on the Latin American strand which in turn is detailed in the subsection 1.2.

9
many reflections on how economic growth should be understood in a historical perspective of
mutual causation in the economic system. While various historical, political and ideological
factors contributed to the structuralist view, Keynesian criticism of the neoclassical
economics and its argument regarding state interventionism was principally important.

Paul Rosenstein-Rodan, Ragnar Nurkse, Arthur Lewis, Albert Hirschman, Gunnar

Myrdal and Hollis Chenery belong to the handful of economic thinkers associated with early
structuralism or pioneers of development10. Their seminal contributions challenged the
neoclassical view of market efficiency to structural change and recognised particularities
through which the manufacturing industry has a central role to support and propel economic
development. A further theoretical contribution comes from Latin American structuralism,
which is mainly related to the Economic Commission for Latin America and the Caribbean
(ECLAC), whose works merged into a coherent school of thought in late 1950s. In the light of
historical experiences, the main thoughts presented in this Latin American version are
encapsulated in the works of Raul Prebisch and Celso Furtado, focused on the specific
challenges faced by developing countries given an world economy divided by two poles, the
“centre” and the “periphery”, and the distinctive structure of production present in these ones
(Prebisch, 1949; Furtado, 1964). Problems relating to dualism in international trade,
technology disparities, balance of payments constraint and state interventionism were all
emphasised.

1.1.The Early Structuralist approach to manufacturing: First insights

In economic theory, many studies associate the emergence of the Early structuralism
with the publication of Rosenstein-Rodan’s “Problems of Industrialization of Eastern and
South-Eastern Europe”11. In this study, Paul Rosenstein-Rodan assigned particular emphasis
to the transformative power of industrialisation in the economic system (Rosenstein-Rodan,
1943). In a similar line of thinking, Nurkse (1953), Lewis (1954), Hirschman (1958), Myrdal
(1957) and Chenery (1960, 1979) pointed out that the study of long-term economic growth is
a “sector- specific” process and consequently involves an increase of the industry share,
which, in turn, provides the highest potential of productivity, spillover effects, forward and
backward linkages, as well as technological and pecuniary externalities. Hence, their focuses

10
See for instance, Blankenburg, Palma & Tregenna (2008) and Ancochea (2007).
11
Rosenstein-Rodan (1943).

10
were essentially on the internal special properties of manufacturing and on the way in which
these properties spread to the economy as a whole, stimulating the process of economic
growth.
Although not always emphasised by the literature, the essence of these classical
contributions relied especially on Allyn Young’s ideas concerning long-term determinants of
economic growth which were further extended in their seminal studies. These pioneers of
economic development also focused on the identification of bottlenecks and rigidities that
block the industrialisation process in underdeveloped economies. However, in spite of general
agreement amongst these pioneering economists that industrialisation was the most efficient
means to support economic development, the emphasis on growth economics through
industrialisation engendered a debate over whether “balanced” or “unbalanced” growth was
the best strategy to extract and propel economic development through the special properties of
manufacturing.

1.1.1. Industrialisation: Balanced and unbalanced growth theories

The Early Structuralist approach to manufacturing is particularly associated with

Rosenstein-Rodan’s path-breaking research in economic development, which stresses the
conditions for economic growth in line with Nurkse (1953). Paul Rosenstein-Rodan and
Ragnar Nurkse supported the balanced growth theory based on “classical” arguments
concerning long-run determinants of the economic growth, particularly dynamic externalities
and increasing returns, as advanced by Allyn Young. This type of argument gave rise not only
to the role of demand complementarities and increasing returns (to scale) in manufacturing
industries, but also various arguments that justify industrial policy, especially of selective
type, on the basis of the existence of interdependence between different activities (Chang et
al., 2013).

Rosenstein-Rodan (1943) states that a remarkable feature of high-income economies,

i.e. developed countries, is a structured and dynamic industrial sector. Unlike developed
economies, underdeveloped countries were characterised by the absence of a structured and
dynamic industrial sector. As a matter of fact, since industrialisation tends to be concentrated
in developed countries, massive and planned investments coordinated by the state are sine qua
non conditions for the creation of a new institutional environment and, consequently, the
successful carrying out of industrialisation in underdeveloped countries. In this way,

11
Rosenstein-Rodan (1943) describes what later became known as the “big push theory”, i.e. a
large-scale development programme geared towards jump-starting the economic growth
through the industrialisation process of an underdeveloped economy. Rosenstein-Rodan states
that free market mechanisms would only keep or even increase the distributive inequality
between countries, showing the growing gap between developed and underdeveloped nations.
In his own words “the market mechanism does not realize the ‘optimum’ either in one nation
or between nations because it relies on such unrealistic assumptions” and “obscures the nature
of the development process” (Rosenstein-Rodan, 1984, p. 209).

Regarding planning, the author highlights two crucial points. The first is related to
labour training policies coordinated by the state to transform peasants into industrial workers.
This assertion is based on the observation that the automatism of laissez-faire never worked
properly in this field. In other words, from the perspective of an individual firm, investing in
training labour is very risky since if workers move to another firm a considerable loss of
capital may occur. Although not a good investment for a private firm, it is the best investment
for the state when analysed the economy as a whole (Rosenstein-Rodan, 1943). The second
and most important argument in favour of such a large investment unit refers to the
complementary influence between different industries that potentiates the dynamic effects of
external economies and balances the process of economic growth. The expansion of the
market through the creation of a planned complementary system of industries reduces the risk
of demand shortage and, since risk can be considered as cost, it reduces costs and provides the
most important set of arguments in favour of a large-scale industrialisation12.

In such a way, a big, comprehensive and balanced investment package between

manufacturing sectors performed by the state, i.e. “the big push” – using Rosenstein-Rodan’s
terminology – is the key to the economic development process through positive linkages
effects in the productive chain that enhances the dynamic effects of external economies.
From this perspective, industrialisation has a central role in economic development not only
because of the terms of trade differential, as noted by Prebisch, but also due technological and
pecuniary external economies which are the main source of increasing returns to scale – a

12
As argued by Rosenstein-Rodan (1943) in his shoe factory example, the diversity of human wants creates a
necessity for a planned effort to generate a sufficient and sustained expansion of the market.

12
central aspect in the development process – and in turn, are much higher in manufacturing
industry than in agriculture13.

In a similar view to Rosenstein-Rodan, Nurkse stressed that economic growth is “not a

spontaneous and automatic affair14”. With this assertion in mind, Nurkse (1953) describes the
forces that limit the development process in underdeveloped countries. The so-called “vicious
circle of poverty” is illustrated as “a circular constellation of forces tending to act and react
upon one another in such a way as to keep a poor country in a state of poverty” (ibid, p.4).
This dynamic, translated in a low level of investment and capital accumulation, operates both
on the supply and demand side. In this way, from the supply side a low level of investment
arises from the small amount of savings available in the economy as a result of its low income
level which, in turn, is a consequence of a low level of productivity. Moreover, low
productivity is a direct result of small amounts of capital used in the production process and is
related to the low domestic savings presented in the country. From the demand side, similar to
Rosenstein-Rodan, the greatest obstacle to development was the atrophy of the domestic
market through low demand for goods due to low income level in the economy which, in turn,
discourages the formation of capital. The low level of capital used in the production process is
associated with a weak level of investments that implies a low level of productivity existing in
the country. When the productivity per worker is low, the real income is consequently low
and the poverty vicious circle is complete. Additionally, it is important to emphasise that
when analysing the underlying causes of the scarcity of capital, Nurkse (1953) did not treat it
just as an issue of resource availability.

The author recognises that underdevelopment was linked to the kind of product
produced by a specific country and how it was traded in the international market. In order to
break this circle, a wave of capital investment in various industries should be carried out. This
would enlarge the market size, increase productivity and provide an incentive for the private
sector to invest. As pointed out by Nurkse, the only way out of the dilemma is a “more or less

13
Although Rosenstein-Rodan (1943) clearly approaches the discussion of technological externalities, the
discussion of pecuniary externalities discussion is less so. However, Rosenstein-Rodan (1961, 1984) explain it
taking into consideration their horizontal and vertical dimensions on both demand and supply sides. According
to Rosenstein-Rodan, any expansion of the market through the process of industrialisation leads to external
economies, both pecuniary and technological. Pecuniary externalities are market-transmitted or inter-industry
interdependencies. Horizontal pecuniary external economies occur between firms producing consumer goods,
while vertical pecuniary external economies occur between suppliers and final goods producers. In terms of
technological externalities, manufacturing is recognised as a source of effective knowledge (contemporaneously
also termed as learning by doing).
14
Nurkse (1953, p.4).

13
synchronized application of capital to a wide range of different industries. Here is an escape
from the deadlock; here the result is an over-all enlargement of the market (...) most industries
catering for mass consumption are complementary in the sense that they provide a market for,
and thus support, each other” (Nurkse, 1953, p. 11).

In contrast to Nurkse and Rosenstein-Rodan, Hirschman did not support the “balanced
growth theory”, arguing that imbalances generated between sectors could provide corrective
reactions, giving arguments in favour of a theory of “unbalanced growth”. According to
Hirschman (1958), economic growth is essentially an unbalanced dynamic process, in which
successive disequilibria produce the conditions for development in different sectors. In his
“unbalanced growth” theory, the productive structure is linked through forward and backward
linkages to downstream and upstream industries. These linkages represent physical relations
of supply and demand among sectors of the economy. Thus, backward linkages are associated
with the magnitude that each sector demands from other sectors of the economy, while
forward linkages are associated with the extension that each sector is demanded by other
sectors. In this dynamic, manufacturing industry is characterised by both strong backward and
forward linkages, enabling this sector to generate higher economies of scale with positive
effects in terms of productivity gains and cost savings in later stages of the production chain.
From this perspective, as pointed by Toner (1999), Hirschman focused particularly on the
intermediate and capital goods sectors while Rosenstein-Rodan and Nurkse focused
essentially on productivity growth in the consumer goods sector.

Furthermore, while also concentrating on the role of bottlenecks, external economies

and complementarities, Albert Hirschman qualifies the economic development “essentially as
the record of how one thing leads to another” involving not even physical relations of supply
and demand, but also technological linkages. It leads to the first insights on the concept of
spillover effects, which stems from manufacturing to the rest of the economy and is
approached by the contemporary economic developmental literature, e.g., the Kaldorian and
Neo-Schumpeterian strands. Additionally, the author’s opposition to Nurkse’s and
Rosenstein-Rodan’s strategies relies on the idea that too many financial resources and
planning efforts would be necessary to stimulate the economy, concluding that “if a country
was ready to apply the doctrine of balanced growth, then it would not be underdeveloped in
the first place” (Hirschman, 1958, p. 53-54).

14
 In this sense, Hirschman’s strategy rested on the idea that economic policy should
focus on specific industries, i.e. key sectors with a strong interdependence or linkages – both
backward and forward – with other sectors of the economy15. The backward linkage refers to
the potential of a sector to stimulate production and investment of sectors that provide its
inputs, whilst forward linkage relates to the capacity of a sector to induce productive activities
of sectors which demand its output. Therefore, key sectors, i.e. sectors with strong linkages
would be capable of generating higher economies of scale with positive effects in terms of
productivity gains and cost savings in later stages of the production chain. As a matter of fact,
unbalanced growth theory asserts that certain sectors, particularly inside the manufacturing
industries, are the main engines of growth16.

Like Albert Hirschman, Myrdal (1957) centralised his theory on the understanding that
economic development is intrinsically a process in disequilibria, breaking with the
neoclassical statement of “stable equilibrium”17. Thus, Myrdal’s theory of unbalanced growth
is centred on the concept of “cumulative causation” to analyse the problem of development
inequality between nations. Thus, in this dynamic, trade and economic relations between
developed and underdeveloped countries are discussed considering effects that arise from this
interaction and may negatively (“backwash effect”) or positively (“spread effect”) impact the
development of an underdeveloped economy. Furthermore, according to him, economic
development also involves not only economic relationships of supply and demand but also
institutional and political structures, denominated non-economic factors, which operating in a
process of cumulative causation reveals challenges to be faced by underdeveloped countries18.

In the Myrdal’s concept of “circular cumulative causation”, the main idea relies on the
fact that free market forces would tend generally to increase regional disparities. The assertion
made by Myrdal was important because, while international economic inequality grew and
became a common concern in many schools of thought, the neoclassical theory of
international trade insisted on the idea that there was a gradual equalisation tendency of factor

15
Input-output models made possible measurements of linkages. In fact, during the 1960s and 1970s, different
indicators of forward and backward linkages based on the Leontief inverse matrix were developed in order to
fulfil the theoretical approach.
16
According to Ancochea (2007), the selectivity of some sectors as proposed by unbalanced growth economists
received different names, such as the “propulsive industry” (Perroux, 1955), “leading sector” (Rostow, 1952,
1978), or the “development block” (Dahmen, 1987). For instance, see (Hirschman, 1987).
17
To Myrdal, neoclassical trade theories were “never developed to comprehend the reality of great and growing
economic inequalities and of the dynamic processes of under-development and development” (Myrdal, 1957,
p.51).
18
See also Ho (2004).

15
prices and income across countries. Even focused on social aspects of this cumulative
causation, Myrdal’s theory provided the fundamental framework for later complementary
heterodox theories, such as the Latin American Structuralist approach – with a strong
influence in Celso Furtado – and the Kaldorian theory which concentrated in the demand-
supply relationships to the manufacturing sector.

As a general link between all pioneers of economic development, although wrapped in

a critical assessment of the level of state intervention in the economy, both approaches –
“balanced growth theory” and “unbalanced growth theory” – directly or indirectly paid
attention to the role of the industrialisation process as a way to overcome underdevelopment.
In the context of Latin American development problems, it is important to highlight that
ECLAC participated actively in these discussions providing important contributions notably
from the works of Raúl Prebisch, Celso Furtado and Aníbal Pinto. From this background, the
following section seeks to provide the main reflections on the obstacles encountered by
underdeveloped countries in the face of the absence of a dynamic industrial structure.

1.2. Latin American Structuralism: Linking underdevelopment to the centre-

periphery paradigm

In modern economics, Latin American structuralism should be placed in a

methodological tradition which has its origin in the Raul Prebisch’s (1949) study “El
desarrollo económico de la América Latina y algunos de sus principales problemas”. With
Prebisch leading a group of outstanding economists, the Economic Commission for Latin
America and the Caribbean (ECLAC) sparked remarkable insights and explanations regarding
the causes of Latin American underdevelopment19. Latin American structuralist writers
challenged the neoclassical theory through a critique of the prevailing international trade and
proposed a theory of peripheral capitalism incorporating core elements presented in the
French20 and Anglo-Saxon structuralist traditions, as well as in Keynesian thinking21 (Furtado,
1967; Love, 1995, 1996; Sunkel, 1989, Palma, 1987, Blankenburg, Palma & Tregenna, 2008).

19
The French structuralism represented by innovative contributions of François Perroux (1939) defined
structural economics as a science which analyses the relations characteristic of an economic system situated in
time and space. According to him, economic analysis should incorporate institutions and structures over time.
See Blankenburg, Palma & Tregenna (2008).
20
Perroux (1950) anticipated important elements of ECLAC’s theory, particularly with regards to the centre-
periphery model, which was an extension of the concept of economic systems where “the economic world is

16
 Based on this theoretical background, the basic analytical components of ECLAC and
other Latin American structuralists were grounded in historical methodology, the study of
domestic determinants of economic growth and technological progress, as well as an
evaluation of arguments in favour and against state intervention (Bielschowsky, 1998). In this
sense, many prominent works followed ECLAC thinking and provided important insights,
criticisms and complementarities for the understanding of the Latin American
underdevelopment. Through a sharp critique of the neoclassical economics and its idea that
specialisation based on comparative advantages, whatever its nature, was a superior solution
for economic growth, the Latin American Structuralist school gave life to an important
interpretation where the productive structure matters to the pace and scope of the development
process. Comparing commodity-producers economies and industrialised countries, Prebisch
(1949) noted that productivity was essentially higher in the manufacturing sector than in the
primary activity. This dichotomy in levels of productivity between the productive structure of
developed (centre) and underdeveloped (periphery) countries, the so-called structural
heterogeneity, was also approached by Furtado (1959, 1961) and Pinto22 (1965, 1970).

For Furtado (1961), the mainspring of the capitalist development is technological

progress through a process of incorporation and diffusion of new techniques with a
consequent increase in the production and productivity23. Therefore, underdevelopment is
seen as a partial and blocked version of the development, either because of the uneven spread

conceptualised in terms of hidden or explicit relationships of ‘force’, ‘power’ and ‘constraints’ between
dominant and dominated entities” (Blankenburg, Palma & Tregenna, 2008, p. 2). Perroux’s theory is based on
different analytical levels, i.e., markets, firms and international economy and clarifies interactions between
different entities with distinctive structures of power and the consequences in terms of trade and finance.
According to Perroux growth does not appear everywhere and all at once; “it appears in points or growth poles
with variable intensities; it spreads along diverse channels and has varying terminal effects for the whole of the
economy” (Perroux, 1955, p 308). Thus, the intensity and magnitude of growth poles are determined by the
fundamental role played by propulsive industries, which are in turn highly innovative and, according to Perroux,
sources of technological progress. These industries constitute the pillar of economic development and generate
positive effects (spread effects) in other regions, such as the increase in income and employment and at the same
time produce structural change through economic growth. Moreover, through the development pole theory,
Perroux also exerted a fundamental influence on the ECLAC’s division of the world between centre and
periphery and Furtado’s early work including his doctoral dissertation at Sorbonne. See Furtado (1995).
21
The Latin American structuralist school was also concerned, to a greater or lesser extent, with a coalition of
the industrial bourgeoisie, the middle class and the urban working class, under the coordination of the
developmental state to propel the economic development. According to the structuralist approach, it is the
developmental state which is the main agent of change since it is the only institution capable of transcending
sectional interests and thereby able to pursue the national interest (Kay, 1989).
22
Although the concept of structural heterogeneity was a central element in works of Raúl Prebisch or in Celso
Furtado in the form of “dualism”, it was with Anibal Pinto that the concept of structural heterogeneity solidifies
during the 1970s. See for instance Pinto (1970, 1971, 1976).
23
In a complementary approach, Tavares (1972, p. 50) highlights the problem to create technical progress
endogenously and the consolidation of a diversified productive structure with increasing share of national
content in domestic production.

17
of technical progress or the limited transmission of productivity gains to wages. According to
him, in developed countries, dynamic growth is headed by technical progress while in
underdeveloped countries it is determined primarily by external demand for imports. While
the centre countries internalised the new technology by developing an industrial capital goods
sector and by spreading the improved technology to all economic sectors, the periphery
remained dependent on imported technology which in turn was mainly confined to the
primary export sector. Consequently, a sizeable low-productivity pre-capitalist sector
continued to survive in the periphery producing a continuous surplus of labour and
consequently keeping wages low. Without the processes of industrialisation, the asymmetry
between the centre and periphery would not only perpetuate but also deepen.

In relation to this dynamic, Pinto (1965) highlighted the persistence of structural

heterogeneity. According to Pinto, developed countries had a much more homogeneous level
of productivity than the periphery. The heterogeneity in underdeveloped countries – expressed
in sectors where the productivity is high or “normal” with others where the productivity is
lower or several times lower in the same economic structure – would generate problems of
underemployment in face of the occupational mismatch. As the occupational structure is a
mirror of the productive one, economies with high productivity tend to generate more
employment while structures with very low productivity tend to generate underemployment
(Rodriguez, 2006).

Pinto (1970) identifies three levels of productivity in Latin America: i) the primary
sector with low productivity and low earnings, keeping similarities with the prevailing forms
of production since the colonial period; ii) the middle sector that is neither the upper end nor
the lower end and thus is near to the average of the economy as a whole; and iii) the modern
sector with high productivity levels and gains, which are similar to the average of developed
economies. Given these different levels of productivity, the rate of technical progress
incorporation and productivity increase would be significantly higher in central economies
vis-à-vis peripheral economies, which in turn are specialised in primary products. In this way,
a shift from less productive sectors to those with higher productivity – notably the
manufacturing industry – would promote an increase in the aggregate productivity, a stimulus
in the technological diffusion and an increase in real wages24.

24
During the 1980s, Fernando Fajnzylber provided important contributions in relation to the underdevelopment
theory emphasising the Latin American bottlenecks, especially regarding technical progress and productivity.

18
 While various writers contributed to the Latin American structuralist paradigm,
Prebisch’s original ideas were pivotal in launching a critical perspective on the neoclassical
approach to the mutual profitability of free trade between developed and developing
countries, whose influence was very remarkable in Latin America. In his thinking, a key
structural economic characteristic of peripheral economies refers to the deterioration in their
terms of trade over time due different income-elasticity of demand – also known as “dynamic
disparity of demand”. Thus, contrary to what the comparative advantage theory suggested,
prices of primary products produced and exported by peripheral countries, such as in Latin
America, tended to present an antagonistic evolution when compared to prices of
manufactured products exported by industrialised countries. This means that the centre’s
imports of primary products from periphery rise at a lower rate than its national income, while
the periphery’s imports of manufactured goods from the centre grow at a faster rate than its
income. Since demand for manufactured goods increases more rapidly than the demand for
primary goods, the well-known Engel’s law, there is a tendency to deteriorate the terms of
trade of those economies specialised in the production and export of primary goods in
comparison to central industrialised economies.

In other words, prices of manufactured goods would be structurally higher in relation

to primary products. This meant that peripheral economies would have to export more to
achieve the same value of industrial exports over time. In central economies adjustments
along the global economic cycle are made through export quantities, due to the high level of
industrialisation. On the other hand, in peripheral economies, adjustments occur through
export prices due to the primary specialisation25. In contrast to the free trade doctrine, these
movements would be gradually accentuated in the absence of a dynamic industry. Thus,

Fajnzylber (1983) explained the low technological dynamism that characterised Latin American industrialisation
through a convergence of structuralist thinking, the French regulation school and evolutionary economics.
According to Fajnzylber, an economy which does not have an “endogenous nucleus of technological dynamism”
cannot overcome underdevelopment. Moreover, since the sector of capital goods materially incorporates
technological progress, policies to strengthen this sector should be carried out to establish an endogenous
nucleus of technological dynamism and stimulate the diffusion of technology to other sectors as well as reverse
the Latin American structural deficit in the current account. Without a strong developmental state to build this
“endogenous nucleus of technological dynamism” as occurred in developed countries and in late industrialising
countries (LICs) such as East Asia, including Japan, and in absence of an industrial vocation, i.e, business
leadership capable of inducing transnational companies to build this “endogenous nucleus of technological
dynamism”, underdevelopment would be maintained. According to Fajnzylber, in Latin America, the problem
with transnational companies (TCs) was the establishment of productive structures based on technology
transferred by headquarters which therefore did not contribute to the process of technological innovation. To
clarify the understanding of how to overcome the inheritance of past mistakes, the author defends that Latin
America should not only focus the macroeconomic stabilisation and debt reduction, but also push the
technological frontier inducing TCs to adopt innovative domestic behaviour.
25
In a complementary way, see Prebisch (1950, 1959) and Singer (1950).

19
overcoming underdevelopment would not be possible through the international division of
labour, in which peripheral countries would be doomed to a specialisation in primary
products. In this sense, industrialisation was seen as a way to modify this process. Thus,
through a productivity increase, the deterioration of the terms of trade could be reduced, the
technological progress incorporated and a process of income distribution promoted26.

The limits of spontaneous industrialisation in developing countries clearly revealed the

need for an active state intervention in the industrialisation process. Although the state should
have the capacity to promote particular sectors through the creation of public companies, it
should especially focus on planning (Ancochea, 2007). For this purpose, many other Latin
American “structuralists” sought to map different stages of the industrialisation process in a
similar manner, often distinguishing between the so-called stages of industrialisation, i.e.
internalising the production of consumer goods, consumer durables, intermediary inputs and
capital goods. Therefore, it is not by chance that during 1950s and 1960s this argument gave
rise to the support of import substitution industrialisation (ISI) model, principally in Latin
America. ISI was a trade and economic policy programme based on the premise that a country
should attempt to reduce its foreign dependency through the local production of
manufacturing goods. For this purpose, the developmental state would coordinate the process
of industrialisation from light to heavy industries through imports of intermediary and capital
goods necessary to obtain a diversified and interdependent productive structure. In each stage
of the industrialisation process, a gradual replacement of imported goods for domestic
production would spread technological and productivity gains over the economy.

Broadly speaking, the idea expressed by the Latin American Structuralism was that,
despite the spread of modernity, backwardness and wide differences in labour productivity
between economic sectors and subsectors, and between regions and segments of the
population, tended to be maintained and sometimes expanded (Bielschowsky, 2009).
According to these authors developing countries could be characterised by a dual structure
where a late agricultural sector and a modern industrial sector would coexist. The
manufacturing importance vis-à-vis a concentration in primary commodity exports was a
central concern of the structuralist approach associated with the ECLAC. In this way,
industrialisation is seen as the only way for developing countries to catch up. As a general

26
These dynamics were also pointed out also by Furtado (1959). In this sense, Furtado’s works are closely
connected to Prebisch’s, especially regarding endogenous dimensions of underdevelopment and its determinants.
In general, the economic thinking of ECLAC and Celso Furtado addressed particularly the different stages of
industrialisation in a similar manner, often distinguishing between the so-called phases of industrialisation.

20
trend, these aspects would be a central issue in the Latin American structuralist strand. The
Kaldorian theory which concentrated on the demand-supply relationships in the
manufacturing sector complements this view giving further elements to explore the
importance of the industrialisation process, more specifically the manufacturing sector.

2. Kaldor’s stylised facts and the contemporaneous debate

Kaldor showed the central role that manufacturing sector plays in economic growth in
two lectures: one in Cambridge in 1966 entitled Causes of the Slow Rate of Growth of the
United Kingdom (Kaldor, 1966); the other at Cornell University in the same year, published
as Strategic Factors in Economic Development (Kaldor, 1967). Inspired by studies of Allyn
Young, Gunnar Myrdal and Adam Smith and recovering the empirical regularities pointed out
by Kuznets, Rostow, Chenery and Syrquin, Nicholas Kaldor argued that it is not possible to
understand the development process and growth rate differences between countries without
taking a sectoral approach. In a complementary line of research to Furtado, Hirschman,
Rosenstein-Rodan and Prebisch, Kaldor noted that the manufacturing sector was imbued with
special growth-enhancing properties that triggers a process of cumulative causation that are
not shared by other sectors27.

Kaldor’s view offered support for the special role played by the manufacturing sector
given some special characteristics which forwarded itself to a special theoretical framework
for understanding the causal relationships between industrial development and economic
growth. Kaldor introduced the concept of dynamic economies of scale, such that the faster the
growth of manufacturing output, the faster the growth of manufacturing productivity. He
attributed these dynamic economies to Arrow’s (1962) notion of ‘learning by doing’ and
argued that this occurred principally in manufacturing and not in services or agriculture. In
this way, challenging the neoclassical statement of constant returns to scale across sectors,
Kaldor noted that poor developing countries tend to specialise in land-based sectors –

27
In this case, Kaldor’s explanation is highly influenced by studies of Allyn Young, Gunnar Myrdal and Adam
Smith. In his own words, “to explain why certain regions have become highly industrialised, while others have
not we must introduce quite different kinds of considerations – what Myrdal (1957) called the principle of
‘circular cumulative causation’. This is nothing but the existence of increasing returns to scale – using the term
in the broadest sense – in processing activities. (…) As Allyn Young (1928) pointed out in a famous paper,
Adam Smith’s principle of ‘division of labour’ operates through the constant sub-division of industries, the
emergence of new kinds of specialised firms, of steadily increasing differentiation – more than through the
expansion in the size of the individual plant or the individual firm” (Kaldor, 1970, p. 340). All these assumptions
stimulated many studies which tried to provide a formalisation of Kaldor’s ideas of economic growth. In a well-
known work, Dixon & Thirlwall (1975) provided a systematic formalisation of the cumulative causation model.
Appendix 1.1 addresses this issue.

21
agriculture and mining – subject to diminishing returns, while richer developed countries
specialise in increasing returns activities (static and dynamics) such as manufacturing and
sophisticated service activities associated with them including banking, finance and insurance
(Thirlwall, 2013).

Unlike the neoclassical theory of growth, i.e. theories of exogenous and endogenous
growth developed since Solow (1950), where the theoretical framework relies exclusively on
the supply-side vision, Kaldor’s argument considers the demand side28, particularly the role of
aggregate demand, to explain the economic dynamic. On the basis of the interaction involving
demand and supply conditions in agriculture, manufacturing and services, Kaldor derived
generalisations concerning the relationship between the growth of output, employment and
productivity in different sectors of the economy. These theoretical formulations became
known as “Kaldor‟s growth laws” or “stylised facts” and became an important turning point
in the literature of economic growth.

Many empirical tests centred on Kaldor’s laws to explain the differences in

international growth rates were made and replicated for different groups of countries and
periods of time since then, following the evolution of econometric methods for static- and
dynamic-model data panels. Under this framework, many contemporaneous and prominent
Kaldorian analyses emphasised a strong relationship between changes in the sectoral
composition of an economy and economic growth. In a complementary way, the
contemporary developmental literature has emphasised the bundle of interactions that
connects manufacturing and economic growth, particularly considering distinct stages of
development and the intersectoral relationship between services and manufacturing.

2.1. Manufacturing as engine of growth: from industrialisation to de-

industrialisation

Over the years, the complex relationship between production structure and economic
growth has been the subject of considerable debate among economists. The ever-increasing
body of policy reports, academic papers and manufacturing national strategies covering
economies with different income levels have showed the virtues of manufacturing. The so-
called stylised facts are supported by important empirical regularities that recent research on

28
Different from the neoclassical view, in the Kaldorian framework the production used in a modern capitalist
economy is itself goods produced within the system given the effective demand.

22
patterns of economic growth has highlighted29. In this sense, recovering the stylised facts
formulated by Kaldor, the first law states that “manufacturing is the engine of growth”. It
means that the faster the growth rate in manufacturing output is, the faster the growth rate of
the economy as a whole will be.

In dialogue with this hypothesis, using different econometric methods such as cross-
section, panel and time-series data analysis (for some individual countries), many empirical
studies were strongly supportive of Kaldor’s first law (Thirlwall, 2013). Kaldorian analyses
showed that regression coefficients were statistically significant at the 95 per cent confidence
level and above, and are strong in the case of East Asian and Latin American economies
(Hansen and Zhang, 1996; Mamgain, 1999; Libanio, 2006; Wells & Thirlwall, 2003;
Dasgupta & Singh, 2006). Unlike manufacturing, in agriculture and services the relation
between GDP growth and the growth of other sectors does not exist, at least not in the same
magnitude. In fact, studies did not find a correlation between the growth of agriculture and the
growth of GDP in a causal sense. Furthermore, econometric tests have showed a strong
negative correlation between GDP growth and the excess of agricultural growth over non-
agricultural growth. Additionally, the coefficient for services was lower than those of
manufacturing (Thirlwall, 2013).

Other empirical studies have been supportive of the Kaldorian literature. Rodrik
(2007) found that since 1960 the economic growth of developing countries economic is
strongly associated with the development of modern industrial sectors. Szirmai & Verspagen
(2011) and Szirmai (2012), taking into account a sample of 21 advanced economies and 67
developing countries in the past 50 years, found that manufacturing has functioned as an
important engine of growth in developing countries. These analyses have been in dialogue
with Fagerberg & Verspagen’s (1999) work which noted that developing countries are those
that benefit more from expansions of the manufacturing sector. They test the importance of
the most dynamic segments of manufacturing in economic growth and confirm the
importance of the flexibility to shift towards manufacturing productions.

The following Figure (1.1) illustrates a positive correlation between the rate of GDP
growth and the increase in manufacturing’s share in the global manufacturing GDP. Those
countries that achieved the fastest economic growth during the period are the countries where

29
Kaldor developed this stylised fact through a study of developed countries over the period 1952-54 to 1963-64
(Kaldor, 1966).

23
the increase shift towards manufacturing has been most intensive. The numbers bellow
illustrates the dichotomy between Asia and Latin America in terms of strategy of economic
development. In Asia, the share of manufacturing in GDP grew at much higher rates resulting
in significant increases in the GDP. In an opposite trend, Latin America and developed
countries showed exactly the reverse path, de-industrialising and achieving very modest rates
of GDP per capita growth.

Figure 1.1 – Annual GDP growth and change in the share of manufacturing value added
in the global manufacturing GDP, 1970-2010

9%
Change in the share of manufacturing value added in

7% Korea
the global manufacturing GDP

5% Nics 1
Taiwan
Singapore
Nics 2 South-Eastern Asia China
3% India

1% Mexico
Brazil
Latin America
Japan Chile
-1% Argentina US
Developed economies
Hong Kong
-3%
0% 1% 2% 3% 4% 5% 6% 7% 8% 9%

Annual GDP growth

Source: Own calculations based on data from GGDC 10-Sector Database and UNCTAD.
Note: GDP = Gross Domestic Product. According to UNCTAD database, developed countries include Australia,
Austria, Belgium, Canada, Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece,
Hong Kong, Iceland, Ireland, Israel, Italy, Japan, Luxembourg, Malta, Netherlands, New Zealand, Norway,
Portugal, San Marino, Singapore, Slovakia, Slovenia, South Korea, Spain, Sweden, Switzerland, Taiwan, United
Kingdom, United States. Nics 1: Korea, Taiwan, Hong Kong, and Singapore. Nics 2: Malaysia, Thailand,
Indonesia, and the Philippines. South-Eastern Asia: Brunei Darussalan, Cambodia, Indonesia, Lao, Malaysia,
Myanmar, Philippines, Singapore, Thailand, Timor-Leste, Vietnam. R2 = 74%; all variables are significant at the
5% level.

Nevertheless, Szirmai & Verspagen (2011), Fagerberg & Verspagen (1999) and
Szirmai (2012) have noted that even though manufacturing plays a central role in the process
of economic development, some countries have presented a decreasing share of
manufacturing followed by an increasing share of services in their sectoral composition. With
the intention of evaluating this trend, an important issue related to the stage of development
should be considered. As Rowthorn & Wells (1987) noted in a seminal study, countries

24
follow a broadly similar trajectory of economic development. Thus, as development gets
under way, in low-income countries, the share of agriculture in the value added and
employment fall and there is a rapid increase in the share of manufacturing. The transition
from a reasonable diversified structure of production to a mature economy is known as
‘industrialisation’ and mostly represented by the so-called middle-income countries. At a
certain point, when the economy reaches a certain level of income per capita, the share of
manufacturing stabilises and then starts to fall back taking the form of an ‘inverted-U’
(Rowthorn, 1994) 30. This trend is also followed by a corresponding increase in the share of
services in national employment31 and value-added, and is often described as natural process
of de-industrialisation32.

Figure 1.2 – The ‘inverted-U’ relationship between manufacturing and income per
capita
50
45
Middle Income
40
35
Manufacturing Share (%)

30
25
20 Low Income
High Income
15
10
5
0
0 5,000 10,000 15,000 20,000 25,000 30,000

GDP per capita (Geary-Khamis dollars - 1990)

Source: Mckinsey (2012).

30
Rowthorn & Wells (1987) defined de-industrialisation as a decline in manufacturing employment first in
relative terms and then, at least in some countries, also in absolute terms. Tregenna (2009) defined the de-
industrialisation as the consistent reduction of both the share of employment and the value added of
manufacturing industry in total employment and GDP, respectively. In this section “de-industrialisation” will be
analysed solely from the point of view of manufacturing value added.
31
In terms of employment, the statistics also indicate that in most countries the growth rate of labour
productivity in the manufacturing sector has been faster than in services and in the economy as a whole.
Consequently, the relative decline of manufacturing employment has been mainly the result of rapid productivity
growth in this sector (Rowthorn & Coutts, 2013b). See also Rowthorn & Coutts (2013a, 2013c).
32
The “inverted U” curve also reflects the consumption pattern in each stage of the income per capita. During
early stages of the economic development, i.e. low-income countries, agriculture presents the biggest share in the
economy because food represents most of the household consumption. Thus, as countries go through the
economic development inputs such as steel and cement are needed for infrastructure as well as machinery and
transportation equipment for the productive process of industries. When the country reaches the stage of middle-
income economy, a higher income per capita triggers additional forms of consumption, particularly in services.
An increase in spending on services such as education, health care, travel, banking and many others begins to
take up gradually a higher share of incremental income.

25
 Many authors such as Robert Rowthorn have done seminal studies to understand this
natural shift in terms of sectoral composition, contrasting the so-called natural de-
industrialisation, i.e. the natural consequence of the industrial dynamism in an already
developed economy, with the so-called negative de-industrialisation, which is defined as “a
product of economic failure and occurs when industry is in severe difficulties and the general
performance of the economy is poor” (Rowthorn & Wells, 1987, p.9). Through this line of
thinking, Rowthorn & Coutts (2004), Palma (2005), and Pieper (2003) stated that several
developing countries are de-industrialising at a much lower level of per capita income than
observed historically in today’s developed countries. In order to consider if the rise of
services is a symptom of economic failure and a harbinger of impending impoverishment or if
the decline of manufacturing might be seen as something natural or even as a sign of
development, some considerations must be taken into account.

2.2.Statistical illusions

Firstly, a non-negligible part of the extent of de-industrialisation – decreasing relative

importance of manufacturing – seems to be due statistical illusions, in the sense that it reflects
changes in statistical classification rather than changes in real activities. Therefore, the extent
of deindustrialisation has been overestimated due to the outsourcing of some services that
used to be provided in-house by manufacturing firms and thus were counted as manufacturing
output, e.g. catering and cleaning, research, design, IT, accounting, telecommunications,
engineering activities, logistics and legal services (Palma, 2005, Rowthorn & Coutts, 2004)33.

Although these outsourced activities are still the same, now they are counted as part of
services output, rather than manufacturing output. Consequently, services become more
important without a real increase. In this manner, experts agree the outsourcing effect has
been a considerable source of de-industrialisation in middle/high-income countries principally
from 1980s when the neoliberal policies took momentum. This potential overestimation of
manufacturing’s decline in industrialised countries is also discussed by McCarthy &
Anagnostou (2004) and Vittucci (2008). Another handful of theorists such as Laplane & Sarti
(1997), Carneiro (2008) and Rocha (2011) have highlighted that this statistical effect also
affected Latin American countries. These studies noted that neoliberal policies, particularly
regarding market deregulation and trade liberalisation, led a wide range of manufacturing

33
Rowthorn & Coutts (2004) named it as “statistical artefact”.

26
sectors to undertake defensive adjustments characterised by the process of outsourcing which
shifted value added from manufacturing to services, contributing to statistical illusions
outlined above34.

2.3.Servicification

Secondly, although the explanation for the premature de-industrialisation outlined

above obscure the traditional distinction between services and manufacturing, empirical
studies have shown that middle income countries – characterised by a premature de-
industrialisation – and high-income countries tends to present strong correlation between GDP
growth and the increasing share of services, particularly more sophisticated and high-value-
added service activities, which in turn has shown a strong dependence on manufacturing.
Although economists have also observed the difficulty of measuring interactions between
services and manufacturing, due to the process of outsourcing, input-output analyses revealed
strong intersectoral interactions and interdepencies between these two sectors35 (Lodefalk,
2010; Nordås & Kim, 2013).

Pilat & Wolfl (2005) quantified the interaction between manufacturing and services in
the economy. This analysis has showed that the character of European manufacturing seems
to change over the years interacting more with service industries than before, that is,
manufacturing has been using more intermediate services and employs a rising number
service-related workers. This trend has been not only a result of outsourcing that
overestimates the use of services in manufacturing, but also the increasing interdependence
between some knowledge-intensive services and manufactured products. In the case of the
latter, the most remarkable example is the service subsector of Information and

34
Additionally, as pointed out by Chang (2014, p. 261), “seeing the share of manufacturing in their output
falling, some manufacturing firms have applied to be reclassified as service firms”. In this sense, individual
manufacturing firms have been reclassified as service firms, even though they still engage in some
manufacturing production. According to the UK government’s Department for Business, Enterprise and
Regulatory Reform (BERR) up to 10% of the fall in manufacturing employment between 1998 and 2006 in the
UK may have been due to this reclassification effect (Chang, 2012).
35
As showed by Mckinsey (2012, p. 7), depending on the segment, 30 to 55 percent of manufacturing jobs in
advanced economies are service-type functions, and service inputs make up 20 to 25 percent of manufacturing
output. Moreover, for every dollar of output, US manufacturers use 19 cents of service inputs, creating $900
billion a year in demand of services, while services create $ 1.4 trillion in US manufacturing demand. In China
this interdependence is around $500 billion in services demand and $600 billion in manufacturing demand.
Additionally, although manufacturing exports drives more than 80 percent of German exports, services and
manufacturing contribute nearly equal shares of value added to the country’s total exports.

27
Communications Technology (ICT). However, although services now contribute more as
providers of intermediate input and service related workers to the performance of other
industries, their role remains more limited than those of the manufacturing sector. Park &
Chan (1989), Rocha et al. (2014) and Magacho et al. (2014) have confirmed Hirschman’s
intuition that manufacturing has larger multiplier indices than other sectors.

As suggested by Guerrieri & Meliciani (2005), the country’s capacity to develop its
services depends on the specific structural and technological composition of its manufacturing
sector. Some knowledge-intensive services are spin-offs from manufacturing production,
given that the manufacturing sector itself has been the key source of new productive
knowledge to the rest of the economy. Moreover, the manufacturing sector creates demand for
the growth of high-productivity services such as finance, engineering, design, accounting,
consultancy, telecommunication and transport. Therefore, services growth is closely
connected to the manufacturing sector and consequently a weakening manufacturing base
would eventually lead to a decline in the quality of those services (Chang, 2014).

2.4.Productive Fragmentation

Last but not the least, in recent decades, the geography of industrial production has
gone through a revolution, with the breaking-down of the ‘value chain’ by multiproduct
Transnational Corporations (TNCs), leading to the process of de-industrialisation in advanced
economies. Since around the year 2000, the relocation of labour-intensive assembly-end part
of the value chain to developing countries has significantly fragmented the international
production in coordinated networks also known as global value chains (GVC) (Unctad, 2013).
In this new productive environment countries not only specialise in terms of industries but
also of activities carried out for production of a particular product. Therefore, the value of a
final product sold on the market embodies more and more value added from other countries
due an increasing process of international vertical specialization (Stehrer, 2013).

In the background to this process, the dominant view stresses that international
productive fragmentation is an important source of firm efficiency through improvements in
competitiveness both in domestic and international markets and by cost savings. Through this
productive fragmentation, companies could focus on the most dynamic levels of the global
value chain such as R&D, product design, marketing and services. Furthermore, companies

28
responsible for tangible activities would also benefit from the learning flow with global
buyers that would improve their production process and product upgrading. Moreover,
countries specialised in production activities could also move to levels with higher value
added in the global value chain. Therefore, the stage of economic development would be
closed related to the process of acquiring new functions which generate higher incomes (and,
conversely, ceasing to perform low-income activities). At the country level, the process of
productive re-allocation to low and middle income nations would give to companies more
access to new consumer markets. In this process, low-income countries could also experience
rapid economic growth processes stemming from the structural transformation of the large-
scale migration of workers from subsistence sector to manufacturing. Middle-income
countries in turn could consolidate their productive structure and, in the course of economic
development, move up in the global value chain.

Figure 1.3 –Value added in the Global Value Chain

Source: Elaborated by the author based on OECD/WTO (2013)

However, many studies have stressed that the process of embracing the global value
chain has generated doubtful results. Thus, the decline of manufacturing in advanced
countries like the US and EU members as well as in Latin American economies due the
process of productive fragmentation nowadays concerns policy-makers and scholars
particularly with regards to the risks of losing the interdependence between production
activities and technological innovation. The proponents of the manufacturing renaissance

29
have been warning that a knowledge economy which loses interaction with its productive
structure may lose the capacity to innovate next-generation technologies and products.

Therefore, prominent scholars have pointed out the fact that the off-shoring of
production operations is all too often followed by a deterioration of the so-called “industrial
commons”36, i.e. damage to other parts of the industrial system which includes “reduced
operations by local suppliers of materials, components and production technologies; a decline
in process engineering skills, manufacturing know-how and leadership; a deterioration of
prototyping, test-bed and pilot-manufacturing infrastructure” (O’Sullivan & Mitchell, 2013, p.
43). In this way, losing “industrial commons” may cause a decline in important technological
capabilities that stems from the interconnection between product development, next-
generation production technologies and process engineering. Consequently, due to the fact
that cutting-edge technologies often rely on elements of the “industrial commons”
underpinned by dynamic manufacturing sectors, this process risks reducing the country’s
capacity to compete in some of the most important new industries.

Pisano & Shih (2009) exemplified this point illustrating that US companies
overestimated the advantages of outsourcing development and manufacturing work to
specialists abroad and cutting their spending on basic research. Moreover, the outsourcing and
off-shoring slogan comprised not only low-value tasks like simple assembly or circuit-board
stuffing, but also sophisticated engineering and manufacturing capabilities that support
innovation in a wide range of products and process. Therefore, due to the process of losing
knowledge, skilled people, and supplier infrastructure, the US reduced its capacity to
manufacture a vast range of cutting-edge products.

According to Ezzel & Atkinson (2011), the current challenges faced by some
industries in the US illustrate this point well. In the case of the solar panel industry, they
found that in order to lower costs, US companies re-allocated semiconductor foundries to
Asian countries such as Japan, India, Taiwan, South Korea, and especially China. However, it
not only deteriorated the silicon-processing and thin-film-deposition capabilities in the US but
also undermined the process of manufacturing solar panels. In the same way, US companies
are falling behind in the software industry. Initially, companies outsourced only relatively
36
Pisano & Shih (2009) coined the term ‘industrial commons’ based on the Marshall study (1920). Moreover,
based on the seminal Marshall study on industrial districts, Kaldor also pointed out that one of the most
remarkable features of industrial districts is the special concentration of manufacturing activity that culminates in
important advantages for industries as a whole due to the availability of specialised skills and ready
communication of trade and managerial know-how (Kaldor, 1970, 1972).

30
mundane code-writing projects to Indian firms to lower software-development costs.
However, over the years, Indian companies developed their own software-engineering
capabilities and started to attract more complex activities to India like developing
architectural specifications and writing sophisticated firmware and device drivers (Pisano &
Shih, 2009). The consequence of these trends is the slump of the US industrial base, “the
hollowing out of advanced production supply chains, and the loss, for many US industries, of
their industrial commons” (Ezzel & Atkinson, 2011, p. 15).

Furthermore, the critics of productive fragmentation have widely documented that

production and R&D cannot be disassociated because the proximity of research, development,
and manufacturing is very important to leading-edge manufacturers. As stated by Berger
(2011), there is a close connection between R&D and manufacturing. In this interdependence,
for example, strategies for making processes more efficient stem from the close synergy
involving R&D engineers and manufacturing. Therefore, off-shoring manufacturing is pulling
high-end design and R&D capabilities out of the US and consequently diminishing the
country’s capacity to create new high-tech products. For this reason, Ezzel & Atkinson (2011)
highlight that nowadays 90 percent of all electronics R&D now takes place in Asia due to
some extent to the scale of production needed to be able to afford general R&D. They also
point out that from 1998 to 2007, US corporations invested more than 2.65 times faster in
overseas R&D than domestically. Furthermore, almost “every US brand of notebook
computer, except Apple, is now designed in Asia, and the same is true for most cell phones
and many other handheld electronic devices” (Pisano & Shih, 2009). Thus, these examples
show that R&D, design and manufacturing cannot be disassociated.

For this reason, especially after the outbreak of the global crisis in advanced countries,
particularly the US but also in European economies, turned attention particularly to the need
of re-strengthening the manufacturing base together with the objective to ‘bring
manufacturing back home’, which was lost during the productive fragmentation era that
stimulated outsourcing and off-shoring practices. The European Commission has set an
ambitious goal aiming for a manufacturing share of 20% in 2020 (which is now around 15%)
and there is an active debate in the US today about the need of re-industrialisation to
guarantee further economic growth and high-quality job creation (Stehrer, 2013)37.

37
See European Commission (2010).

31
 3. Manufacturing as the main source of productivity growth

Through a sharp critique of the neoclassical theory, Kaldor postulated with regards to
productivity growth in the manufacturing sector. The origin of Kaldor’s second law is an
observation made by Verdoorn in 1949 that there is a positive relationship between the rate of
growth of output in manufacturing and the rate of growth of labour productivity in
manufacturing. Influenced by Young (1928), Kaldor noted that increasing returns to scale are
pervasive to manufacturing activities. In other words, according to Kaldor, the rate of
productivity growth in the manufacturing sector depends on its rate of output growth due to
the operation, within this sector, of static and, above all, dynamic returns to scale38 which are
also a source of technological progress (Kaldor, 1981). Static returns are mainly related to
economies of scale internal to the firm, where the large scale of production allows a reduction
on the average cost. Dynamic returns refer to the induced effect that output growth has on
capital accumulation through increased productivity derived from learning–by-doing,
technological change, external economies in production, and so forth (Libanio, 2006).

In this way, by highlighting the relevance of the sectoral composition to economic

growth, it is possible to verify that primary production and services do not possess the same
properties as the manufacturing sector. It means that the first one tends to present lower
returns to scale while in the second one, the scale of production tends to keep constant returns.
It happens because the returns tend to be higher when the production scale is greater, as in the
manufacturing sector. Therefore, there are several studies that tested the Kaldor-Verdoom law
over many different time periods and across variety of countries and industries. The causal
relation between the growth of manufacturing output and labour productivity growth in
manufacturing as a result of static and dynamic returns to scale has been tested for developed
countries, such as United Kingdom (Hildreth, 1989; Harris & Lau, 1998), the US (McCombie
& De Ridder, 1983; Bernat, 1996), Japan (Knell, 2004), the European Community (Fingleton
& McCombie, 1998; Angeriz, McCombie & Roberts, 2009) and developing economies,
particularly for Asia (Timmer & Szirmai, 2000) and Latin America (Libanio, 2006). The
results indicate substantial returns to scale, especially dynamic increasing returns, in

38
In his seminal study, Verdoorn (1949) demonstrated the existence of increasing returns both across industries
within one country and in total industry across countries. Verdoorn was a member of the Research and Planning
Division of the Economic Commission for Europe in Geneva, directed by Kaldor between 1947 and 1949, but
did not receive widespread recognition until 1966, when Nicholas Kaldor explicitly referred to it and coined the
term Verdoorn’s Law in his Cambridge Inaugural Lecture (Kaldor, 1966).

32
manufacturing industry. When the same tests are fitted to other activities, there is no evidence
of increasing returns in agriculture and services, especially in developing economies.

Another important stylised fact is Kaldor’s third law, which holds that the faster the
growth of manufacturing output, the greater the rate of labour transference from other sectors
of the economy (where productivity is lower) to manufacturing industries (where productivity
is higher). Thus, overall productivity growth is positively related to manufacturing output
growth and negatively related to employment in non-manufacturing sectors. Moreover, the
manufacturing sector increases the productivity of the system as a whole because it withdraws
labour from the agricultural sector where there is a lower marginal product. Therefore, when
the surplus of labour becomes exhausted in the agriculture sector, and levels of productivity
tend to equalise across sectors, the degree of overall productivity growth induced by
manufacturing output growth is likely to slow down39. In this sense, Kaldor stresses that this
process is characteristic of economies in transition from “immaturity” to “maturity”, where an
“immature” economy is defined by a large amount of labour available to be transferred to
industry (Thirwall, 2013). This is why growth rates tend to be fastest in the initial stage of
development, and decelerate as economies mature and become more service-oriented.

Additionally, over the years many empirical studies have tested Kaldor’s third law.
Even in face of the difficulty of measuring labour productivity growth in the non-
manufacturing sectors, particularly service type activities and public goods – such as
education and health – results obtained by Hansen & Zhang (1996) for 28 regions of China,
Wells & Thirwall (2003) for Africa showed that there is a strong negative relation between
employment growth in non-manufacturing sectors and overall productivity growth. Moreover,
emphasising the crucial role of the manufacturing sector in economic growth, Kaldor states
that industrialisation through manufactured goods endows special elements which trigger
spillovers not only in the same sector, but also in primary production and services i.e.
intersectoral spillovers. As indicated by Szirmai (2012) and Tregenna (2007), the
manufacturing sector is one of the primary sources of technological advance in the economy.
It is inside manufacturing industries that most product and process technologies are
developed. Important spillover effects in modern economies arise from manufacturing and
spread to other sectors, such as the service sector. Therefore, for instance, advances in IC

39
However, as noted by Thirlwall (2013, p. 51) , “manufacturing output growth is never likely to be constrained
by a generalised shortage of labour, because labour is a very elastic factor of production in terms of hours
worked, participation rates of males and females, and the possibility of international migration”.

33
hardware technologies produced in the manufacturing sector (silicon chips, glass fibre cables)
fuel technological change in the software producing and software using service sectors
(Szirmai, 2012).

Complementary views on the role of productivity increase in the manufacturing sector

for macroeconomic growth were further presented by Cornwall (1976, 1977), recovering
Kaldor’s ideas of manufacturing as a leading sector. In this sense, John Cornwall clearly
refers to manufacturing as a driving force for productivity improvement in a whole range of
sectors, through technological interdependence as well as input-output linkages between
sectors (Verspagen, 2000). Both Kaldor and Cornwall state in their perspectives that the
manufacturing sector embodies such special properties of a prime sector leading to economic
growth. Moreover, this link between manufacturing and technological change, as stressed by
Cornwall, rests on Neo-Schumpeterian literature which draws attention to the crucial role
played by investment in technology to the catching-up of underdeveloped countries. In this
way, Cornwall (1977), Fagerberg & Verspagen (1999) and Szirmai (2012) have emphasised
the limitations of a development strategy based on the production and trade of low value-
added products.

4. Manufacturing really does matter for the equilibrium in the balance of payments

In contrast to the neoclassical model of international trade, free trade does not generate
income increase and factor price equalisation among countries. As noticed by Prebisch in the
1950s, the neoclassical literature regarding balance of payments did not recognise that an
imbalance between the income elasticities of demand for imports and exports would be a
central explanation for the substantial constraint on economic growth for developing
countries. In dialogue with the work of Prebisch, Kaldor attached great importance to the
export performance, thus reinforcing the link which would be developed in the literature on
growth constraints imposed by the balance of payments (McCombie & Thirlwall, 1994).

Following the stylised facts proposed by Kaldor, the fourth law deduced from
Harrod’s foreign trade multiplier40 shed light on economic growth led by demand and its
limitations by the balance of payments equilibrium. Known also as Thirlwall’s law, due to
Thirlwall’s (1979) pioneering work, it places emphasis on balance of payments constraints on
40
According to Harrod (1933), the production of a country is determined by the external demand for its goods
and tends to be a multiple of such demand that is represented by the reciprocal proportion of domestic income
spent on imports.

34
growth, where the country’s long-term growth rate is approximately given by the ratio
between the rate of growth of exports and the income elasticity of demand for imports. In
other words, the sustainability of growth depends on the country’s ability to maintain the
competitiveness of its exports, which in turn depends on the capacity of the manufacturing
sector to increase productivity.

However, in spite of the recognition that manufacturing is the leading sector in

economic growth, its relevance in the basic model of balance of payments was overshadowed
since all exports and imports are aggregated together in the standard model, i.e. income
elasticities of demand for exports and imports, which ‘drive’ the model, are aggregate
elasticities. For this reason, Araujo & Lima (2007) integrating Pasinetti’s structural economic
dynamics (SED) to the balance of payments-constrained growth model created a multi-
sectoral version of Thirlwall’s Law, in which changes in the productive structure affects the
overall economic growth rate41. Moreover, this model shows that each country’s growth rate
is directly proportional to the rate of exports growth in a sectoral perspective. Therefore, this
proportionality is related inversely to the sector income-elasticity of demand for imports and
directly to the sector income-elasticity. As stressed by the authors, even if the sectoral
elasticities and the growth rate of world income are constant, it is still possible for a country
to raise its long-term growth rate by favourably changing the sectoral composition of its trade.

Following the multi-sectoral approach, recently a number of studies have been

exploring the connection between the sectoral composition of each country’s trade and the
differences in income elasticities of demand across sectors. Gouvêa & Lima (2010) dividing
primary products from manufacturing products and taking into account the technological
subdivisions of manufacturing tested this multi-sectoral model for four Latin American
countries (Argentina, Brazil, Colombia and Mexico) and four Asian countries (South Korea,
Malaysia, Philippines and Singapore) over the 1962–2006 period. The authors used the
sectoral elasticities to estimate year-by-year the evolution of the aggregate income elasticities
of exports and imports. The results showed that unlike Latin American countries (except
Mexico), Asian countries have successfully changed their composition of exports and imports
to technology-intensive manufacturing sectors in a way that led their weighted income

41
Although Pasinetti’s (1981, 1993) structural economic dynamics (SED) recognises explicitly the role of
demand- led structural change in economic growth; in his model there is no explicit balance of payments
constraint on demand.

35
elasticity of exports to grow faster than their weighted income elasticity of imports and
consequently impacted positively on the balance of payments equilibrium.

Complementarily, Gouvêa & Lima (2013), in another disaggregated study, have

estimated sectoral export and import functions in a panel of 90 countries over the period
1965-1999. Similar to the previous study, their findings showed that technology-intensive
manufacturing sectors have a higher income elasticity of demand for exports. Additionally, as
noted by Cimoli et al. (2010) in a study involving 29 developed and developing countries, the
inequality between nations was reduced in countries which sought to transform their
economic structure towards sectors with a higher income elasticity of demand for exports
relative to imports. These sectors would encompass both, as they call, ‘higher Schumpeterian
and Keynesian efficiency’, i.e. respectively products with a superior demand properties and
technical characteristics. Based on these considerations, the next section draws attention to
this dynamic, more specifically, to the role of technological progress in the economic
development.

5. Technological dynamics, innovations and economic growth

According to Schumpeter and his intellectual heirs (sometimes called the neo-
Schumpeterian school), the capitalist system is characterised by cyclical dynamics generated
by successive waves of innovations penetrating markets. His ideas, particularly with regards
to innovation, technical progress or structural change, became a contemporary subject in
economic policy and scientific debate and influenced not only a vast amount of economic
literature in mainstream economics, but also the heterodox strand. A remarkable wave of
Schumpeterian studies based on Nelson & Winter’s (1982) groundbreaking research arose to
qualify the crucial role of technology in economic growth. Building on the work of Joseph
Schumpeter (1912, 1942), they argue for an evolutionary theory of production (and economic
change), which delved inside the “black box” of the production function in order to
understand how innovation occurs and affects competition and economic growth (Mazzucato,
2013). In this approach there is no vague production function as in the so-called new growth
theory (including the early early neo-Schumpeterian models – such as in Aghion & Howitt
(1998) and Grossman & Helpman (1991), since the process of production and competition

36
involves a complex process of differentiation among firms based on their different abilities to
innovate42.

The evolutionary and Schumpeterian approach to study the complexity of innovation

has led to a policy view where the so-called national innovation systems (NIS) of a country,
as firstly defined by Freeman (1982a, 1982b), are constituted by the institutional environment
where firms of different type are embedded in as system at sectoral, regional and national
levels (Mazzucato, 2013). Over the years the NIS has been used as an analytical framework
for policy analysis in both developed and underdeveloped countries. As a result, research and
policy activities clearly focusing on systems of innovation can be found in most countries and
a rapidly growing number of studies of specific national systems of innovation, encompassing
sectoral systems of innovation, have been produced (Cassiolato & Lastres, 2008).

Many studies have provided different explanations of the concept of NIS. However,
the main definition states that NIS is shaped by collective and individual contributions of
different agents to the development and spread of new technologies. It congregates a sequence
of elements and relations that relate production, assimilation, use and diffusion of knowledge
(Lundvall, 1992). The emphasis is not on the stock of R&D, but on the circulation of
knowledge and its diffusion throughout the economy. In a micro and meso perspective, a
structured NIS provides an institutional environment to promote technological progress and
structural change.

5.1.The catching-up hypothesis: From imitation to innovation

Over the years important contributions have been made, linking the Schumpeterian
framework and policies for catching-up in an economic development perspective. Remarkable
studies such as Fagerberg (1988a, 1988b), Perez & Soete, (1988), Dosi et al. (1990),

42
This dynamic also reveals a central feature of the Schumpeterian thinking, i.e, the so-called “creative
destruction”. According with this concept the economic development is a dynamic phenomenon where new
combinations - understood as a synonym of innovation - incessantly revolutionise the economic structure. The
innovatory process involves the development of different products and processes, as well as a permanent search
for new technologies, organisational capabilities and markets providing to the innovative firm extraordinary
gains. Therefore, in the process of production and competition, the introduction of basic innovations leads to a
process of “creative destruction” in which sectors associated with the “old” technologies decline and new sectors
emerge and grow (Schumpeter, 1912). This dynamic assumes an evolutionary character which endogenously
recreates economic structures through an uninterrupted process of innovations. The capitalist dynamic involves
competitive forces in constant search for capital gains, in which even leading companies with high market power
may be aware that remaining in such a situation requires constant investment.

37
Abramovitz, (1986), Silverberg & Soete (1994), and Silverberg & Verspagen (1995) have
highlighted that the innovatory dynamics sets the pace of economic growth where the ultimate
goal is the catching-up. Thus, according to the “catching-up hypothesis”, a country’s
technological progress diffuses through the interface between innovative firms, responsible
for introducing technological innovations in the economy, and imitative ones, responsible for
the transmission of innovations throughout the economic system based on their activities of
“technological imitation” (Abramovitz, 1986).

Therefore, the innovative heterogeneity also divides countries according to the

capacity to promote technological innovations. While “leading economies” are responsible for
the frontier of the scientific knowledge, the “followers” – countries with less developed
scientific basis – can only improve their technological capabilities through the incorporation
of technological progress developed in leading countries or based on improvements of the
technological progress achieved by leading countries, which characterises “opportunity
windows” (Oliveira, Jayme Jr. & Lemos, 2003). In both cases, technological caching up
involves relatively smaller costs than those costs related to innovation for leading countries.

Moreover, stressing the importance of productivity growth for the economic

development – as in the Kaldorian theory and the structuralist approach – Perez & Soete
(1988) point out that an efficient rate of technology incorporation results in a growth rate of
labour productivity in follower countries, i.e. developing economies43. The way in which
these countries absorb and adapt technologies from leading countries will determine their
productivity growth. In this process, the cumulative causation is generated though the impact
of knowledge accumulation on productivity growth (Nelson & Winter, 2002). This dynamic
elucidates the essence of the “catching-up hypothesis”, where the gap between leading and
follower countries determines the potential of technological progress for backward economies
(Abramovitz, 1986, Fagerberg, 1988a). In this sense, the process of catching-up takes place
when a backward country is able to maintain over time a technological progress higher than
that of leading countries, due to a significant efficiency in absorbing new technologies
(Olivera, Jayme Jr. & Lemos, 2003).

Although the technological gaps offer opportunities to catch up, they are not sufficient.
The success depends on “social capabilities” which allow developing countries to obtain
“advantages from the gap”. In this way, as pointed out by Freeman (1995) and Nelson (1993),

43
This assertion elucidates the development path followed by South Korea as showed by Kim (1997).

38
such characteristics are related to the institutional framework established in a determined
structure by its National Innovation System (NIS). In this case emphasis is particularly put on
the role of historical processes, reporting differences in socio-economic capabilities for
distinct development paths and promoting systems of innovation with very particular local
features and dynamics. The innovative performance depends not only on the scientific and
educational infrastructure of a country, the magnitude of R&D, labour force capabilities,
among others, but also on how they interact between them and other variables, as well as all
other forms by which industries in a country acquire, use and diffuse knowledge.

Additionally, the Neo-Schumpeterian literature considers that NIS cannot be replaced

by foreign technology since it has a local character. That is because the development of the
NIS in a certain economy affects the degree of technological sophistication of its products and
consequently its exports. Thus, despite the recent globalisation process, NIS remains
fundamental to the development of technical progress and its diffusion in a country (Freeman,
2004; Nelson, 1996; Dosi, Fabiani & Freeman, 1994). Innovation capacity derives, thus, from
the confluence of social, political, institutional, and cultural specific factors and from the
environment in which economic agents operate in a specific country. Therefore, the chances
of a country to catch up depend on its NIS capabilities in relation to “mature countries”
(Cassiolato & Lastres, 2008; Oliveira, Jayme Jr. & Lemos, 2003)44. A movement of structural
change in favour of an intensive economic growth requires investments to build an
institutional structure that supports learning and innovation which is essential to technological
upgrading (Nelson & Winter, 1982; Fagerberg, 1994; Freeman, 1995, Abramovitz, 1986;
Albuquerque, 1999).

44
Jayme Jr. & Resende (2009) also states that the institutional structure summarised in a developed NIS
provides opportunities for productive diversification and spillovers towards the technological frontier, i.e.
technological opportunities for other sectors in the economy. On the one hand, a developed NIS and a diversified
industrial structure have the propensity to offer better conditions for gains in international trade through exports.
An economy with this structure may generally present at least three features regarding its exports: I) the conquest
of new markets will be as diversified as its exports; II) more stability of the value exported since it involves
distinct products and consequently reduce the risk of market fluctuations; III) an increase in the income
elasticity of demand for exports since that the demand for imports tends to grow in times of world economic
growth, opening more opportunities for countries which have a diversified range of exports. On the other hand, a
less developed NIS tends to present a more specified productive structure. Consequently, the range of imports
will be more diversified and the proportion of domestic market attended through the international supply higher.
This dynamic tends also to affect the income elasticity of imports. Indeed, the country where the NIS is
relatively less developed exposes a structural external vulnerability because the income elasticity of export
demand is lower than the income elasticity of import demand.

39
 5.2. Manufacturing shaping a hierarchy of national innovation systems: The case
of Machinery industries

The analysis of historical and national trajectories, taking into account the productive,
financial, social, institutional and political contexts, as well as micro, meso and macro
spheres, shows that an efficient NIS is a necessary condition to produce the learning processes
that allow the structural change towards high-tech sectors, which are centralised in the
manufacturing industry (Nelson & Pack, 1999; Freeman, 2003; Lastres, Cassiolato & Maciel,
2003). Through the convergence of the neo-Schumpeterian literature and concepts presented
in the structuralist economics Freeman & Soete (1997) and Lundvall et al. (2002) present a
chronological overview of this dynamic, analysing, among other features, the role of “leading
sectors” in the economy. They show how creative destruction has emerged from the
manufacturing sectors and promoted a movement of structural change, i.e. changes measured
ultimately by variations in the share of industrial sectors in production or employment, since
the First Industrial Revolution.

Lundvall et al. (2002) showed that the convergence of the structuralist approach and
Schumpeterian thinking was already presented in Perroux and his French followers. French
structuralism had developed an analysis of the importance of the structure of national systems
of production for economic dynamics, some of it rooted in the Marxian schemes of extended
and intensive reproduction. Like Hirschman, they assumed that different sectors affect growth
differently and that the most dynamic elements in the system (the growth poles) were located
upstream, particularly in the manufacturing sector. This led them into ordering national
systems in a hierarchy. It was assumed that countries such as the US, Germany and United
Kingdom had a stronger economy than France because their economies were more
industrialised and their production systems were particularly specialised in the production of
machine tools. Machine tools were the catalyst for the industrial revolution, which emerged in
Britain in 18th century. Since then, machine tools are the origin of almost every manufacturing
process and for this reason recognised as mother machines. Moreover, the historical
experience of the Great Powers of the industrial era – the US, the UK and Germany – gave
rise to a perspective where well-functioning machinery sectors were the driving force of a
strong manufacturing sector and a long-term innovative interaction between industrial sectors.

The relevance of the machine tools industry for economic growth were also
documented by Rosenberg (1963), taking into account a historical perspective from 1840 to

40
1910. Rosenberg showed that the US, the UK and Germany have been those nations that have
controlled, within their territories, the global production, reproduction and destruction of
machinery niches. As presented by Rynn (2010), by 1913, 82.4 % of the global production of
machine tools was centralised in these three countries which had the most dynamic
manufacturing development at that time45. The US produced 50% of the world’s machinery,
Germany 20.6% and the UK 11.6%. In 1925 the aggregated number rose to 84.3%, where the
US corresponded to 57.6%, the UK to 13.6% and Germany to 13.1%. The US could be
considered a super power in terms of machinery during this time period; the US, Germany,
and the UK were the Great Powers because they dominated the production of machinery.
During the period of World War II, the USSR overtook the UK as the third biggest producer
of machine tools. Certainly the technological global race evidenced in the years that followed,
particularly during the Cold War, highlighted the importance of a leading capital goods sector
as locus of technical change and economic hegemony.

Rosenberg (1963, p. 416) clarified the significant role played by the capital goods
industry in promoting economic dynamism, particularly regarding the introduction and
diffusion of technological change, through two key aspects46. Firstly, the aspect of “external
adaptation” states that “all innovations whether they include the introduction of a new product
or provide a cheaper way of producing an existing product—require that the capital goods
sector shall in turn produce a new product (capital good) according to certain specifications”.
In other words, the industry of capital goods needs to maintain its production at the edge of
the technological frontier in an adaptive and innovative way to develop dynamic systems of
production that set the rhythm and path of economic growth.

Secondly, the “internal adaptation” aspect refers to the internal motivation that
machinery producers improve their own techniques of production that affect the price of their
machinery output – cost reduction – and therefore is an important determinant of investment
activity throughout the economy and, consequently, it determines the rate at which
technological innovations are introduced and diffused in other manufacturing industries.
Furthermore, since cost reduction in the machinery industry is a form of capital saving for the
economy, it also raises the marginal efficiency of capital of other industries.

45
In May 1927, the League of Nations held an International Economic Conference. As part of that conference,
Dr. Karl Lange presented a memorandum that discussed the state of the machinery industries.
46
Fajnzylber (1983) also emphasises the central role played by the sector of capital goods for technological
progress, and to create an endogenous nucleus of technological dynamism in the economy.

41
 Over the years the sector of capital goods underwent significant technical
advancements enabling intra and inter-sectoral spillover effects in the economy and pushing
the technological frontier. The machine tools industry also has a significant role in the
industrialisation process transferring production know-how and technology to other
manufacturing sub-sectors which enhances industry’s capacity to develop and produce new
products and increases the productivity level and competitiveness of the country’s
manufacturing base. As a knowledge-intensive sub-sector of the manufacturing industry,
machine tools also enable the transfer of the latest advancements in information and
communication technologies or material sciences into production systems, which increases
the efficiency of the productive process and develops new materials which are used later in
new fields of application such as in railway vehicles, ship building, aerospace and automobile
industries. Since the engineering know-how in production technologies accumulated in the
machine tools industry is a competitive advantage, it also benefits first-mover firms in the
development of many other new products and processes (Saxena & Sharma, 2014; CECIMO,
2011).

These unique characteristics found in the machine tools industry have a strategic place
within the economic dynamic. Consequently, throughout the twentieth century, countries that
reduced the capacity to make manufactured goods became dependent on imported machinery
for the domestic industries and consequently lost economic dynamism. Furthermore, countries
that faced a decline in their machine tools were also those who lost the higher shares in
manufacturing. Due the global shift of machine tool towards Asian countries, the US and
Europe have clearly lost global market share in the last four decades. Among developed
countries the US underwent the most expressive decrease in the machine tools production
losing around 77 percent. While the US experienced a constant fall in the last four decades,
Japan and Germany presented an increase until 1995 reaching respectively 23.5 and 16
percent of the global machine tools production. From 1995 onwards all the most
representative producers of machine tools in Europe, i.e. Germany, the United Kingdom and
Italy, faced a constant fall in their market share, dropping respectively from 22.6% to 16%,
from 8.5% to 8% and 2.7% to 1.7%. In an opposite trend China followed an upward trajectory
becoming the world’s biggest producer of machine tools with 32% of the global production,
i.e. one third of the worldwide machine tools production is centralised in one country47.

47
These data were taken from Rynn (2010) and Gildemeister Annual report 2010.

42
Therefore, it is not by coincidence that China has showed an impressive dynamism over
recent decades showing the relevance of manufacturing to today’s world economy.

6. Concluding remarks

This chapter is a study which makes an effort to combine different theoretical strands
on development regarding the importance of the manufacturing industry to economic growth.
Through a confluence of the Keynesian-Kaldorian, Structuralist and Neo-Schumpeterian
frameworks, this chapter argued that the manufacturing industry presents some special
properties which are not found in other sectors. The theoretical discussion presented through
this heterodox triad shows how these theories incorporate different levels of economic theory,
i.e. micro (firm), meso (sector and sub-sectors) and macro (economy) and are mutually
complementary about the special role of the manufacturing industry in economic
development. Therefore, this chapter constituted the theoretical core of this dissertation.

The first section described the Anglo-Saxon Structuralism also known as Early
Structuralism and the Latin American strand. As a general characteristic among both
structuralist strands, the economic development is narrowly linked to a radical transformation
in the structure of production to suppress obstacles, bottlenecks and other rigidities of the
underdevelopment. Based on the hypothesis that the industrial structure affects both the pace
and the direction of economic development, the structuralist literature highlighted the
importance of the industrialisation as a process of structural change where the manufacturing
sector plays a central role. The structuralist strand states that without a dynamic
industrialisation, it is not feasible to increase employment, productivity and income per capita
and, consequently, to reduce poverty. The main argument stresses that the development
process involves a production reallocation from low productivity to high productivity sectors
where increasing returns to scale prevail. Inserted in this theoretical background, economic
structuralism has provided many reflections on how economic growth should be understood
in a historical perspective of mutual causation in the economic system.

The second section sought to describe the Kaldorian approach to growth, understood
as “laws” where Kaldor argued that it is not possible to understand the development process
and growth rate differences between countries without taking a sectoral approach. In a
complementary line of research to Furtado, Hirschman, Rosenstein-Rodan and Prebisch,

43
Kaldor noted that the manufacturing sector was imbued with special growth-enhancing
properties that trigger a process of cumulative causation that is not shared by other sectors.
The so-called Kaldor’s stylised facts were brought back in many contemporary studies that
validated the Kaldorian approach to today’s world economy. Moreover the section reviewed
the debate regarding the importance of the manufacturing industry for economic dynamism,
highlighting the debate of three sources of de-industrialisation, i.e. the so-called statistical
illusions, servicification and productive fragmentation. In this sense, this section showed that
a non-negligible part of the extent of de-industrialisation reflects changes in statistical
classification rather than a real decreasing relative importance of manufacturing.

Additionally, it is also emphasised that despite the considerable rise of services in

recent years, manufacturing is still the main engine of growth because many services
essentially are spin-offs from manufacturing production. Furthermore, the increasing trend of
productive fragmentation over the world has not only decreased the manufacturing share in
many economies but also negatively affected these countries through a deterioration of the so-
called “industrial commons” that caused a loss of high-value added activities, such as R&D
and design, as well as the capacity to generate technological innovation. The Kaldorian
section still explores the role of manufacturing for productivity growth stating that the rate of
productivity growth in the manufacturing sector depends on its rate of output growth due to
the operation, within this sector, of static and, above all, dynamic returns to scale. This section
ends showing the role of manufacturing for the equilibrium of balance of payments and
highlighting the importance of manufacturing technology-intensive sectors in this dynamic.

The third section investigates the Neo-Schumpeterian route to development, exploring

relations between innovation, economic dynamic and catching-up in a sectoral specific
approach. From this discussion, it is emphasised that the way in which a developing country
absorbs and adapts new technologies from leading countries will determine its economic
dynamism. However, the technological catching-up depends on its National Innovation
System. Consequently, the more developed the country’s NIS is, the larger the range of goods
produced in the global technological frontier will be. This process stimulates the performance
of exports value and mitigates the value of imports, generating a pattern of exports which can
overcome the structural external vulnerability and the underdevelopment. Additionally,
through the convergence of the neo-Schumpeterian literature and concepts presented in the
structuralist economics, studies showed the role of the manufacturing sector as a locus of
technological change since the First Industrial Revolution. From this perspective arose the

44
role of the machine tools industry in economic growth, particularly regarding the introduction
and diffusion of technological change to the economy.

45
Appendix 1.1 - The Kaldor-Dixon-Thirlwall cumulative causation model

In contrast to the neoclassical growth theory, which emphasises the importance of

mechanisms that generate convergence, Kaldor (1966, 1970) highlighted the weight of
cumulative mechanisms that slow down convergence or make economies diverge. A few
years after Kaldor had developed broad ideas about the mechanisms behind the cumulative
causation model, Dixon & Thirlwall (1975)48 presented the first formalisation of the standard
cumulative causation model through four equations in order to clarify its structure. The basic
model is very recognised and represents that the rate of growth of output is a function of the
growth of exports. It is given by:

y = γx (1.1)

Where (y )denotes the rate of growth of output in time(t). The term (x ) is the rate of growth
of exports and (γ) is the elasticity of output growth with respect to exports growth.

This equation underlines the role of external demand in an open economy. In this
sense, a diversified industry in terms of manufacturing goods, which are highly demanded
internationally, tends to increase the country’s output growth.

The equation of export demand can be defined as:

X = P  P (Z  ) (1.2)

Where (X  ) is the quantity of exports, ( ) is the price of domestic goods, ( ) is the price
of foreign produced goods, () is the price elasticity of demand for exports, () is the cross
elasticity of demand for exports, (Z ) is the level of the world income and () is the income
elasticity of demand for exports.

This, for discrete changes, gives the approximation:

 =  +  +  (1.3)

Assuming that η = δ, as proposed by McCombie & Thirlwall (1994), the growth rate
of exports, in turn, can be expressed as:

 =   −  " +  (1.4)

48
See also Dixon & Thirlwall (1979).

46
Where lower case letters represent variables in terms of rates of growth.

While  and  are determined exogenously, the rate of change of domestic prices is
given by the mark-up pricing equation, taking into consideration that firms set their prices. It
can be specified as:

 = # − $ + % (1.5)

Where w , r and τ respectively represent the rate of change of money wages, the rate of
growth of average labour productivity and the mark-up on unit labour costs49. Finally, the
linchpin of the system to complete the model stresses that the rate of labour productivity is
partly dependent on the growth of output itself, i.e. “the Kaldor-Verdoorn law”. Thus, the
following equation (6) is Verdoorn’s law, which connects the rate of labour productivity
growth and that of output growth. It has been specified as:

$ = $) + *(+, ) (1.6)

Where ($) ) represents the rate of autonomous productivity growth and ( *) the Verdoorn
coefficient. This equation exposes the fundamental mechanism for cumulative causation in
the model. This equation allows the possibility of cumulative growth. The growth of output
improves the productivity of labour, reducing the price of domestic products (8), leading to
new rounds of export growth (7) and more output growth (4). Combining equations (4), (7),
(8) and (9), the expression of the equilibrium growth rate of output is given by:

-./(01 231 4 51 2671 "4891 ]

+ = ;4-/<
(1.7)

The equation (10) shows that, since  < 0, the rate of growth of output varies positively as a
function of the autonomous rate of productivity growth ($) ), the growth of foreign
prices ( ), the growth of world income ( ) and the income elasticity for exports (). In
contrast, output growth varies negatively with the rates of growth of money wages (# )and
mark-ups on unit labour costs (%). In this view, the Verdoorn coefficient ( *) constitutes the
heart of this model. Moreover, it gives an important contribution to the understanding of why
economic growth differs between countries. Therefore, manufacturing production has a
central role in determining the productivity of the economy.

49
It is important to emphasise that the Dixon and Thirlwall model tries to interpret a developed country – such as
Britain – where wages and labour productivity are assumed to be high.

47
CHAPTER 2 – FALLING BEHIND AND MOVING
AHEAD: THE BRAZILIAN AND SOUTH
KOREAN PROCESS OF INDUSTRIALISATION

“Don't listen to ‘comparative advantage’ advice. Whenever we wanted to

do anything the advocates of comparative advantage said, ‘We don't have
comparative advantage.’ In fact, we did everything we wanted, but whatever we
did, we did well.”

Sung Sang Park, Governor of the Bank of Korea from 1986 to 198850.

“The best industrial policy you can have is not have one.”
Pedro Malan, Brazil's Finance Minister from 1995 to 200351.

Introduction

Comparisons of the Latin American and East Asian industrialisation experiences are very
commonplace. At a first glance, until the 1980s, Brazil and South Korea presented similarities
in terms of political and economic development. During the 1960s and 1970s, both countries
were governed by dictatorial regimes and had their development path based on the so-called
national development plans which sought to strengthen the heavy industrial base –
metallurgical, chemical and metal-mechanic industries. Under the developmental agenda,
although through different strategies, the results achieved by a state-led industrialisation were
not negligible, showing a strong economic and industrial growth with both economic blocks
increasing their share of manufactured products in the global value added. From 1950 to
1980, the Brazilian share increased from 1.65% to 3.12% and in South Korea this number
rose from 0.33% to 0.76%52.

50
Quoted in Robert Wade, “East Asia’s Economic Success, Conflicting Perspectives, Partial Insights, Shaky
Evidence. World Politics, Volume 44, Issue 2 (Jan., 1992), 270-320.
51
Quoted in David Trubek, “Toward a new law and development”, The World Bank Legal Review, volume 4,
2012, 287.
52
Source: GGDC 10-Sector Database.

48
 However, in spite of the vigorous growth, in both economies, after the 1980s some
structural problems came up, reversing this similar economic dynamism. In both countries
exogenous shocks like oil crises in 1973 and 1979, the Volcker interest rates shock in 1979
and the debt crisis in 1982 triggered a severe imbalance in the public finances and a deep
current account deficit. On the one hand, Brazil as well as the rest of Latin America was badly
affected by these shocks due to its large current account deficits and its large stock of external
debt. This vulnerability, associated with an abrupt cut-off in bank financing, plunged the
country into a serious crisis and caused a deep disarticulation of the state to provide active
policies for the national industry. Consequently, economic growth was seriously retarded
giving rise to the commonly used term “lost decade”. On the other hand, Korea dealt with the
crisis in a much less traumatic way since its strategy of industrialisation was relatively less
dependent on foreign debt and the Japanese banks “recycled” the country’s liabilities.
Therefore, the Korean state and corporations consolidated and expanded the domestic
industrial base through an industrial partnership with Japan.

In contrast to Korea, the Brazilian economic weakness caused by exogenous and

endogenous shocks gave rise to the neoliberal view that spread through the perception that
severe macroeconomic imbalances and the institutional framework of a model based on the
Import Substitution Industrialisation (ISI) did not promote the development of a competitive
industrial base as observed in international markets. The adoption of the neoliberal agenda
supported by the World Bank and the International Monetary Fund (IMF) exposed Latin
American economies to a doubtful development strategy based on the free market doctrine to
conduct the development path53. In a different development strategy, Korea, as well as other
Asian countries, did not embrace the neoliberal agenda in such “fundamentalist” way. In fact,
the state directed and coordinated investment decisions to expand the productive structure and
achieve levels of technology compatible with competing in external markets. While the
Korean strategy of development based on “getting prices wrong” resulted in a trade and
productive dynamism never observed before in the emerging world54, Brazil and Latin
America as a whole followed a perverse strategy through neoliberal reforms that subjected the
country to low rates of economic growth followed by cycles of stop-and-go over the last three
decades. This debate has reopened different interpretations of the development model adopted

53
See, for instance, Frenkel et al. (1992).
54
See Amsden (1989).

49
by Brazil and South Korea as well as endogenous and exogenous dimensions that shaped their
industrial structures.

Focusing on by factors which include the sectoral composition of countries and the
role of the state in guiding the development path, the main contribution of this chapter is an
identification of factors that determined different routes of economic dynamism followed by
Brazil and South Korea from the 1980s onwards. Thus the chapter is divided in eight sections,
apart from this introduction. The first section elucidates briefly the whole of the
developmental state for economic growth. The second section describes the developmental
era in both countries in a comparative perspective detailing the development plans
implemented from 1950s to the late 1970s. The third section makes an attempt to balance the
role of public and private sectors in the formation of these late industrialised countries. Thus,
seeking to answer why the decoupling in growth rates was much more drastic in Latin
American countries vis-à-vis Asian economies, the fourth section explains the genesis of the
debt crisis. The fifth section analysis the reason behind the technological decoupling between
these two economies and the trajectory of technological learning undertaken by South Korea.
The rise of the neoliberal era is described in the sixth section to give the necessary support to
understand the economic policy change after the 1980s. Therefore, the outcomes in terms of
industrialisation and sustained economic growth in a long term perspective are approached in
the seventh section highlighting the dichotomy between Brazil and South Korea in terms of
labour productivity, catching-up and exports. The last section presents some concluding
remarks.

1. The Developmental State promoting a Big Push in economic growth

In the economic literature on development, there are a vast range of studies stressing
the transformative power of industrialisation in the economic system. Classical Structuralist-
Kaldorian contributions by Rosenstein-Rodan (1943), Nurkse (1953), Lewis (1954),
Hirschman (1958), Myrdal (1957), Kaldor (1966, 1967) and Chenery (1960, 1979) pointed
out that the study of long-term economic growth involves the role of industrialisation as
engine of economic growth. These “sector- specific” approaches emphasised the special
properties of manufacturing (such as highest potential of productivity, spillover effects, strong
forward and backward linkages, as well as technological and pecuniary externalities) and the
way in which these properties spread to the economy as a whole stimulating the economic

50
growth. However, the economic literature has also stressed that successful cases of
industrialisation were not a spontaneous process. Therefore, the developmental literature has
discussed the active role played by the state in the industrialisation process of many latecomer
economies55.

Although the concept of developmental state has been applied in a number of contexts,
contemporaneously, the state’s commitment to developmentalism has been expressed through
common policies and institutions designed to achieve national economic development. In this
sense, Chang (1999, p.183-192) defines a developmental state as “a state which can create and
regulate the economic and political relationships which support sustained industrialization (...)
This state takes the goals of long-term growth and structural change seriously, ‘politically’
manages the economy to ease the conflicts inevitable during the process of such change (but
with a firm eye on the long-term goals), and engages in institutional adaptation and innovation
to achieve these goals.”

Additionally, a central element in this context has been the 'pro-investment'

macroeconomic environment for industrial growth. As a general characteristic, the
instruments implemented by the developmental agenda comprised controls over interest rates
and credit allocation, including particularly long-term financing with low interest rates. In
addition to financing, a set of tax and tariff measures such as customs tariff protection, tax
exemptions, and special depreciation allowances were applied to provide protection to the
nascent industry. Moreover, measures to boost enterprises’ profit margins and therefore the
potential investible resources available to corporations were widely implemented. In the case
of South Korea, for instance, it is also possible to mention policies for technology acquisition,
incentives to activities of R&D and the promotion of cartels for specific purposes such as
standardisation, specialisation and exports (Akyuz et al., 1998).

As pointed out by Coutinho (1999), the sequence of sectoral priorities, as suggested by

Hirschman, and companies to support, in order to carry forward the development agenda,
were part of development process that involved financial institutions (e.g. development banks)
and planning (e.g. committees, ministries, departments). In this process, the state’s initiative
and leadership was a general rule varying from case to case according to the interaction and
55
Even the World Bank at the height of the neoliberal revolution acknowledged that in most Asian economies,
especially those in East Asia, the State did more than get the “prices right”. Actually, the intervention was
carried out basically with three policy instruments: (i) the promotion of specific industries, (ii) directed-
investment programmes and (iii) the promotion of exports. For details, see ‘The East Asian Miracle' published in
1993 by the World Bank.

51
coordination with the private sector. In short, elements of planning, selection, intervention and
dirigisme were present in all successful cases of rapid industrialisation. The creation of
competent bureaucracies of the state, organised under meritocratic criteria to operate the most
important institutions, constituted an indispensable condition for coordination of industrial
policies.

Furthermore, the implementation of development plans, especially in the most critical

stages of heavy industrialisation, demanded a clear concentration of political power in the
executive authority that was structured around a certain core of the bureaucratic system able
to organise and give coherence to the multiple instruments used for the implementation of
sectoral programmes. In this sense, both in Brazil and South Korea fit paradigmatically in the
pattern described above, evidently under peculiar political and historical contexts. The most
important divergences from the pattern of industrial development in these economies refer to
the interrelation between the state and private spheres as well as distinct forms of international
trade insertion and investment pattern (Cheng et al., 1998). Therefore, while in South Korea
the domestic private capital, represented by the Chaebols, played a decisive role in the
evolution and development of Korean industry, in Brazil economic growth and investments
were driven by state-owned enterprises and multinational companies established in the
country during the 1950s and 1960s. Both dynamics are analysed in the next two sub-
sections.

2. Mapping the Golden Age of the Industrialist Era

2.1. Brazil: 1930-1980

For the five decades between 1930 and 1980, the concern with the development of the
productive structure remained a central element in the analysis of economic growth. The
period of transition from a primary export model to a strategy of Import Substitution
Industrialisation (ISI) was widely discussed by theorists such as Furtado (1959), Tavares
(1972) and Cardoso de Mello (1982), Suzigan (1988), Serra (1982), Malan & Bonelli (1983).
In Brazil, between the 1930s and the end of the Second National Development Plan (PND II)
in 1979, this theoretical framework was extremely important for the development strategy
adopted through industrial diversification supported by the state. From 1930 until 1980, the
Brazilian economy had a remarkable growth record achieving one of the highest rates of

52
growth in the world. During this period gross domestic product grew at an annual average rate
of 6.5%, while real GDP per capita increased at an annual rate of 3.8%. In this context
throughout this fifty-year period, the growth strategy adopted was based on ISI policy. Under
the ISI regime, Brazil based its strategy of development in a strongly state-led orientation
which was theoretically formalised by the Latin American structuralist view which
incorporated relevant arguments of the infant industry56, and the Keynesian economics
approach.

As presented by Furtado (1959) and Tavares (1972), the Brazilian development until
1930 could be defined as an outward orientation, via a primary-export model, whose demand
was determined exogenously. In this dynamic, the agricultural-export sector was responsible
for the growth of the product, the formation of national income and the capacity to import.
With the primary sector crisis in the late 1920s, together with the effects of the Great
Depression in the 1930s, there was a strong decrease in export revenues – due to the decrease
of exports earnings as a result of the plunge of coffee prices – and, consequently, a sharp drop
in the import capacity that represented a reduction of 50% in the quantum of imports. This
phenomenon forced the country to develop new productive activities, with the support of the
domestic demand previously met by imports (Tavares, 1972). Therefore, a new dynamic
started to be shaped with domestic demand playing a central role in the strategy of
development57. In contrast to the previous period, the external sector lost relative importance
in the composition of national income whereas the participation and dynamism of the
domestic market increased in the economy. Moreover, the external sector which was
previously a source of demand started to contribute to the expansion and diversification of the
productive structure supplying capital goods and intermediate goods.

Based on this structural dynamic, during Vargas’ government (1951- 54), the
apparatus that enabled the process of a state-led industrialisation started to be consolidated
and inspired by the opportunities created by the reconfigured international division of labour
in the post-war period. This view was closely in line with the framework of historical analysis
formulated by the new generation of economists coalesced at the Economic Commission for
Latin America (ECLA), created by the Economic and Social Council of the United Nations in

56
See Hamilton (1790) and List (1841).
57
Furtado (1959) denominated this phenomenon as a ‘displacement of the dynamic centre’.

53
194858 (Ioris & Ioris, 2013). The perspective in vogue was that the Brazilian economy had to
establish the basis for a more dynamic process of industrialisation. During Vargas’s
government, industry advanced though import substitution of consumer goods by national
private capital and some intermediate goods through state-owned enterprises (SOE).

Inspired by the ECLAC developmental thinking, in 1951, the Industrial Development

Commission (CDI) was created to formulate a plan of industrialisation alongside with specific
projects to create and expand priority sectors such as energy, metallurgy, mineral processing,
chemical, textile, rubber and building materials. Its creation was a strengthening of the
government's role as an economic agent and favoured the collaboration between industries
and the state59. Moreover, in 1952, the Brazilian government concerned with the country’s
infrastructure led to the creation of the Brazilian National Bank of Economic Development
(BNDE)60. The BNDE started out as a vehicle to provide long-term financing for the renewal
of large infrastructure projects such as energy, steel and transportation, as well as providing
resources for heavy industrialisation. At the same time, many state-owned enterprises were
created to support and propel the ongoing process of industrialisation particularly in areas less
appealing to the private sector. A remarkable example is Petrobras, which was established in
1953 to undertake activities of exploration, production, refining of oil and its derivatives.

During the 1950s, credit policies, multiple exchange rates and preferential tariffs for
imports for specific sectors were extensively adopted to boost the industrial growth. In this
way, an important step taken by the Brazilian government was the resolution (Instruction 70)
issued by the Superintendence of Currency and Credit (SUMOC) in 1953, which ranked
imports according to their essentiality61. This normative instruction established five exchange
rate categories, depending on the item being imported. Imports considered of national interest,
particularly fuel, wheat and print paper, were subsidised while other items were strongly
discouraged by over-taxation. Furthermore, the government offered a variety of incentives to
attract Foreign Direct Investment (FDI), such as tax cuts and subsidies. Thus, in 1955, the

58
The commission is nowadays described as the Economic Commission for Latin America and the Caribbean
(ECLAC), with its headquarters in Santiago, Chile.
59
The CDI was abolished in 1954 with the end of the Vargas government, but reborn in 1956 during Juscelino
Kubitschek’s government with the name of the Development Council.
60
The acronym was later changed to BNDES when “social development” was added to the name in 1982.
61
This instruction represented a stimulus for at least three reasons. First, by rising domestic prices of specific
imports it consolidated protection for producers of industrial goods. Second, it provided concession of exchange
subsidies for capital goods and basic inputs required for the process of industrial development. Third, with the
additional revenues obtained by the auctioning of foreign currency to importers of ‘non-priority’ goods, it raised
public funding for government investments in infrastructure (Studart, 1995).

54
Superintendence of Currency and Credit (SUMOC) issued a resolution (Instruction 113)
which allowed imports of capital goods without exchange cover to be counted as direct
foreign investment62. In practice, under a regime of multiple exchange rates, these resolutions
provided a subsidy for imports operations and together they served as the two most important
tools of commercial policy to support the process of import substitution.

In 1956, the recently elected President Juscelino Kubitschek launched the Targets Plan
(Plano de Metas – 1956-60) to accelerate the process of industrialisation. The slogan in vogue
at the time was that Brazil would advance “fifty years in five”63. The main aim was to achieve
the heavy industrialisation with ISI deepening in sectors such as automotive, shipbuilding,
heavy electrical equipment and machinery, allowing a significant expansion of the capital
goods sector. With regard to basic industries, the plan particularly addressed industries such
as steel, non-ferrous materials, heavy chemicals, petroleum, pulp and paper. As result of
developmental policies the investment-to-GDP ratio grew from an average of 14.91% in
1951-55 to 16.04% in 1956-60. In this investment promotion agenda, a central role was
played by the state, whose participation in national investment increased from 23.80% in
1951–55 to 29.78% in the period 1956–60. During the Targets Plan, the Gross Domestic
Product (GDP) grew in average 7.7% and the manufacturing industry rose 11.37% per
annum64.

In the early 1960s, however, the Brazilian economy showed serious economic
imbalances alongside with an unstable political period. In the economic side, the constant
increase in public expenditure during the 1950s led to a severe budget deficit, which was
largely financed by the expansion of the monetary base, causing a sharp increase in inflation
rates (Britto, 2008). On the political side, President Quadros’s resignation in 1961 initiated a
serious political crisis that culminated in a military coup in 1964. Due to this critical scenario,
both the GDP growth and investment rate presented a volatile trend. These variables
combined with a significantly lower import coefficient, led many economists to diagnose the
end of ISI growth model.

With the rise of the military regime, the Government Economic Action Plan (PAEG)
was launched to give emphasis to short-term policies to combat inflation, inherited from the
62
Since the Brazilian Central Bank was created in 1964, prior to this year, the responsibilities of a central bank
were shared by the National Treasury, the Brazilian Bank and the Superintendence of Currency and Credit
(SUMOC), which was later expanded to become the Central Bank.
63
See Lessa (1983) and Lafer (2002).
64
See Appendix 2.1.

55
Target Plan, and make structural reforms. Thus, from 1964 to 1966, under the mandate of
General Castello Branco, the military government undertook a tight fiscal and monetary
policy to tackle increasing inflation and public deficit. The military government conducted
structural reforms which encompassed three pillars, i.e. finance, taxation and external sector65.
Therefore, the financial reform implemented by the PAEG sought to create mechanisms of
long-term financing to avoid the inflationary financing of the public sector and allow the
private sector to recover the industrial investment. The financial reform led to the creation of
the Central Bank, and the establishment of a series of indexation mechanisms to protect long-
term contracts from inflation. Moreover, the BNDES changed its focus from lending to public
projects to financing private companies66. In terms of tax reform, the state implemented the
value added tax (VAT) that increased govern revenues as percentage of GDP from 17% in
1964 to 20.9% in 196667. Despite the success of the Plan, particularly regarding stabilisation
and reforms, the country was still trapped in a cycle of mediocre economic growth.

Therefore, in 1967, under the administration of the General Costa e Silva, there was a
radical change in the economic policy. The Minister of Finance, Antonio Delfim Neto,
promoted expansionary monetary and fiscal policies that were successful in inducing
economic recovery. These policies, associated with the return of external liquidity, restored
the mechanisms of the state to provide funding for public and private agents. From 1968,
reaching an investment rate of 18.7% of GDP, ISI returned to deepen and to promote the
country’s economic growth, under a strong developmental thinking. The First National
Development Plan (I PND: 1972-1974) was launched to promote investments concentrated
especially in the durable goods sector and infrastructure. Consequently, the economy
experienced unprecedented economic growth, fuelled by private consumption and public
investment in infrastructure.

This period corresponds to the so-called “Brazilian miracle” (1968-1974) in which the
country grew an average of 11.7% per annum. During the peak of the economic miracle, from
1970 to 1973, the capital goods industry grew at an annual rate of 22.5%. In this interregnum,

65
For more details about reforms implemented during the 1960s, see Simonsen (1970) and Lara-Resende (1989).
66
In 1965, in order to support the development of the machinery and equipment industry, the first subsidiary of
BNDES, namely Government Agency for Machinery and Equipment Financing (FINAME) was created. In the
ISI model, the domestic machinery industry was seen as a strategic sector to overcome the dependency on
foreign imports of capital goods and therefore foster industrial growth. FINAME had the objective of providing
medium- and long-term funding for the development of the machinery and equipment industry. It comprised
financing purchase and sales operations and exports of Brazilian machinery and equipment, as well as imports of
goods of the same nature not produced domestically.
67
Source: Brazilian Development Bank (BNDES). For historical series, see BNDES (2001).

56
the durable consumer goods industry maintained growth rates of 25.5% per annum. However,
given the macroeconomic instability caused by the first oil shock in 1973, the "economic
miracle" ended imposing a new challenge to the government that took office in 1974.
Therefore, the development of a new developmental project faced a scenario in which the
Brazilian industrial structure was not fully diversified and the country was still very
dependent on import of capital goods and basic inputs, such as oil.

Under the developmental aegis, the Second National Development Plan (PND II:
1975-1979) emerged as an attempt to maintain the economic growth even in face of a strong
international instability. This last economic programme of the Brazilian developmental era
was announced in order to advance the industrialisation model in an environment marked by
the absence of a consolidated and coordinated long-term funding system. The new programme
guidelines focused on continuing ISI, particularly accelerating it in intermediate and capital
goods, and correct productive weaknesses. The industrial policy also sought to propel sectors
related to basic inputs, food and energy as a way to overcome the external constraint.
Thereby, the measures adopted sought to generate profound changes in the productive
structure (Castro & Souza, 2004; Castro, 2001).

The main pillars of the Brazilian industrial policy were the combination of trade
protectionism, promotion of investments by state-owned enterprises and financing granted by
the BNDES. These measures sought to coordinate the expansion of the domestic supply and
maintain high rates of investment in a gradually adverse scenario (Carneiro, 2002). After the
first oil shock in 1973, the refusal of II PND to adopt a strategy more cautious in face of the
dramatic external environment led to a gradual replacement of the private investment by the
public sector to maintain the high rate of economic growth68. Therefore, public investments
increased from 23.5% in 1974 to 28.5% of the total investment in 1979, while private
investments for the same period fell from 60% to 55% (Serra, 1982; Coutinho & Reichstul,
1982). Although distant from the technological frontier in many industries, ISI
regime resulted in a diversified industrial structure, deeply regulated, integrated and driven by
the domestic market which resulted in a profound change in the productive structure and a
progressive de-linking from the international economy with increasing shares in the global
GDP.

68
See also Lessa (1977).

57
Table 2.1 - Brazilian Economic Development Plans and Economic Indicators
Sectoral Composition
Economic GDP Export/ Import/ GFCF/ Primary
Manuf. Services
Development Plan Growth GDP GDP GDP Sector
Targets Plan
(1956-1960) 8.12 5.86 6.22 16.04 11.62 20.42 62.25

PAEG
(1964-1968) 5.30 6.46 5.86 16.10 10.32 21.07 62.41

I PND
(1972-1974) 11.35 7.60 10.40 20.85 7.24 22.80 63.37

II PND
6.42 7.06 9.10 22.54 6.20 21.67 64.58
(1975-1979)
Source: Elaborated by the author, using data from The Brazilian Central Bank, Brazilian Institute of Geography
and Statistics (IBGE), Ipeadata, World Bank – World Development Indicators, The Conference Board and
Groningen Growth and Development Centre and UNCTAD.
Note: GDP = Gross Domestic Product, GFCF = Gross Fixed Capital Formation and Manuf. = Manufacturing.
PAEG = Government Economic Action Plan, First National Development Plan = I PND and Second
National Development Plan II PND.

2.2. South Korea: 1945-1980

As widely documented in the literature, South Korea is one of the most successful
trajectories of economic development in the second half of the twentieth century (Jones &
Sakong, 1980; Bénabou, 1982; Pack & Westphal, 1986; Amsden, 1989, 2001; Canuto &
Ferreira, 1989; Chang, 1994, 1998, 2002; Canuto, 1994 and Rodrik, 1994). However, the
immediate post war period was characterised by extreme economic disorganisation and
stagnation due to geopolitical turbulences. In 1945, Korea emerged from World War II under
US occupation after a long period of Japanese domination. Shortly after the proclamation of
the Republic in 1948, the country was the scene of a new war with North Korea that lasted
from 1950 to 1953. During this period, South Korea struggled to take the first steps in a
process of industrial development. At that time, the manufacturing sector was very incipient
and the weak industrial bourgeoisie was entirely dependent on the state.

After the Korean Armistice in 1953, South Korea moved from a majority agrarian
economy to an industrialised country, showing levels of income per capita and well-being
much higher than any other developing nation. In the 1950s, under President Syngman Rhee’s
government, South Korea adopted ISI model and focused in the following measures: (i)
support for the non-durable consumer goods sector, particularly with low capital intensity,

58
though favoured credits and import licenses; (ii) the creation of national capitalist groups
through subsidised privatisations of several companies, including banks; (iii) the
implementation of a broad agrarian reform in order to reduce social tensions and create a new
social base to support the dictatorial regime; (iv) efforts to reduce illiteracy and promote basic
education. Over this decade, Rhee’s dictatorial regime led this process based on the support
given by the US and economic groups that were favoured during the period (Coutinho, 1999).

Additionally, as explained by Frank et al. (1975), from 1953 to 1960, most of South
Korea's imports were financed by foreign aid grants from the United Nations Korea
Reconstruction Agency (UNKRA) and the United States bilateral assistance programme. Both
sources of foreign aid were used to import food and essential industrial raw materials as well
as capital goods. Between 1954 and 1960, foreign assistance, excluding donations by foreign
voluntary organisations, financed more than 70 percent of total imports. Furthermore, about
74 percent of South Korean investment was financed by foreign aid from 1953 to 1960. From
1953 to 1960, GDP grew on average 4.37% per annum and the only bad year was 1956 with
0.48% of GDP growth. However, this rapid economic growth also resulted in a surge of
inflation. The consumer price index (CPI) increased on average 40.8% per annum between
1953 and 1957, reaching its highest point in 1956 when the inflation rate peaked to 68%69.
Therefore, concerned about inflation, both the South Korean Government and the Office of
the Economic Coordinator (OEC)70 established a plan to control the rampant inflation rate.
After the implementation of the programme in 1957, the annual rate of domestic inflation
declined sharply.

In 1960, the lack of political support and a wave of popular uprisings led by students
culminated in the resignation of President Rhee. After a year of political instability, in May
1961, a military coup overthrew the Chang Myon government that had come to power
following the student revolution. Park Chung Hee, presided over South Korea from 1961 until
his assassination in 1979. During this period, General Park’s administration resulted in an
impressive economic growth in parallel to a dynamic process of industrialisation through
successive five-year plans. The remarkable characteristic throughout this process was a
coordinated state planning associated with a sequence of clear objectives that resulted in a
substantial increase in the economy's investment rate. The investment rate increased from

69
Source: OECD statistics database. The national reference year for the CPI index is 2010.
70
The US government established the Office of the Economic Coordinator under the UN Command in Korea to
coordinate between the US and Korean governments and other aid agencies (Kim & Kim, 2014).

59
11.70% in 1961 to 39.00% in 1979, and the GDP grew in average almost uninterruptedly
10.01% per annum71.Additionally, this period was marked by a considerable shift towards the
adoption of the strategy of export promotion.

In the beginning of the 1960s, the South Korean industrial structure was still incipient
and poorly diversified. Moreover, it was based essentially on the production of non-durable
products. In this way, to overcome the problem of foreign exchange shortage and US
dependence, General Park launched a programme of investments (first five-year plan, from
1962 to 1967) to expand the manufacturing industry with strong incentives for exports. The
strategy was based on taking advantage of the alliance with the US government to penetrate
the large North American market. The textile and clothing sector led this first export effort,
complemented by other light industries. Therefore, a series of policy reform programs were
introduced. One of the first acts of the government of Park Chung Hee was to nationalise all
commercial banks and establish many state-owned specialised banks72.With this control over
finance, the government could closely manage and monitor progress of industrial investment,
all development projects, and export performance (Amsden, 1989). Moreover, seeking to
improve competitiveness, the currency was devalued and a unitary floating exchange rate was
adopted (Seong, 2001). Tax subsidies were also widely implemented (indirect taxes
exemption, income tax reductions, tax premiums linked to performance targets, drawback,
etc), estimating its volume as equivalent to 10% of exports in the period (Coutinho, 1999).

The second five-year plan (1967-1971) came to reinforce the Government’s

commitment with the export-oriented growth strategy. Exports as share of GDP increased
continuously from 11.02% in 1967 to 14.16% in 197173. It is important to highlight that,
although export promotion was the main aim of South Korean development policy,
investments in infrastructure and the domestic market also received attention during the
period. As the industrialisation process progressed, the need for a structured heavy industrial
base to provide intermediate inputs domestically became increasingly evident. For instance,
the Korean government created the national steel company, POSCO, in order to foster the
development of a national industry. Therefore the industrial policy advanced to giving the
necessary support for the development of industries such as chemicals, machinery, and
iron/steel. Thus, as explained by Amsden (1989), a central characteristic of the South Korean
71
Source: Bank of Korea, Economic Statistics System.
72
Although pressures to liberalise in the 1980 led the government to privatise commercial banks, the South
Korean government maintained its control over commercial banking system (Amsden, 1989).
73
Source: Bank of Korea, Economic Statistics System.

60
development relied on the combination of strategies to promote exports and the ongoing
process of ISI.

At the beginning of the 1970s, President Park announced the third five-year plan
(1972-1976) for the transition to a heavy and chemical industrialisation. Industries receiving
particular attention included iron/steel, nonferrous metals and petrochemicals. Additionally,
this plan sought to provide the basis for shipbuilding, automobiles, machinery and equipments
industries. The programme was supported by a broad range of policy instruments. In 1973, the
South Korean government established the National Investment Fund (NIF) to supply funds to
support the heavy and chemical industrialisation, as well as to help increase exports, through
low-cost financing. Furthermore, since Korean entrepreneurs' capacities to obtain foreign
capital was very limited due to the low creditworthiness of domestic firms, the government
began to guarantee the reimbursement of all foreign loans, regardless of whether they were
initiated by public or private companies (Seong, 2001). Additionally, to support the export-
oriented growth further, export incentives were expanded. In this way, tax incentives were
deepened and the Export-Import Bank of Korea (Korea Eximbank) was created in 1976 to
finance export operations with favourable interest rates (Coutinho, 1999). Along with these
policies, the government overhauled the education and training systems to promote and secure
engineers and skilled workers into the heavy and chemical industrialisation74.

The fourth five-year (1977-1981) plan strengthened and fostered the development of
industries designed to compete effectively in the world’s industrial export markets. Therefore,
based on efforts undertaken in the previous plan, these major strategic industries comprised
technology and skilled labour-intensive industries such as shipbuilding, automobiles,
electronics, machinery and equipment. In the late 1970s, the measures adopted in the third and
fourth five-year plans resulted in a rapid change in the industrial structure. Under the
developmental perspective, South Korea achieved a substantial increase of the manufacturing
share in the GDP. From 13.17% in 1950 the manufacturing share in the GDP rose to 27.62%
in 1979, particularly due to the massive efforts of the developmental agenda to consolidate the
industrial base comprising light, heavy and chemical industries. In an intra-industrial analysis
both the productive and export structure shifted gradually from labour-intensive sectors to

74
Aiming to support the heavy and chemical industrialisation, in terms of education policy, “the government also
introduced a skills licensing system to encourage every Korean worker to possess at least one skill. In addition,
for each field of engineering the government actively recruited outstanding Korean scientists abroad and
established a modern laboratory where research on the improvement of production technologies was encouraged
in collaboration with industry researchers and university professors” (Seong, 2001, p. 10).

61
capital-intensive ones by the end 1970s. Although costly to establish, from infant industries
these capital-intensive sectors became progressively more competitive in the international
market.

Table 2.2 - South Korean Economic Development Plans and Economic Indicators
Sectoral Composition
Economic GDP Export/ Import/ GFCF/ Primary
Manuf. Services
Development Plan Growth GDP GDP GDP Sector
1st Five-Year Plan
(1962-1966) 8.34 6.70 16.32 16.04 41.68 15.25 40.41

2nd Five-Year Plan

(1967-1971) 11.46 12.65 23.81 25.34 30.45 19.25 46.15

3rd Five-Year Plan

(1972-1976) 10.50 24.81 30.40 27.16 27.43 24.24 44.23

4th Five-Year Plan

7.44 28.53 33.23 34.02 21.52 27.44 44.05
(1977-1981)
Source: Elaborated by the author, using data from The Bank of Korea (BOK) and The World Bank – World
Development Indicators.
Note: GDP = Gross Domestic Product, GFCF = Gross Fixed Capital Formation and Manuf. = Manufacturing.

3. The role of the State and private sector in the industrialisation process

3.1. Creating National Champions: The South Korean Chaebols

In South Korea, a relatively small domestic market size and scarce natural resources
endowments culminated in an outward-oriented industrial policy in order to re-orient
industrial production in favour of exports. In this perspective, the export orientation was an
inevitable strategy to promote economic development given that on one hand the domestic
market was insufficient, particularly for sectors that exhibit increasing returns to scale, and on
the other hand extractive industries were not internationally competitive to generate foreign
exchange. Therefore, the production and exports of manufactured goods sought to overcome
the lack of foreign exchange to support the industrial development which usually requires
imports of capital goods, components and raw materials not produced domestically. Thus,
taking advantage of its special political status with the United States government, South
Korea focused on the large US market as its main export target (Coutinho, 2000)75.

75
See also Amsden (1989) and Zysman & Doherty (1995).

62
 Over the decades, South Korea sought to plan and prepare its manufacturing industry
for exports, taking into account the structure of the economy and opportunities in world trade.
This trajectory also represented an industrial upgrading strategy from 1950s to 1980s. Thus,
during the 1950s and 1960s, the industrial policy focused on light manufacturing
characterised by unskilled-labour-intensive sectors, then in 1970s shifted to sectors marked by
skilled labour and economies of scale, i.e. particularly heavy and chemical industries. Finally,
in the 1980s, dynamic sectors with high technological content and specialised labour became
part of the industrial structure. In short, South Korea directed their efforts to constitute a
dynamic competitive advantage rather than by their static comparative advantage in cheap and
unskilled labour.

Another important characteristic of the South Korean development was the

tight control over the FDI inflows in the 1950s and 1960s. Therefore, the South Korean
government sought to develop privileged national economic groups in order to propel the
process of industrial growth. The large diversified corporate conglomerates – the so called
Chaebols – were structured over a robust state-controlled financial sector and closely
resembled the Japanese zaibatsu (family-owned conglomerate), which is different from the
current Japanese corporate ‘families’ centred on a bank, namely keiretsu (Chang, 2003). The
Chaebols were largely controlled by their founding families, centralised in ownership,
generally formed by subsidiaries to produce components for exports and heavily dependent on
government loans and loan guarantees in their early years.

Articulated through trading companies, these Korean conglomerates were very

effective tools for the exploration and development of foreign markets along the
industrialization process. Initially, the Chaebols exported traditional manufactured products,
and subsequently as outsourced suppliers began to export parts and components of complex
products. Later, the Chaebols focused on exports of consumer and capital goods, to assume
the position of world leaders in several markets. Recently, these South Korean conglomerates
internationalised their production via direct investments abroad. The following Table presents
the ten largest Chaebol groups from the late 1950s to 2000.

63
Table 2.3 – The 10 largest Chaebol groups in South Korea from late 1950s to 2000
Rank Late 1950s Mid-1960's 1974 1983 1990 1995 2000
1 Samsung Samsung Samsung Hyundai Hyundai Hyundai Hyundai
2 Samho Samho LG Samsung Daewoo Samsung Samsung
3 Gaepung LG Hyundai Daewoo Samsung Daewoo LG
4 Daehan Daehan Hanjin LG LG LG SK
5 LG Gaepung Ssangyong Ssangyong Ssangyong SK Hanjin
6 Tongyang Samyang SK SK Hanjin Ssangyong Lotte
7 Keukdong Ssangyong Hanhwa Hanhwa SK Hanjin Daewoo
Hankook
8 Hwashin Daenong Hanjin Hanhwa Kia Kumho
Glass
Dong-Ah
9 Donglip Panbon Kukje Daelim Hanhwa Hanhwa
Const.
HanilSyn.
10 Taechang Tongyang Daelim Lotte Lotte Ssangyong
Textile
Source: Lim (2010, p. 41).

Several studies pointed out that the establishment and expansion of Chaebol
conglomerates were inspired, directed and supported by the state (Amsden, 1989; Wade,
1992; Chang 1993). Although many measures were implemented to promote South Korean
industrial development, such as devaluation of the domestic currency and restrictions on
foreign investment76, the expansion of South Korean conglomerates relied on reinvestment of
profits and domestic finance, both public and private. The government directed large volumes
of credit to Chaebols through banking institutions, such as The Korean Development Bank,
The Korea Long-Term Credit Bank and The Korea Export-Import Bank, which were directly
or indirectly controlled by the state and provided a predominant share of investment capital in
industry77. As a matter of fact, credit was channelled by the state into specific sectors through
numerous policies and institutions to enhance economic performance and encourage the
formation of export driver clusters. Complementarily, the expansion of South Korean
conglomerates depended on foreign loans (about 20 times the size of FDI), half of which were
public.

Furthermore, in this process, a remarkable characteristic of the South Korean industrial

policy over other developing countries, particularly those in Latin America, relates to the role
of the state to discipline, penalise and reward big business according to their performance. In

76
As stated by Laplane et al. (2013) and, until 1980, an essential characteristic of the economy was restrictions
on foreign direct investment inflows, relaxed only in cases of new technologies and managerial expertise
transfer. According to Leung-Chuen (2001), from 1965 to 1985, foreign direct investment was less than 2% of
gross capital formation.
77
These specialised banks received funds partly from the government, from private deposits and by issuing
bonds in international financial markets.

64
other words, differently from Latin America, incentives and resources to guide private sector
activities were result-oriented, i.e. contingent upon performance. As highlighted by Amsden
(1989), in cases of underperforming, badly managed or bankrupt firms, the state refused to
bail them out and allowed better managed ones to acquire them. In addition to important
measures mentioned before, the state awarded good performers with licenses in other
industries, leading to further diversification. Moreover, the state propelled high-risk infant
industries, rewarding entrants with licenses in more lucrative sectors. Therefore, Kim (1993)
pointed out that this development strategy not only resulted in rapid economic growth and
productive diversification of the South Korean industry, but also affected market
concentration. Thus, in the late 1970s, 93% of all commodities and 62% of all shipments were
produced in monopolistic, duopolistic or oligopolistic market conditions, under which the
three largest producers dominated more than 60% market share78.

3.2. Contrasting developmental patterns

In contrast to South Korea, Brazil has always been recognised internationally for its
large domestic market. Therefore, with an inward orientation, from the 1950s to the end of the
1970s, Brazil achieved high rates of growth through effective industrial policies that
succeeded in a strong industrialisation process, especially in emerging sectors of the national
economy79. This strategy was sustained by the tripod of transnational companies (TNCs),
state-owned enterprises and private domestic firms. During, for instance, the Brazilian
Miracle, the investment rate grew from 16.20% of GDP in 1967 to 20.37% in 1973 and a
large expansion of GDP marked the economy. In this process, the manufacturing share in
GDP reached its historical peak of 23% in 1973.

In this process a strong pro-cyclical movement of investments by state-owned

enterprises (SOEs) marked the pattern of economic growth. Between 1966 and 1973, SOE
investments increased about 19% each year80. The military government had an active
industrial policy and created SOEs with the explicit purpose of developing new industries
essentially to produce basic industrial inputs and to expand the country’s infrastructure. In the

78
At that time this monopolistic pattern was a central characteristic in the South Korean economy. In Taiwan,
for instance, the economy was mostly based on private small and medium enterprises (which had very limited
and overall unsophisticated access to credit) and large state-owned firms. In comparison to Latin American
countries this market concentration did led to the development of rentier capitalism.
79
See also Furtado (1959), Tavares (1972) and Serra (1982).
80
Data calculated using constant domestic prices. For more details, see Trebat (1983).

65
Brazilian developmental agenda, state-owned enterprises were considered a “shortcut to
industrialization—an expediency forced upon policymakers by the absence of a well-financed
domestic private sector and by Brazil’s reluctance to allow transnational corporations into
certain” sectors such as infrastructure (Trebat, 1983, p. 116). Additionally, it was in this
period that the military government (1964-1985) promoted the highest expansion of SOEs.
Figure 2.1 shows the number of SOEs by their year of creation.

Figure 2.1 – Number of state-owned enterprises created in each year (1950-1990)

25
Brazilian
Miracle
20

15

10

0
1950

1952

1954

1956

1958

1960

1962

1964

1966

1968

1970

1972

1974

1976

1978

1980

1982

1984

1986

1988

1990
Source: Musacchio & Lazzarini (2014)

In the developmental era, Brazil – and Latin America as a whole – also sought to
deepen its industrialisation by opening the economy to a massive FDI inflow from the United
States, Western Europe, and eventually Japan for the production of manufactured goods81.
During Kubitscheck’s government there were, for example, policies that aimed to combine
foreign investments in the most dynamic segments of the manufacturing sector (e.g.
automobile) with national raw material suppliers. Even in the face of Brazilian trade
protectionism, TNCs established industrial plants due the promising large domestic market to
explore. In this way, foreign investments were strongly pro-cyclical and associated with
opportunities opened by the industrialisation process particularly during Kubitschek’s
government, the Brazilian economic miracle, and also in the final stages of the industrial

81
Whereas foreign investors in Latin America traditionally had concentrated on export-oriented projects in
mining, oil, and agriculture, post-war FDI emphasised import-substituting investments in advanced
manufacturing industries like automobiles, chemicals, machinery, and pharmaceuticals whose output was
destined primarily for the relatively large domestic markets in Latin America (Gereffi, 1989).

66
structure composition by the time of the Second National Development Plan (II PND) –
during the General Geisel’s Government.

However, it is important to note that the Brazilian pattern of industrial development

generated a productive imbalance in favour of TNCs. Despite some successful cases in the
manufacturing industry, such as Embraer in the aircraft industry, the industrial policy in
Brazil did not foster the creation of nationally-controlled industries as in South Korea.
Moreover, the industrial bases were built on imported ready-made technologies instead of
ones domestically developed through a gradual process of absorption, adaptation and
innovation. Consequently, the economy became gradually more dependent of transnational
companies and state-owned enterprises. During the late 1970s, TNCs were responsible for one
third of industrial production. Moreover, more than one half of the production was
concentrated in specific sectors such as electrical equipment, transport equipment,
pharmaceutical and chemical products. In terms of categories, TNCs had a very significant
share in the production of durable consumer goods and capital goods, but a smaller presence
in intermediate goods and consumer durables. As with the South Korean Chaebols, these
enterprises operated with a high market concentration wherein the 100 largest transnational
companies represented two-thirds of total production in Brazil (Gonçalves, 1996).

Another factor that differentiates Brazil from South Korea is the country's relationship
with banks. As mentioned before, the South Korean state facilitated the accumulation of
capital through public banks (socialising banking risks and controlled margins) that provided
means to business groups to continually increase their debt levels (with long-term financing
and low interest rates) in order to successively undertake ambitious investment projects in
consonance with the five-year development plans. In contrast, in Brazil, banks were never
nationalised and the private banking system was relatively distant to the industrialisation
process, limited to short-term financing for working capital without offering long-term and
relevant credits for large investments. Therefore, it was up to the public banking, i.e. BNDES
and Banco do Brasil (BB), to provide long-term financing to the development of the national
industry. In the case of transnational companies (and large state-owned enterprises),
additional resources of long-term financing were obtained through external loans and
financial resources provided by parent company (Coutinho, 1999).

67
 4. External Debt and Exogenous Shocks

In spite of different trends regarding the interaction with the external market, in terms
of productive structure, both Brazil and South Korea achieved the phase of heavy
industrialisation. However, during the 1970s, the massive effort to climb the ladder of the
world production in a turbulent economic environment profoundly affected the public
accounts. In both Brazil and South Korea, although in different proportions, given the lack of
foreign exchange, the military government had gradually accumulated external debts with
international private banks to finance the current account deficit and maintain the booming
economic growth. This excessive external financing increased economic vulnerability and
exposed these countries to external shocks. However, South Korea was less affected, as its
path of “self-reliance” and import substitution industrialisation relatively made it less
insolvent in terms of external debt-exports ratio, allowing a leeway in its economy (Maddison,
1985; Singh, 1985).

The two oil price shocks in 1973 and 1979, and the Volcker interest rate hikes in 1979
had huge deleterious effects in both economies. The former affected the trade balance of oil
dependent developing countries while the latter triggered the surge of international interest
rates that increased the debt burden of emerging countries. For instance, the Brazilian external
debt in real dollars rose sharply from US$ 14.8 billion in 1973 to US$ 55.8 billion in 197982.
At the same time, the net current account deficit increased from US$ 2.0 billion (2.48% of the
GDP) to nearly US$ 11.0 billion (4.79% of the GDP)83. To make matters worse, Brazil's
inflation rate peaked dramatically to 77.25% in 197984. In the case of South Korea, adverse
consequences of the economic environment were similar, although of a different magnitude.
The external debt in real dollars increased from US$ 4.3 billion in 1973 to U$S 20.3 billion in
1979. Moreover, since South Korea was dependent on imported oil to foster the development
of petrochemicals industries, the increase of oil prices affected negatively not only the
inflation that reached 28.7% in 1980, but also the industrial productive capacity.
Consequently, in 1980, the investment rate in GDP dropped 3.40 percentage points and the
country showed a negative economic growth of 1.70%.

82
Source: Ipeadata.
83
Source: Brazilian Central Bank.
84
Source: Ipeadata.

68
Figure 2.2 – External Debt, total (% GDP) and Total external debt/exports (ratio), 1961
- 2013
External Debt, total (% GDP)

Brazil South Korea

60 60

50 50

40 40

30 30

20 20

10 10

0 0
1961
1965
1969
1973
1977
1981
1985
1989
1993
1997
2001
2005
2009
2013

1961
1965
1969
1973
1977
1981
1985
1989
1993
1997
2001
2005
2009
2013
Total external debt/exports (ratio)
Brazil South Korea
5 5
4.5 4.5
4 4
3.5 3.5
3 3
2.5 2.5
2 2
1.5 1.5
1 1
0.5 0.5
0 0
1961
1965
1969
1973
1977
1981
1985
1989
1993
1997
2001
2005
2009
2013

1961
1965
1969
1973
1977
1981
1985
1989
1993
1997
2001
2005
2009
2013
Source: Elaborated by the author, using data from The Bank of Korea (BOK), Collins & Park (1989), World
Bank, Brazilian Central Bank and OECD.

The nail in the coffin of such macroeconomic imbalance was Mexico's default in 1982
that culminated in an abrupt credit crunch for developing countries in the second half of that
year. In this context, a recessive economic environment plagued the economies with high
external debt. This problem associated with protectionist policies implemented by developed
countries during the 1980s complicated still further the economic environment for emerging
economies. Additionally, a strong economic integration took place between OECD economies
under the aegis of financial globalisation. It led to a heavy process of international mergers
and acquisitions associated with high direct investment flows by large companies from

69
industrialised nations. In this way, the rise of regional trading blocs and the intensification of
intra-industry and intra-firm trade marked the period. In short, the process of greater
interdependence and integration that took place in the 1980s was concentrated within the
OECD and a handful of eastern Asian countries, including South Korea (Coutinho, 2000).

The restrictive access to international finance and foreign-exchange constraints began

to obstruct output growth, investment and fiscal revenues. Indeed, Latin American countries
were derailed by exogenous shocks mainly because of their high external debt rather than
their closed trade regimes (Hughes & Singh, 1991). Moreover, in a strategy of development
based on deficit-financing to keep up growth impetus, the liquidity contraction triggered a
fiscal and financial fragility of the state that culminated in a deep disarticulation of the state to
provide active policies for national industry (Carneiro, 2002; Belluzzo & Almeida, 2002).
Furthermore, the overall scenario of uncertainty marked by the threat of hyperinflation and
disarticulation of the state blocked the private sector from engendering economic recovery.
Unlike South Korea, due to a strategy based on transnational companies and state-owned
enterprises, the domestic business sector was unable to centre itself on large private groups
capable of competing internationally (Coutinho, 2000). The glorious decades of the Brazilian
developmental era, marked by the highest levels of manufacturing share in GDP and
investment rates, ended with one of the most radical policy shifts in the history of developing
economies. It moved away from its state-led ISI policy into a massive process of economic
‘liberalisation’ and de-regulation (Palma, 2005).

South Korea, in contrast, adopted a more pragmatic and progressive neoliberal agenda.
Given the assassination of General Park Chung Hee in 1979 and the debt crisis, in early
1980s, the South Korean economy went through a period of recession and restructuring.
However, unlike Brazil, in a much more constructive way, the country implemented measures
to overcome the crisis and strengthen the ambitious growth strategy. Therefore, after this
turbulent period, under the government of General Chun Doo Hwan (1980–1987), the
economy achieved a rapid recovery. In this process the decisive alliance between the South
Korean government and the Japanese economic system was decisive. On the financial side, as
explained by Canuto (1994) and Coutinho (2000), the Japanese banks had a decisive role in
face of the US credit crunch giving the necessary means to promote a new dynamic decade.
As described by Iqbal (1988), the US net loans by large US banks declined from US$ 2.3
billion in 1981 to US$ 0.7 billion in 1983, creating a slack taken up by Japanese banks and by
alternative instruments such as bonds and notes. From 1981, bonds and notes rose from less

70
than 2% of annual gross medium- and long-term borrowing to around 23% by 1984 (ibid,
147). On the productive side, the country linked its leading industrial sectors to the South
Korean production system, enabling the country advance in technological learning and
manufacturing upgrading85.

Throughout the 1980s the Korean government introduced additional measures to

enhance industrialisation policies and strengthen the economy’s industrial pillars. The state
did not expose domestic producers to international competition in an ideological
fundamentalist neoliberal way, but instead directed and coordinated investment decisions
(‘disciplining’ the capitalist élite) to achieve levels of technology and productive structure to
make them able to compete in external markets. In face of the international economic
recovery led by the American growth after 198386, South Korea and other Asian NICs
invested massively in their manufacturing goods exports without a harmful counterpart on
credit levels and inflation as in Latin America. Moreover, the Korean developmental state
guided the capital accumulation process in order to overcome the obstacles to peripheral
industrialisation.

5. Technological Learning and Industrial Development

During the 1980s, while Brazil struggled in the ‘lost decade’, the South Korean
government seeking to embark on a new cycle of expansion encouraged the country’s large
Chaebols to establish partnerships with Japanese leading companies. As a bargaining chip,
South Korea offered to Japan its capacity to mobilise resources and qualified working force to
produce components and/or assemble electronic products under the original equipment
manufacturer (OEM) agreement. At the same time, Korean enterprises started to acquire
complete production units from Japan to accelerate the learning process mainly in the
electronics sector (consumer goods, telecommunications, information technology (IT),
memory semiconductors and capital goods). Moreover, Japanese engineers were hired to
transfer manufacturing know-how, essentially new knowledge and techniques, to their South
Korean colleagues (Coutinho, 1999).

85
In this regard, another special peculiarity was the emergence of salaried managers responsible for exploiting
the borrowed technology and focus on shop floor management to optimise technology transfer (Amsden, 1989).
86
From 1983 to 1989 the US grew on average 4.43%. In this period, the most expressive GDP growth was in
1984 when the country registered an expansion of 7.26%. GDP growth rate at constant dollars prices. Source:
The World Bank.

71
 Despite the neoliberal rhetoric at that time, the state continued determining the
direction and priorities of the development process87. Therefore, the fifth five-year plan
(1982–1986) turned to the development of high-tech industries, particularly, IT and electronic
industries which received essential support from the state. A strong thrust was given to
electronics, including production of memory chips used in computers, consumer goods, and
telecommunications products. Inspired by the Japanese industrial policy, the Government
established Deliberation Councils in order to act as a state instrument to identify and target
industrial opportunities, evaluate process of merging and acquisitions, and approve financing
suggestions proposed by lending institutions (Chang, 1994). Furthermore, large Chaebols that
were already present in the electronic consumer goods sector since the 1970s, such as
Samsung and Lucky Goldstar, received further incentives to undertake new projects.
Additionally, from the mid-1980s, in an effort to lessen the South Korean trade deficit with
Japan, the government strengthened the incentive structure for the automobile industry and
capital goods industry sector 88(Coutinho, 2000).

Given the long-term planning of the South Korean economic policy, the sixth five-
year plan (1987-1991) undertook policies to improve competitiveness combined with
measures to enhance the technological capacity of the production system. As stated by Chang
(1993), the constant upgrading of the industrial structure based on the development of local
technological capabilities was seen by South Korean policy-makers as the most concrete way
of achieving sustained growth and structural change. In this sense, technology acquisition and
licensing contracts were encouraged. Moreover, R&D incentives were expanded in an effort
to promote the private sector’s R&D investment. Furthermore, tax credits for corporate
expenditures on human resources development were also implemented (Lee, 2000). Under
clear guidelines, the government sought to encourage the private sector to take initiatives in
order to prepare South Korea’s industrial base to reach the world's technological frontier. In a
Schumpeterian perspective, the state provided the necessary means to constitute a dynamic

87
As argued by Palma (2004), the economic reforms implemented in East Asia were less fundamentalist when
compared to Latin America due the distinct economic level of economic fragility during the debt crisis and
external political pressure from the US government.
88
In terms of trade policy, during the 1980s, the large US market was the main focus of the South Korean export
programme. For this purpose, the government sought to avoid its currency’s overvaluation against the dollar.
Moreover, the South Korean government took advantage of its special political relationship with US and
expanded the exports. The growth of the US economy pulled Korea’s exports generating a substantial
improvement in the trade balance. The surplus of dollars helped to balance the South Korean deficit with Japan.

72
national innovation system (NIS) for the development and spread of new technologies89. The
following Table reveals that these efforts resulted in a drastic jump in terms of innovation
activities, principally from 1980 to 1985.

Table 2.4 - Major indicators of innovation activities

1970 1975 1980 1985 1990 1995
Total R&D/GNP 0.39 0.44 0.58 1.56 1.88 2.71
Private R&D/GNP 0.09 0.15 0.21 1.17 1.52 2.20
Governamental R&D/GNP 0.30 0.29 0.37 0.39 0.36 0.51
Number of Researchers 5,628 10,275 18,434 41,473 70,503 128,315
R&D expediture/researcher
(Thousand Won) 1,874 4,152 15,325 27,853 49,514 73,574
Researcher/10,000 popular 1.80 2.90 4.80 10.10 16.40 28.60
Number of corporate R&D Labs 0 12 54 183 966 2,270
Applied Patents 1,846 2,914 5,070 10,587 25,820 78,499
Utility Models 6,617 7,290 8,558 18,548 22,654 59,866
Industrial Design 4,522 6,707 10,075 18,949 18,769 29,978
Trademark 5,124 9,476 13,558 26,069 46,826 71,852
Source: Lee (2000).

In this strategy, the alliance with the Japanese economy was decisive. From Japan’s
point of view, the outsourcing of certain products under OEM agreements to countries within
the Asian axis represented a crucial strategy. It helped to maintain the Japanese
competitive edge in export markets, especially after the yen overvaluation due to the Plaza
accord in the second half of the 1980s. In the South Korean counterpart it was a great
opportunity to assimilate the last stages of the productive process and to develop internally its
own technologies. Subsequently, South Korean companies started to develop gradually their
Own Brand Manufacture (OBM), replacing the OEM strategy. For example, while supplying
Japan with intermediary inputs such as petrochemical products, paper, steel products, and
non-ferrous metals, South Korea acquired capital goods, especially machinery, from leading
Japanese companies and redeveloped them according to its own industrial needs. As a
developing economy, South Korea took advantage of the technological gap with Japan to
successfully absorb, assimilate and adapt cutting edge process and products and consequently
boost the economy in terms of productivity and GDP growth.

In short, South Korea’s partnership with Japan was fundamental not only for the
recovery of the economy from the debt crisis, but also provided the means to move ahead
enhancing the industrial structure toward the new paradigm of microelectronics90. Through a

89
For a detailed analysis on the dynamic involving technological dynamics, innovation and economic growth,
see the first chapter of this thesis.
90
For more details about technological revolutions, see Perez (2002).

73
strategic export dynamism and industrialisation, South Korean policy makers sought to
maximise the benefits of FDI inflows and technology that they could extract from
transnational corporations, to ensure efforts to strengthen domestic capacity (UNCTAD,
2002). According to Amsden (2009), the development of technological capabilities was the
most important factor of differentiation between South Korean developmental trajectory and
the Latin America one. In Latin America, the limited success attained by scientific-
technological development occurred due to the absence of technological learning policies
linked to the expansion of the industrial base and exports promotion. As Fajnzylber (1983)
pointed out, ISI was marked by a ‘frivolous protectionism’, in the sense that it did not protect
the domestic industry with the perspective to assimilate and develop cutting-edge
technological industries and strengthen competitiveness91.

This situation got worse in face of the external debt crisis of the 1980s that did not
allow Brazil to keep up with the new technologies and the country’s manufacturing industry
became delayed compared to other latecomer economies. In contrast, the South Korean state
played a central role coordinating the private sector (“disciplining” the capitalist élite),
establishing priorities and managing instruments and incentives to establish a dynamic
industrialisation and rapid technological progress. During the debt crisis, unlike Brazil, the
economic policy comprised a strategy to enhance the manufacturing technological upgrading.
Therefore, through a close relationship with Japan in terms of commercial, technological and
financial agreements, the South Korean state strengthened the Chaebols in order to build a
competitive electronic complex to become a world leader in terms of brands, technologies
endogenously created and large companies with a worldwide presence (Coutinho, 1999).

91
In broad terms, it is important to emphasise that this pattern of technological assimilation was a common trend
in other Asian countries. For example, China has engaged itself in the transformations of global productive
structures through a clear orientation of its national development policy aiming not only to compete in various
markets, but also to absorb and develop capabilities from the technology transfer of multinational companies
operating in the country, in order to create competitive national groups (Nolan, 2001). The Chinese dynamism
focused not only on producing cheap manufacturing goods but also on the huge market for the world production
of machinery and equipment with high-level technology. The high-performing East Asian countries recognised
the imperative of joining the world economy through the promotion of labour-intensive manufactured exports,
but clearly did not limit its expansion in this statement. Therefore Asian countries successfully expanded the
manufacturing sector in a process of assimilation, incorporation and diffusion of new products and techniques,
with a consequent increase in the production and productivity which changed the underdeveloped position in the
global GDP share. In this way, it is no mere coincidence that Asia constituted much more technology intensive
industry than Latin America.

74
 6. The “fundamentalist” neoliberal shock

The abrupt and very painful internal/external macroeconomic adjustment in Brazil –

much larger than suffered by South Korea – opened the way for a radical and widespread
change in the economic paradigm of the region and the emergence of the neoliberal agenda92
(Palma, 2003). Both exogenously and endogenously, the neoliberal view spread through the
perception that severe macroeconomic imbalance and the institutional framework of a model
based on ISI did not promote internalisation of the changes observed in the external
environment. Authors such as Franco (1998), Bacha & Bonelli (2005), Krueger (1985) and
Balassa (1988) attributed to the ISI model and the strategy based on the infant industry
protection the relative prices distortions which, in turn, resulted in low rates of growth,
investment and productivity. According to the neoliberal camp, ISI led to an inefficient
allocation of resources and impacted negatively the level of competitiveness of the economy.
Through this interpretation the state-led industrialisation based in the prevailing economic
policy started to be harshly questioned by a new view in the literature known as the
"Washington Consensus" or, in the words of Franco (1998), "the new economic model"93.

Therefore, the period which started in the 1980s with neoliberal reforms contrasts with
the preceding decades when economic growth was mainly based on industrial diversification,
with the manufacturing industry playing an important role in the development strategy94.
Since the 1980s, the adoption of the neoliberal “fundamentalist” agenda supported by the
World Bank and the International Monetary Fund (IMF) has exposed not only Brazil, but all
of the Latin American economies to a doubtful development strategy based on the free-market
doctrine95. In the face of the neoliberal ascendancy, industrial policies and foreign trade
interventions were gradually denied. In this sense, policies in favour of more trade and
financial openness were implemented in pursuit of a reduced role of the state in the economy.
These guidelines were strongly marked by an extreme move towards trade and financial
liberalisation, market deregulation, attraction of new flows of FDI, privatisation of state-
owned enterprises, and withdrawal of selective industrial policies.

92
Palma (2000, p. 6), states that the rise of the fundamentalist neoliberal thinking in Latin America was mostly a
consequence of the substantial economic weaknesses of these economies. Likewise in the 1930s, “an external
shock that found Latin America in an extremely vulnerable position not only brought about the need for a very
painful internal and external macroeconomic adjustment, but also laid the foundations for a radical and
widespread change in economic thinking”.
93
See also Kuczynski & Williamson (2003).
94
See also Hughes & Singh (1991).
95
See, for instance, Frenkel (1992).

75
 The implementation of economic reforms in Brazil began in the late 1980s, after the
onset of reforms in other Latin American countries such as Argentina, Mexico and Chile
(Abreu, 2003). The main objective of the reforms was to dismount the ISI apparatus and
establish a market-oriented model. In other words, neoliberalism emerged in Brazil to reverse
almost every aspect of the growth strategy previously followed by the country. Thus, it was
taken for granted that the recovery of economic growth would be strictly connected to the
reduction of state intervention. From this perspective, “market deregulation and trade
liberalisation would not only stimulate tradable production but would also significantly
increase private investment, while higher interest rates would boost domestic saving and
reverse capital flight” (Palma, 2005, p.18). The recovery of private investments would
ultimately lead to an increase in the investment-to-GDP ratio, interrupting the downward
trend observed in the 1980s.

The restructuring of production and trade would take an essentially positive character
resulting in significant allocative and in technical efficiency as a result of increased scale of
production. Moreover, it would contribute to the resumption of the economic growth
trajectory in a more specialised production structure in line with the country's comparative
advantages, allowing the shift of emphasis from domestic to foreign markets as a driver of
economic growth (Moreira, 1999). Moreover, the neoliberal view interpreted the weakness of
the domestic industry as a consequence of the protected market and vertical integration of the
productive system, which in turn, would block access to high-quality inputs, impeding
domestic industry from adopting large-scale production methods. The industrial policy at that
time turned to the adoption of horizontal policies vis-à-vis vertical ones. Furthermore, the
exportability of the economy would increase in face of the specialisation in sectors in which
the country has comparative advantages, thus reinforcing the necessity to promote the
development through the market mechanisms. Additionally, it was hoped to have the support
of foreign capital to head a new investment cycle with greater specialisation, modernisation,
productivity, technology transfer and international integration.

These propositions have turned to a structural change in the pattern of growth based on
production diversification supported by ISI, with a strong state directing. The model focused
on the productive capacity expansion with the protection of the domestic industry was
abandoned in favour of policies for market openness. The domestic industry was exposed to
international competition without any sectoral selectivity. It was expected of the Brazilian
industrial sector to produce in terms of price and quality compatible with the world market,

76
enhancing the competitiveness through market mechanisms. The process should not be driven
by demand anymore, particularly by public investment, but based on changes on the supply
side and grounded on competition and technological innovation (Barros & Goldstein, 1997).

Although lacking a pragmatic interpretation, the neoliberal view supported by the

World Bank and IMF gradually promulgated the idea that the adoption of the neoliberal
agenda would replace positively outdated elements of the “old economic model”. These
interpretations changed the way that economists and policy makers understood the problems
faced by peripheral countries and their possible solutions96. Even though distant from a sense
of reality, these studies suggested that developing countries which adopted a market led
strategy, i.e. liberalisations and great exposure to the free-market forces, achieved high levels
of economic growth.

Therefore, the “Structuralist-Kaldorian” way to visualise the problems in the

peripheral countries was left out97. The Kaldorian links running from growth to productivity,
related with the presence of underutilised resources during the growth process, have also been
ignored, as well as the links between productivity convergence, domestic production linkages
and external balances (Ocampo & Parra, 2006). Policies designed to promote the
manufacturing sector lost strength during this period and were virtually abandoned as a result
of the then-prevailing maxim that “the best industrial policy is no industrial policy” (Stallings
& Peres, 2000). As pointed out by Palma (2013, p. 3), “what characterises Brazil’s (and the
rest of Latin America’s) economic reforms was an attitude of ‘throwing in the towel’ vis-à-vis
the previous state-led import substituting industrialisation strategy (ISI)”98. In this context, the
rentier characteristic of the country’s capitalist élite was reinforced undermining any
perspective of sustained economic growth99.

In contrast, as indicated in the vast literature on East Asian industrialisation, the

strategy followed by South Korea (and most East Asian countries) has been remarkably

96
See Williamson (1989).
97
See, Amsden (2001), Lall (2001) and Palma (2004).
98
According to Palma (2004), Latin America moved to “a standardised environment in which the only issue at
stake was who would best (and most faithfully) implement the 'new' model”.
99
The term ‘rentier’ is applied in a Keynesian perspective. According to Keynes, the rentier preference for
liquidity affects negatively the level of productive investment and employment. In this perspective, for example,
the economic dynamic is undermined by rentier forces that operates in a speculative manner and influence the
economic policy for its own interests. For that reason, Keynes advocated for 'the euthanasia of the rentier' and
the creation of a new kind of capitalist culture of cooperation between private and public authorities (Keynes,
1936).

77
different from trade liberalisation, in the traditional (and “fundamentalist’) sense100. In South
Korea, the economic reforms were implemented mainly as a pragmatic mechanism to
strengthen the existing development model (Palma, 2003). It means that, in the same period,
South Korea managed to transform its initial catching-up efforts into a dynamic and
sustainable process of capital expansion and development. Rather than a pure and simple
adherence to neoliberalism, the Asian experience has been characterised by the maintenance
of a strong state control over the socioeconomic transformations and by the pragmatism in the
management of intermediate and instrumental objectives such as the economic
internationalisation and administration of the fiscal and monetary policy (Cunha & Acioly,
2009). In this sense, after the 1980s the history in the global development would was very
different from the previous decades, with all gains of economic growth obtained practically
only by the South Korean economy, or in terms of region – vis-à-vis Latin America – by
developing Asia101.

7. Brazil falling behind and South Korea moving ahead

7.1. Brazil: FDI flows and the loss of industrial dynamism

As aforementioned, after the golden age of the industrialist era, the neoliberal agenda
promulgated by free market prophets in the IMF and World Bank and their faithful apostles –
particularly in Latin America – led to a productive restructuring of the Brazilian economy. In
terms of FDI, the massive inflow in the country was prompted by a dramatic process of
privatisation, especially in the second half of the 1990s. Therefore, the amounts of FDI
directed to the country were much higher than previous decades mainly due to cross-border
mergers and acquisitions (M&A) involving SOEs. Sales of stated-owned enterprises increased
in the second half of the 1990s, peaking in 1998 at the height of the privatisation process, and
remained high in the early 2000s, consolidating the presence of international groups in the
national productive structure102. Moreover, FDI inflows were directed to the consolidation of

100
See, Ocampo & Parra (2006).
101
The booming growth clustered in the 1960s and 1970s were widespread in the developing world during the
“golden age”, tended to disappear in the 1980s, except in East Asia, and became somewhat more common in the
1990s, but at levels far below those of the “golden age” (Ocampo & Parra, 2006).
102
During the National Programme of Desestatisation (Plano Nacional de Desestatização – NPD), the sales
revenue of federal state-owned enterprises since the creation of PND in 1990 until December 2008, totalled U$S
30.8 billion. This amount, plus the amount of debt transferred to the private sector of around $ 9.2 billion,
represented a total result for the NPD in the order of $ 40 billion. With regard to telecommunications, the sales

78
oligopolistic structures on a global scale, the size of which is evidenced by the amount of
M&A operations. This profile of FDI flows helps us understand why investment flows
remained high even after the Asian crisis in 1997, the Russian crisis of 1998 and even the
Brazilian crisis that resulted in the devaluation of the Real in 1999 (Sarti & Laplane, 2002).
After this period, the data reveals a transformation of the pattern of M&A, i.e. the process was
not restricted to the purchase of state-owned enterprises, and thus a representative amount
M&A operations involved the private sector between 2003 and 2008 (Carneiro, 2002 ).

Figure 2.3 - Brazil: Foreign Direct Investment, 1990-2008

50
45
40
35
US$ billions

30
25
20
15
10
5
0

FDI M&A

Source: Elaborated by the author, using data from Unctad.

FDI in the form of M&A involving state-owned enterprises and the private sector
culminated in an increasing foreign participation in the domestic economy structured on the
existing stock of productive capacity, with the aim of directing the production mainly to the
domestic and the regional markets. Therefore, especially in the second half of the 1990s, there
occurred an intense change in the ownership structure of the companies. However it did not
trigger a notable expansion of the productive capacity, i.e. greenfield investments, and
consequently did not raise the investment rate of the economy. In contrast to previous
decades, when investment in new production plants were the main mechanism to reach new
markets, the FDI during the 1990s turned to acquisitions of Brazilian enterprises as the main
mechanism to access the domestic market.

revenue for the same period was around U$S 29 billion, totalling U$S 31.1 billion when added U$S 2.1 billion
of transferred debts. The overall result of privatisations carried out in Brazil from 1990 to 2008 reached US$
105.8 billion distributed at federal and state levels.

79
 As noted by Laplane et al. (2001), the increased participation of transnational
corporations in the Brazilian productive structure and foreign trade did not result in significant
structural change in favour to the expansion of the manufacturing sector neither in a positive
evolution in terms of international insertion. In addition, Transnational Corporations (TNCs)
began to integrate the new global productive paradigm expressed in the concept of global
value chains. Thus, TNC operations began to be fulfilled in a more fragmented value chain
that is geographically dispersed. In this regime, multinational subsidiaries in Brazil operated
mostly as buyers of products from other parts of the global value chain to meet domestic and
regional markets. In rare exceptions, these subsidiaries followed the role of global suppliers
within this international division of labour (Hiratuka, 2003).

Moreover, it is important to highlight some important changes in the sectoral

composition of FDI flows received by the Brazilian economy after the monetary stabilisation
and intensification of neoliberal reforms. Through the comparison of accumulated stocks of
FDI until 1995 and until 2000, it is possible to see the depth of changes during the period.
Thus, the data shows practically an inversion of the percentage between manufacturing and
services. Furthermore, unlike 1990 and 1995, between 1996 and 2000, services received
63.96% of FDI, while the industry and primary sectors received 33.71% and 2.33%
respectively. These numbers confirm the concentration of FDI flows in non-tradable sectors,
which still require remittance of profits and dividends without generating foreign exchange
through exports. Therefore, the boom in FDI flows occurred mainly in the service sector via
privatisations in telecommunications and electricity, as well as the restructuring of the
financial system. Thus, due to the opening of the economy and the privatisation process, there
was a fall in the share of industry in FDI vis-à-vis sectors related to public services.

From 2001 to 2003, the net inflow of FDI fell to $ 12.9 billion, largely influenced by
uncertainties regarding the new government and the turbulent international context after 2001.
The average share of FDI directed to services declined while that in the primary sector
increased. The former fell to 56.5% and the later grew to 7.3%. The manufacturing sector
increased its share to 36.1%, led mainly by the automotive sector. In 2004, with the new wave
of international liquidity and lower domestic economic instability, the FDI inflow reached
US$ 20.3 billion, with significant share of 26.4% (US$ 5.3 billion) in the food and beverage
industry. The sectoral analysis of FDI inflows indicates that, in spite of the concentration in
services, the primary sector showed a substantial growth impacted by the commodity prices

80
boom. The FDI maintained an increasing rate, reaching US$ 21.5 billion in 2005, US$ 22.2
billion in 2006, US$ 33.7 billion in 2007 and US$ 43.9 billion in 2008.

Table 2.5 - Stocks and inflows of foreign direct investment in Brazil, by sector
Stocks (%) Inflows (%)
Sector 1995* 2000* 2001 2002 2003 2004 2005 2006 2007 2008 2009
Agriculture, livestock and min. 2.22 2.33 7.10 3.40 11.53 5.29 10.20 6.13 14.78 29.61 14.70
Manufacturing 66.93 33.71 33.27 40.23 34.92 52.84 29.75 39.33 36.10 31.93 39.17
Automobile 11.60 6.17 7.37 9.36 7.48 4.20 4.30 1.29 2.59 2.20 7.11
Chemical Products 12.79 5.87 7.35 8.38 7.10 6.73 3.55 5.10 2.23 2.46 3.62
Food and Beverages 6.78 4.48 2.67 9.97 3.17 26.38 9.64 3.33 5.39 5.10 1.78
Eletric Mat. and Communic. Equip. 1.88 2.11 5.54 2.90 2.53 1.31 1.84 1.46 2.38 1.92 2.46
Services 30.85 63.96 59.63 56.37 53.55 41.87 60.05 54.54 49.12 38.46 46.13
Post and telecommunications 0.96 18.21 19.63 22.32 21.77 14.66 8.83 5.47 1.81 2.20 4.82
Electricity, gas and water 0.00 6.91 6.85 8.17 5.03 5.82 7.30 10.49 1.83 2.07 3.19
Financial Intermediation 4.87 11.80 10.09 6.77 4.69 4.64 6.01 13.46 17.35 8.92 8.30
Sales (retail and wholesale) 6.72 9.52 6.87 7.64 6.30 5.89 12.92 6.57 8.21 5.84 9.09
Total (%) 100 100 100 100 100 100 100 100 100 100 100
US$ (Millions) 41,696 103,015 21,042 18,778 12,902 20,265 21,522 22,231 33,705 43,886 30,444
Source: Elaborated by the author, using data from The Brazilian Central Bank.
Note: (*) Stock accumulated.

In short, the strategy adopted by the government to attract FDI flows did not propel the
manufacturing sector or exports in the 1990s due to three main reasons. First, FDI was
directed to non-tradable sectors during the privatisation process, such as public utilities.
Second, a large portion of FDI was in the form of mergers and acquisitions, not expanding the
productive capacity. Third, the logic of investment, even in manufacturing sectors, was
primarily associated with the aim to explore the potential of the domestic market without
policies to promote internal intersectoral linkages. In the 2000s, except in 2004, the FDI flows
were directed to other industries but not effectively channelled to tradable sectors, specifically
manufacturing. Furthermore, during this decade, the FDI focused on the primary sector
essentially due to the global commodities prices boom.

7.2.The loss of manufacturing and productive disarticulation

Although from the neoliberal rise onwards the average annual growth rate of the world
economy fell from 4.67% (1950-1979) to 3.34% (1980-2010), in Brazil it was still more
drastic. The late 1980s and the whole decade of 1990s were marked by neoliberal reforms that
took form practically eliminating quantitative controls, non-tariff barriers to trade103, reducing

103
According to Carneiro (2002), non-tariff barriers to trade, which many saw as the main protectionist
instrument, were completely removed after Annex C (a list of 1,300 products whose imports were forbidden
because similar domestic products were available) was abolished.

81
import tariffs104 and denying public policies focused on promoting growth in strategic sectors
for the country’s industrial development. This environment, combined with currency
appreciation and extremely high interest rates to control inflation resulted in a negative
restructuring of the domestic economy in the second half of the 1990s. During this period, to
face the challenges of the new adverse economic environment, companies began to take strict
adjustment measures to rationalise their production by replacing local inputs with imported
ones105. Therefore, as argued by Belluzzo & Almeida (2002), there was a “shrinking” of
supply chains, which were also affected by “predatory” imports. Thus, industrial companies
began to look for ways to improve their competitiveness by cutting costs, replacing local
products with imported inputs and reducing inter-sectoral linkages still under development
(Rocha, 2011).

In the face of an adverse economic environment to expand the manufacturing sector,

short cycles of demand were met by imports106. In an economic environment marked by
currency overvaluation, as demand expanded, this dynamic was constantly reproduced. From
2002 the commodities boom favoured the economy and it was associated with a better
international context, leading to optimistic expectations about the aggregate demand. In this
context, both the levels of investment and the installed capacity showed a rising trend.
However, due to the productive disarticulation legacy and the strategy of economic
development strongly dependent on the primary sector, the levels of investment to expand the
productive capacity were not sufficient to keep up with demand growth. Moreover, the
massive inflow of capital due to both trade relations with Asian economies and speculative
financial transactions resulted in a strong appreciation of the domestic currency. This process
caused a consequent increase in imports of manufactured goods, which rose by 155% at
constant prices between 2002 and 2008107.

During the period of “overrated boom”, not even manufacturing activities associated
with commodity extraction and processing were expanded. Moreover, although still relatively
104
The custom tariffs were reduced based on the country’s structure of comparative advantages. Nominal import
tariffs were reduced by 55.3% between 1990 and 1994, with the maximum tariff not exceeding 40%.
105
See Rocha (2011), for more information on the imported input coefficients.
106
According to the Brazilian Institute for Geography and Statistics (IBGE), the penetration coefficient for
intermediate goods rose from 2.7% in 1990 to 10.5% in 1998. As argued by Laplane & Sarti (2006, p. 276), from
a trade balance perspective, this process “turned the surplus in the trade in manufactured goods registered in the
first half of the decade into a deficit from 1995 on, clearly indicating that it would be difficult to keep the
economy on a growth path. The trade balance was more significantly negative precisely in 1997, when industrial
production was growing at the highest rates, reinforcing the interpretation that the increasing imported contents
of local products were generating an even more pronounced deficit.”
107
Calculated by the author based on data from FUNCEX (Foreign Trade Study Centre Foundation).

82
diversified, most of the companies operating in Brazil are not closely related to each other and
still very concentrated in low-tech segments, expressed by the predominance of resource-
based companies, basic input suppliers and consumer goods producers (Fleury & Fleury,
2009). The absence of a wide manufacturing industry and dense supply chains has limited
growth in terms of technological progress and productivity, causing an excessive dependence
of exogenous expansions.

Given this aforementioned dynamic, perhaps one of the main results of the
“fundamentalist” process of economic reforms was the undermining of any prospect of rapid
and sustained economic growth. The only achievement of neoliberal policies was to establish
an economic dynamic of successive “stop-and-go” cycles derived from a fragile productive
structure. Therefore, from the 1980s onwards, the country’s growth rate collapse was extreme
both in terms of GDP and manufacturing value added growth. The former sharply declined
from 6.74% to 2.75%, while the latter slumped from 8.41% to 1.90% in the respective period.
In 1980, the share of manufacturing in GDP represented 21.13%, however, in 2010 this
number dropped dramatically to 17.96%. Furthermore, data for the Latin American region
confirmed this trend. During the period of 1950 to 1979 the region grew on average 5.39%,
and from 1980 to 2010 this average rate fell by half, i.e. to 2.73%108.

Figure 2.4 – Brazil: Manufacturing Share in GDP and Value Added by Technology
Intensity (%)
40 100%

90%
35
GDP (1950-79) = 6.74% High Tech
80%
Manuf. Share in GDP

30
GDP (1980-10) = 2.75% 70%
25 60%
22.03 21.83 21.13
19.72
20 18.37 17.89 17.96 50% Medium Tech
40%
15
30%
10
20% Low Tech
5 10%

0 0%
1950 1960 1970 1980 1990 2000 2010 1970 1975 1980 1985 1990 1995 2000 2005 2010

Source: Elaborated by the author, using data from The Groningen Growth and Development Centre and
ECLAC-PADI database.
Note: Manufacturing gross value added at constant 2005 national prices (in millions). Classification of
manufacturing industries by technology group based on UNIDO (2013). For details, see appendix 2.3.

108
See Appendixes 2.1 and 2.2.

83
 In contrast, the analysis of the data for South Korea shows a very different
development path. From 1953 to 1979, South Korea registered an average of GDP and
manufacturing value added growth of 8.70% and 9.51% respectively. However, after the
neoliberal reforms, the country did not experience such dramatic loss in terms of economic
dynamism. Although from 1994 onwards the economic reforms implemented in the country
sought to strengthen liberalising policies initiated in the 1980, its nature was totally different
from the huge endeavour seen in Brazil. As pointed out by Palma (2012), economic reforms
in Korea were implemented as pragmatic mechanisms to help lift specific pressing economic
and financial constraints in order to continue and strengthen their existing ambitious
industrialisation strategies. Even during the Asian crisis in 1997, in spite of the re-articulation
between state and private sector, the role of the state was central. To overcome the crisis,
there was a reorganisation of the corporate structure with processes of mergers and
acquisitions gaining momentum. However, it is important to note that in this process the state
played a central whole to stimulate companies to focus on their core activities. Therefore, the
South Korean state remained active either through long-term programmes for the
development of high-tech sectors, or ensuring important long-term funding sources through
public banks.

Figure 2.5 – South Korea: Manufacturing Share in GDP and Value Added by
Technology Intensity (%)
40 100%
GDP (1980-10) = 6.33% 35.24 90%
35
High Tech
30.28 80%
Manuf. Share in GDP

30
70%
25 23.24 60%
GDP (1950-79) = 8.70%
20 50% Medium Tech
17.62
40%
15
30%
10 7.78 20% Low Tech
5 3.34 10%
1.69
0%
0
1953 1960 1970 1980 1990 2000 2010 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010

Source: Elaborated by the author, using data from The Groningen Growth and Development Centre and ECLAC
and UNIDO.
Note: Manufacturing Gross value added at constant 2005 national prices (in millions). Classification of
manufacturing industries by technology group based on UNIDO (2013). For details, see appendix 2.3.

84
 As shown by Figure 2.5, along the period of 1980 to 2010, the country showed an
average rate of economic growth of 6.33% and the manufacturing sector grew on average at
2.37%. Despite the noticeable slow down in manufacturing growth rate, the manufacturing
sector sharply expanded reaching 35.24% of the GDP (in 1980 this number was 17.62%). The
manufacturing sector as an engine of economic growth was also a remarkable characteristic of
other Asian economies. From 1980 to 2011, the “first and second-tier” NICs adopted this
strategy of economic growth as a common denominator. For instance, in the first-tier NICs,
the manufacturing sector increased from 22.52% to 23.10% of the GDP. In a similar trend, the
‘second-tier’ NICs grew from 22.62% in 1980 to 29.14% in 2010109.

7.3.Labour productivity

From a Kaldorian perspective there is a positive relationship between the rate of

growth of output in manufacturing and the rate of growth of labour productivity in the
economy. Therefore, the graph bellow (Figure 2.6) illustrates that from 1950 to 1980 Brazil
experienced a productivity growth similar to South Korea. Moreover, the data reveals that
from 1950 to 1976, Brazil’s labour productivity level was higher than South Korea’s.
However, after 1980 the trajectories of these two developing economies diverged. On the one
hand, South Korea (despite 1997) maintaining a strong dynamism reducing its annual growth
rate productivity after 1980 only 0.80 %, i.e. a declining from an average of 5.19% (1950-
1979) to 4.39% (1980-2010). On the other hand, Brazil’s average annual productivity growth
rate collapsed to the point of becoming negative until 1993. From 1980 to 2010, despite the
widely reported commodities prices boom from 2002 to 2008, the Brazilian labour
productivity growth rate was more than eight times lower than the South Korean one. During
the same period, Brazil’s average productivity level was more than twice lower than those of
South Korea’s, i.e. US$ 11,760 and US$ 26,997 respectively110.

109
In East Asian countries, the manufacture share in GDP grew at much higher rates than in Brazil (or any other
Latin American country) resulting in significant increase in the share of global GDP. More interesting, perhaps,
is the fact that in most East Asian economies the share of manufacturing sector in the GDP did not suffer the
Latin American slowdown or the reduction like in the developing countries. Furthermore, headed specially by
China, India and ASEAN-4 - Malaysia, Thailand, Indonesia, and the Philippines – the Asian axis more than
doubled its share in the global GDP. The gains were not higher because controversially China has become the
most important participant in many essential markets of these countries and therefore made it harder for them to
abandon the industrialisation pattern based on “sub-contracting” and move forward in the direction of industrial
development common to NICs – Korea, Taiwan, Hong Kong, and Singapore (Palma, 2005).
110
Labour productivity per person employed in 1990 US$ (converted at Geary Khamis PPPs).

85
 In terms of productivity “catching-up” with the US, the graph also shows clearly the
dichotomy between Brazil and South Korea. Until 1980 Brazil was ahead of South Korea in
the process of “catching-up”. Indeed, from 1950 to 1980 Brazil engendered a strong process
of productivity growth where labour productivity as share of the US labour productivity
increased by 11.66 percentage points. In other words, from 18.99% in 1950 this number
jumped to 30.65% in 1980. In a similar trend, for the same period, South Korean productivity
grew from 11.86% to 27.52%. However, this productivity trend was drastically reversed after
1980. Thus, while South Korea maintained an upward trend reducing its productivity gap with
the US very rapidly from 27.52% to 65.07% in 2010, Brazil fell behind with productivity
level relative to the US declining to 19.56% in 2010. Additionally, it is important to highlight
that in terms of regions the South Korean and Brazilian trends were followed similarly by the
“first-tier” NICs and Latin America respectively both before as well as after 1980.

Figure 2.6 – Labour Productivity and ‘Catching-up’ with the US, 1960-2010

Labour Productivity ‘Catching-up’ with the US

50,000 80%
NICs 1
South
45,000 70%
Korea
40,000
60%
Labour Productivity as Share of US LP

SK LP (1980
Labour Productivity US$ ( PPP - 1990)

35,000 South
- 2010) =
4.39% 50%
Korea
30,000

25,000 40%

20,000 SK LP (1950 LA
- 1979) = 30%
5.19% Brazil
15,000 Brazil
20%
10,000
NICs 2
BR LP (1980
BR LP (1950 - 2010) = 10%
5,000
- 1979) = 0.52%
3.32%
0 0%
1950 1960 1970 1980 1990 2000 2010 1950 1960 1970 1980 1990 2000 2010

Source: Elaborated by the author adapted from Palma (2011). Data from The Groningen Growth and
Development Centre.
Note: Labour productivity per person employed in 1990 US$ (converted at Geary Khamis PPPs).

86
 7.4.Manufacturing Share and Income ‘Catching-up’

The Figure below illustrates empirically the dichotomy between Brazil and South
Korea in terms of sectoral composition and economic development. During the whole period
(1953-2010), South Korea gradually increased the share of manufacturing in the GDP and
reduced the income gap with the US. Therefore, from 1953 to 2010, the manufacturing share
in GDP relative to US grew from 0.13 to 3.21111. Moreover, except for some periods such as
1995-2000 when the country was affected by the Asian Crisis (1997), the income per capita
showed a constant increase in relative terms to the US. From 8.93% of the US income per
capita in 1953, in 1980 and 2010 this variable reached 22.15% and 70.25% respectively.

Figure 2.7 - Share of manufacturing in GDP and relative income gap with the US, 1950-
2010

South Korea: Progressive Trajectory Brazil: Regressive Trajectory

40 25

2010 24
35

23
05
30
00 60 75
22
% Manuf. South Korea in GDP

% Manuf. Brazil in GDP

70
25 80
95 21
90 65
20 20
85 05
00
80 19 85
15 55 95
50 2010
75 18
90
10
17
70
5 65 16
60
53 55
0 15
0% 20% 40% 60% 80% 0% 10% 20% 30% 40%
Gap GDP Per Capitawith the US Gap GDP Per Capita with the US

Source: Elaborated by the author, using data from The Groningen Growth and Development Centre.
Note: GDP per capita in 1990 US$ (converted at Geary Khamis PPPs) and Manufacturing Gross value added at
constant 2005 national prices (in millions).

111 ?@)3A B C)DE)FE3GDH GD IJK (LBED3MN)

Manufacturing share in GDP relative to US is calculated as:
?@)3A B C)DE)FE3GDH GD IJK (O?)

87
 In contrast to South Korea which moved ahead even after the advent of neoliberal
reforms, Brazil drastically fell behind (in a cyclical fashion) in terms of manufacturing share
in the GDP and income gap reduction with the US. From 1950 to 1980, the Brazilian
economy showed a high volatility in terms of degree of industrialisation and income gap
reduction with the US, but with a substantial improvement of these indicators. In 1950, the
Brazilian income per capita (in PPP) was 17.48% of the US income, and manufacturing
corresponded to 18.37% of GDP. In 1980, the Brazilian income increased to 27.96% of the
US income, and, during the same period, the manufacturing sector reached 21.13% of GDP.
However, after 1980, this overall trend changed dramatically. As showed by the Figure above,
the Brazilian economy embarked on a regressive trajectory both in terms of manufacturing
expansion in GDP and income gap reduction with the US economy. The most dramatic period
was during 1980 to 1995 when the Brazilian economy regressed 15 years both in terms of
income per capita catching-up. From 2005 to 2010, this relative loss was reversed but mostly
by a boom in commodities, finance, and household credit112. Taking into account the
“Kaldorian-Structuralist” perspective, it is not by chance that the persistent difficulty of
overcoming the middle-income barrier appears to be closely related to the manufacturing
sector loss in the GDP.

7.5.Export-trajectories: Two historical distinct patterns

From 1950 to 1980, Brazil and South Korea had the rise of manufacturing in GDP as a
common denominator in the strategy of economic growth. In South Korea this pattern of
development was followed by a strong increase of exports in GDP. From the 1960s until
1987, both the share of manufacturing and exports in GDP grew practically constantly in a
convergent pattern (see Figure 2.8). This trend was also followed by a significant increase of
manufacturing exports in GDP that represented more than 75% of total exports in the GDP113.
By contrast, Brazil pursued a development path based on its large domestic market.

112
It is important to emphasise that the commodity prices boom started in 2002, while the finance and household
credit boomed from 2003 onwards, during the Lula’s government. See, for instance, Gottschalk & Prates (2006).
113
It is important to emphasise that the outward-orientation was a common trend in economic policy principally
in East Asian NICs. Therefore, the development mainspring underpinned over the outward orientation
coordinated by the State which directed the necessary private investment to sustain the export driver in order to
generate foreign exchange via manufactured exports. During this initial phase of export expansion, the rapid
growth of the East Asian NICs – Korea, Taiwan, Hong Kong, and Singapore – was founded on light, labour-
intensive industries like textiles, garments and electronic equipment. In subsequent phases, however, the East
Asian NICs achieved success in much heavier industries like steel, petrochemicals, shipbuilding, vehicle
manufacture and computers that were less well suited to their original factor endowments (Gereffi, 1989).

88
Therefore, different to Korea, Brazil followed a resilient performance of the GDP export-
share. In numbers, while South Korea achieved a share of 23.13% of manufacturing exports in
the GDP in 1980, Brazil by contrast had its exports of manufactured goods during the whole
developmental era under 5% of GDP. Curiously, due to the inward-orientation, the constant
increase of the manufacturing sector performed inversely to exports share in GDP. In other
words, despite a strong growth performance in Brazil, the share of manufacturing in GDP
took place at the expense of the GDP export-share, i.e. in a divergent pattern (see Figure
2.8)114.

Figure 2.8 - Share of Manufacturing and Exports in GDP (%) (1960-2010)

South Korea: Convergent Pattern Brazil: Divergent Pattern

60 25

50
20

40
15
30
10
20

5
10

0 0
1960 1970 1980 1990 2000 2010 1950 1960 1970 1980 1990 2000 2010

Export/GDP Manufacturing Export/GDP Manufacturing

Manuf. Exp/ GDP Manuf. Exp/ GDP

Source: Elaborated by the author, using data from The Bank of Korea (BOK), World Bank, The Conference
Board and Groningen Growth and Development Centre.
Note: Export/GDP = Gross exports/GDP and Manufacturing = manufacturing value added/GDP. Data are
in constant 2005 US dollars.

In terms of exports, despite the rapid process of industrialisation, Brazil did not
overcome the historical dependence of its Ricardian comparative advantages, i.e. natural
resource endowments. Over the years, Brazilian agriculture and mining sectors were not only
important to supply the domestic market but also a central source of foreign exchange. In
theoretical terms, it is important to highlight that Raúl Prebisch and his intellectual heirs at the
ECLAC had during the 1950s and 1960s an economic thinking characterised by a deep
114
Palma (2005) identifies this phenomenon not only as a strategy characterised by a strong anti-export bias, but
also as a consequence of weak demand for Latin American primary commodities in OECD markets since the
Korean War.

89
criticism regarding the growth-enhancing capacity of unprocessed primary commodity
exports in “peripheral” countries. However, this strong developmental thinking – which was
very influential in Latin America at that time – propelled particularly the diversification of the
productive structure rather also exports. Additionally to a negligent export policy, an
important external factor also affected this trade pattern. The massive inflow of foreign capital
contributed to finance the current deficit and consequently eased the pressure to increase the
share of manufacturing products in exports. During the 1970s the Brazilian government tried
to work around the situation through stimulus to manufactured exports, however, this in turn
was highly restricted due the turbulent international environment of the decade. In short, due
internal and external factors, the Brazilian economic policy was substantially ineffective in
increasing exports of manufactured goods when compared to South Korea (and other East
Asian countries)115.

South Korea, in turn, instead of accepting its traditional Ricardian comparative

advantages (essentially associated with cheap labour), sought to manage the industrial policy
to create an institutional environment that allowed the country to follow the Japanese example
of industrialisation and develop a pattern of production and export upgrading over the
decades116. As pointed out by Palma (2005), the extraordinary success of the East Asian
economies was based on external and internal factors. On the external front, following a
“flying geese” paradigm led by Japan, South Korea increased the exports to the OECD
markets, especially the USA, for manufactured goods within a process of productive
regionalisation. On the internal one, South Korea built a structure around a skilled labour
force, property rights, political settlement and institutional capability that allowed the country
to produce competitive manufactured goods. In Kaldorian terms, this dynamic based on the
production and exports of manufacturing products imbued special growth-enhancing
properties that triggered a process of cumulative causation not shared by other sectors and
consequently strengthened the foundations for sustained growth. Not by chance, from the
1950s to 1970s, the South Korean (and East Asian) economy climbed the ladder to establish
the pillars for long-term economic growth.

115
In ECLAC’s economic thinking, a key structural economic characteristic of peripheral economies refers to the
deterioration in their terms of trade over time due different income – elasticity of demand – as “dynamic
disparity of demand”. Therefore, according to ECLAC’s point of view, ISI model was a central measure for
diversification of the productive industry and expansion of exports based on a dynamic manufacturing sector.
For a detailed explanation of the Latin American Structuralism approach, see the first chapter of this Thesis.
116
About the Japanese leadership on the Asian pattern of development see Arrighi et al. (1993), Akyus (1996),
UNCTAD (1996), Ozawa (1991, 1995, 2001) and Palma (2005).

90
 Nevertheless, the South Korean economy also faced turbulences in trade. In the late
1980s to early 1990s, some of its most important manufactured products, particularly
microelectronics, began to experience a persistent drop in prices. It was essentially a result of
rapid standardisation (or “commoditisation”) and an increased supply of its manufactured
products in the global market. In an attempt to reverse this situation the corporate sector
sought to expand the market share, increasing investment in new productive capacity. This led
to overinvestment, which in turn exacerbated the global excess supply in international markets
and further downward pressure on prices117. The declining manufacturing profitability
resulted in a strong change in the finance structure of investment118. From an industrial
structure characterised by the retention of profits, the economy shifted to a situation
characterised by domestic and foreign debt. Therefore, this process caused an increase in the
microelectronics sectoral deficit that rose from around 5% in 1987 to almost 20% of GNP in
1996 (Palma, 2013). In late 1997, South Korea experienced a severe economic downturn
derived from the so-called Asian Crisis. However, despite important ideological changes that
engendered neoliberal reforms, the state maintained an active role, either through long-term
policies for the development of high-tech sectors and exports, or by ensuring important long-
term funding sources through public banks.

Brazil, by contrast, after the ‘lost decade’ embarked in the neoliberal agenda under the
illusion that the market forces per se would create the necessary conditions to the
consolidation of the productive structure and exports promotion. As proved by the facts,
neither the manufacturing share expanded in the GDP nor exports of manufactured products
increased. Moreover, as a result of an extremely weakened production structure resulting from
over a decade of strongly market-oriented policies, production and exports of primary
products became the engine of Brazil’s growth. In 2002, demand for Brazilian commodities
began to rise more intensely because of the growth of the Asian economies, particularly that

117
Palma (2013) suggests that the collapse of exports in microelectronics was significantly explained by the
Taiwanese memory-chips exports that increased massively during the period. For instance, due the excess of
supply in the global market, the price of memory chips fell approximately 80% in 1996 alone. In this process,
South Korea was strongly affected since a significant amount of its exports consisted in this type of product.
118
Chang et al. (1998), also points out that the political and economic shift of the period contributed to worsen
the profitability of the South Korean microelectronics sector. The rise of the neoliberal agenda weakened
domestic sectoral policies and affected the transparency between the State and corporations. The manufacturing
sector became increasingly exposed to state corruption and vested interests. As argued by Chang “ large sums of
money did flow from big business to politicians and top bureaucrats (…) there was a fundamental transformation
in the state–business relationship in Korea, which meant that the major manufacturing sectors became less
insulated from corrupt political exchanges than they had been previously” (Chang et al. 1998, p. 741).

91
of China119. This demand for natural resources and raw materials has pushed up commodities
prices in the world market, benefiting the Brazilian economy. The boom of commodity prices
pulled the Brazilian exports, adding substantial dynamism to the economy but without a
change in the industrial structure to expand the manufacturing sector and increase the export
of manufactured products. Without developmental policies, the current export pattern has
showed little capacity to generate growth-sustaining processes of cumulative causation.
Therefore, it is not by chance that the Brazilian economy has been over decades limited to a
middle-income level. The current productive structure limited the growth-enhancing
characteristics of the manufacturing sector, such as externalities and spillover effects that may
help set in motion processes of cumulative causation that take advantage of dynamic
economies of scale, increasing returns, etc (Palma, 2012).

Furthermore, this productivity and trade relations resemble in some ways the centre-
periphery relationship outlined by Prebisch, where Brazil (as well as Latin American
countries as a whole – with the exception of Mexico and the Central American countries)
exports commodities and raw materials to China, and imports manufactured goods from Asian
countries. In this way, Latin America has assumed the role of the periphery, exporting natural
resources and raw materials, while Asian Countries have gradually assumed the core or
central position, exporting manufactured goods to the region. Brazil’s development has
become strongly dependent on China’s growth and development. Although these structural
bottlenecks had been widely discussed in the Latin America structuralist theory and in the
Kaldorian approach, their contributions seemed to have relatively low influence on economic
policy after the neoliberal rise.

8. Concluding remarks

Taking into account the role of the state in guiding the development path, this chapter
sought to compare the industrial development in Brazil and South Korea and identify factors
(endogenous and exogenous) that determined different routes of economic dynamism in these
economies. Therefore, the chapter showed that a successful process of industrialisation

119
Between 1990s and 2000s, the Chinese economy practically doubled its number both in the GDP growth and
productivity, highlighting its dynamism over the rest of the world. A similar trend was followed by India and
SEAN-4 which, together with NICs, showed their capabilities to expand and conquer other markets. In both
groups, their performance in terms of average of growth rates presented a plateau in 1990s and 2000s with a
peak in 2010.

92
required an active developmental state to expand the manufacturing sector. Both in Brazil and
South Korea during the industrialisation era, policies implemented involved controls over
interest rates, credit allocation, and a set of tax and tariff measures to provide protection to the
nascent industry. Moreover, particularly in South Korea, policies for technology acquisition,
incentives to activities of R&D and the promotion of cartels for specific purposes such as
standardisation, specialisation and exports were also a central characteristic.

Despite general similarities, while in South Korea the domestic private capital,
represented by the Chaebols, played a decisive role in the evolution and development of
Korean industry, while in Brazil economic growth and investments were driven by state-
owned enterprises and multinational companies established in the country during the 1950s
and 1960s. Moreover, while South Korea pursued a process of industrialisation with a clear
outward orientation, Brazil, due its large domestic market and restricted access to OECD
markets, followed a pattern of industrial growth with an inward orientation. Moreover, despite
different trends regarding the interaction with the external market, in terms of productive
structure both Brazil and South Korea achieved the phase of heavy industrialisation in the late
1970s.

Conversely, this massive effort to climb the ladder of world production in a turbulent
economic environment profoundly affected public accounts. In both economies, although in
different magnitudes, given the lack of foreign exchange, the military government gradually
accumulated external debts with international private banks to finance the current account
deficit and maintain the booming economic growth. This excessive external financing raised
economic vulnerability and exposed these countries to external crises. Thus, the two oil price
shocks in 1973 and 1979, and the Volcker interest rate hikes in 1979 drastically affected both
countries. With the outbreak of the debt crisis in 1982, the Brazilian economy, which was
vulnerable due to its large current account deficit and external debt, faced an abrupt cut-off in
bank financing that plunged the country into a serious crisis and caused a deep disarticulation
of the state in providing active policies for national industry. Consequently, during the 1980s,
economic growth was seriously retarded giving rise to the commonly used term “the lost
decade”. In contrast, South Korea dealt with the crisis in a much less traumatic way since its
strategy of industrialisation was relatively less dependent on foreign debt and Japanese banks
“recycled” the country’s liabilities. Therefore, South Korea not only recovered rapidly from
the debt crisis, but also strengthened the manufacturing sector and the production of cutting-
edge technological products. By contrast, Brazil struggled in the debt crisis of the 1980s that

93
did not allow the country to keep up with new technologies and consequently the
manufacturing industry became delayed compared to Asian latecomer economies.

From the 1980s, Brazil embarked on a “fundamentalist” neoliberal agenda which

sought to reverse almost every aspect of the growth strategy previously followed by the
country. The neoliberal policies were strongly marked by an extreme move towards trade and
financial liberalisation, market deregulation, attraction of new flows of FDI, privatisation of
state-owned enterprises and the withdrawal of selective industrial policies. In contrast, the
neoliberal polices implemented in South Koreas were remarkably different from the Brazilian
neoliberal fundamentalist agenda. In South Korea, the economic reforms were much more
pragmatic, seeking to transform the initial catching-up efforts into a dynamic and sustainable
process of capital expansion and development. Even after the 1980s, the South Korean
experience was characterised by the maintenance of a strong state control over the
socioeconomic transformations and by the pragmatism in the management of intermediate and
instrumental objectives such as the economic internationalisation and administration of the
fiscal and monetary policy.

In Brazil, the main result of the “fundamentalist” process of economic reforms was to
reinforce the rentier instincts of the country’s capitalist élites and undermine any perspective
of sustained economic growth. Therefore, the neoliberal agenda established an economic
dynamic of successive “stop-and-go” cycles derived from a fragile productive structure and
lack of effective demand. From the 1980s onwards, the country’s growth rate collapsed both
in terms of GDP and manufacturing value added growth. Moreover from an intra-industry
perspective, the composition of the manufacturing sector stagnated over the decades. In
contrast, South Korea presented a distinct development path. Thus, after the neoliberal
reforms, the country did not experienced such dramatic loss in terms of economic dynamism.
The economy grew on average by more than 6% per year and the manufacturing sector
expanded sharply. Furthermore, in terms of intra-industry technology composition, the share
of high-tech industries rose significantly, reaching around 61% of the manufacturing share.

The deleterious effects of premature de-industrialisation in Brazil vis-à-vis South

Korea’s thriving manufacturing expansion clearly impacted the evolution of labour
productivity and catching-up. Analysing the data there is little doubt that the remarkable
neglect of manufacturing since economic reforms lies at the heart of country’s productivity
stagnation. In contrast to the period from 1950 to 1980 when in both countries manufacturing

94
and productivity expanded considerably, from 1980 to 2010, due to the manufacturing
collapse, Brazilian productivity practically stagnated. Even during the commodities prices
boom from 2002 to 2008, the Brazilian economy productivity growth was extremely poor.
Brazilian average productivity growth was more than eight times lower than that of South
Korean. Moreover, during the same period, Brazil’s average productivity level was more than
two times lower than those of South Korea’s. Thus, while South Korea maintained an upward
trend reducing its productivity gap with the US very rapidly, Brazil embarked on a noxious
trajectory.

In terms of income catching-up, while South Korea moved ahead after the advent of
neoliberal reforms, Brazil drastically fell behind (in a cyclical fashion) in terms of
manufacturing share in the GDP and income gap reduction with the US. The most dramatic
period was during 1980 to 1995 when the Brazilian economy regressed 15 years both in terms
of income per capita catching-up. Taking into account the “Kaldorian-Structuralist” approach,
it is not by chance, that the persistent difficulty to overcome the middle-income trap appears
to be closely related to the manufacturing sector loss in the GDP. Analysing the exports
trajectories the dichotomy between these economies are also very clear. On the one hand,
South Korea pursued an outward strategy of economic growth expanding exports of
manufacturing products, particularly high-tech products, taking into account the special
growth-enhancing properties of this sector. In contrast, Brazil was always historically
dependent on its Ricardian comparative advantages, i.e. natural resource endowments. During
the developmental era, all efforts made to overcome this structural dependence were very
resilient and after the neoliberal rise practically inexistent. Therefore, in Kaldorian-
Structuralist terms, the Brazilian trade composition based on exports of primary products
showed little capacity to generate growth-sustaining processes of cumulative causation and
consequently mitigated any possibility of dynamic economic growth.

95
Appendix 2.1 - Brazil’s Macroeconomic Indicators, 1950-2013 (%)
Manuf. GFCF Sectoral Composition
GDP Export/ Import/ GFCF/ External
Year Exp/ Primary Manuf.
Growth GDP GDP GDP Debt/GDP Public Private Manuf. Construction Services
GDP Sector Growth
1950 6.80 9.20 7.60 - 12.79 - 34.17 65.83 14.05 18.37 5.82 61.76 -
1951 4.90 9.60 11.30 - 15.45 - 23.60 76.40 13.35 18.10 5.70 62.85 5.59
1952 7.30 7.10 9.90 - 14.82 - 23.33 76.67 13.19 17.41 4.92 64.48 5.38
1953 4.70 6.60 5.60 - 15.06 - 22.88 77.12 12.78 18.32 4.57 64.32 9.33
1954 7.80 6.70 6.80 - 15.76 - 24.95 75.05 12.36 18.08 4.63 64.92 9.48
1955 8.80 7.60 6.80 - 13.49 - 24.26 75.74 12.34 18.60 4.92 64.14 10.93
1956 2.90 6.80 5.80 - 14.46 - 20.89 79.11 11.71 18.92 5.36 64.01 5.53
1957 7.70 5.60 6.20 - 15.04 - 30.75 69.25 11.96 18.74 5.58 63.71 5.58
1958 10.80 5.70 6.10 - 16.98 - 32.71 67.29 11.51 20.43 5.80 62.26 16.71
1959 9.80 5.90 6.60 - 17.99 - 28.64 71.36 11.71 21.98 5.92 60.38 12.84
1960 9.40 5.30 6.40 - 15.72 - 35.89 64.11 11.24 22.03 5.86 60.88 10.60
1961 8.60 5.80 6.20 - 13.11 - 44.20 55.80 10.89 22.08 5.64 61.38 11.10
1962 6.60 6.70 8.00 0.19 15.51 17.69 39.27 60.73 10.85 22.64 5.95 60.56 8.19
1963 0.60 8.60 9.00 0.18 17.04 15.04 31.68 68.32 10.79 21.86 5.92 61.42 -0.49
1964 3.40 6.50 5.60 0.36 14.99 15.20 33.69 66.31 10.68 22.25 6.14 60.93 5.29
1965 2.40 7.60 5.40 0.57 14.71 16.79 46.18 53.82 11.79 20.41 6.13 61.67 -4.70
1966 6.70 6.50 5.80 0.46 15.92 13.21 38.03 61.97 9.75 21.25 6.24 62.75 11.67
1967 4.20 5.70 5.80 0.53 16.20 11.00 40.74 59.26 9.87 20.32 6.17 63.64 2.18
1968 9.80 6.00 6.70 0.45 18.68 11.99 33.97 66.03 9.53 21.13 6.30 63.04 14.20
1969 9.50 6.70 6.70 0.60 19.11 12.40 38.04 61.96 9.08 21.32 6.28 63.32 11.19
1970 10.40 7.00 7.40 0.86 18.83 14.66 38.59 61.41 8.61 21.83 6.39 63.17 11.86
(Continue)

96
 Manuf. GFCF Sectoral Composition
GDP Export/ Import/ GFCF/ External
Year Exp/ Primary Manuf.
Growth GDP GDP GDP Debt/GDP Public Private Manuf. Construction Services
GDP Sector Growth
1971 11.34 6.50 8.20 0.90 19.91 16.85 32.86 67.14 8.45 21.97 6.46 63.12 11.87
1972 11.94 7.30 8.90 1.28 20.33 19.51 37.18 62.82 7.85 22.45 6.48 63.23 13.96
1973 13.97 7.80 9.00 1.53 20.37 17.67 28.43 71.57 7.02 23.00 6.52 63.47 16.62
1974 8.15 7.70 13.30 1.83 21.85 18.15 35.68 64.32 6.85 22.95 6.77 63.42 7.74
1975 5.17 7.20 11.00 1.77 23.33 19.34 36.06 63.94 6.77 22.25 6.98 64.00 3.81
1976 10.26 7.00 9.40 1.53 22.42 20.88 47.17 52.83 6.27 22.52 7.20 64.02 12.12
1977 4.93 7.20 7.90 1.73 21.35 21.41 44.50 55.50 6.44 21.77 7.68 64.11 2.28
1978 4.97 6.70 7.90 2.10 22.27 25.94 37.94 62.06 5.85 21.14 7.83 65.18 6.11
1979 6.76 7.20 9.30 2.55 23.36 24.97 29.68 70.32 5.68 20.66 8.06 65.60 6.86
1980 9.23 9.00 11.20 3.19 23.56 27.03 28.01 71.99 5.87 21.13 8.35 64.65 9.11
1981 -4.25 9.60 10.00 3.46 24.31 28.61 30.14 69.86 6.40 19.51 8.08 66.01 -10.38
1982 0.83 7.90 8.60 2.74 22.99 31.52 30.48 69.52 6.48 19.48 7.89 66.15 -0.18
1983 -2.93 12.20 9.70 4.24 19.93 49.48 30.32 69.68 6.85 18.82 6.92 67.41 -5.85
1984 5.40 15.00 8.80 5.34 18.90 53.82 27.12 72.88 7.10 18.93 6.48 67.49 6.16
1985 7.85 12.90 7.50 5.03 18.01 49.82 29.29 70.71 7.27 19.07 6.39 67.27 8.34
1986 7.49 9.20 6.60 4.00 20.01 43.13 27.65 72.35 6.46 19.67 6.98 66.89 11.30
1987 3.53 9.80 6.40 4.42 23.17 42.92 27.20 72.80 6.93 19.47 6.89 66.70 0.95
1988 -0.06 11.70 6.10 5.32 24.32 37.13 26.62 73.38 6.97 18.79 6.67 67.57 -3.41
1989 3.16 8.90 5.50 4.35 26.86 27.77 20.62 79.38 7.03 18.88 6.72 67.38 2.88
1990 -4.35 8.30 8.30 3.53 20.66 26.30 25.23 74.77 7.22 17.89 6.35 68.54 -9.46
1991 1.03 8.30 8.30 4.26 18.11 30.54 24.01 75.99 7.37 17.45 6.12 69.07 -2.36
(Continue)

97
 Manuf. GFCF Sectoral Composition
GDP Export/ Import/ GFCF/ External
Year Exp/ Primary Manuf.
Growth GDP GDP GDP Debt/GDP Public Private Manuf. Construction Services
GDP Sector Growth
1992 -0.54 10.90 8.40 5.22 18.42 35.10 27.02 72.98 7.42 17.47 6.04 69.08 -0.56
1993 4.92 10.50 9.10 5.18 19.28 33.91 24.30 75.70 7.78 17.15 5.81 69.26 -3.65
1994 5.85 9.50 9.20 4.39 20.75 27.31 21.61 78.39 7.52 17.92 5.85 68.71 8.29
1995 4.22 7.30 8.80 3.24 20.54 20.67 18.53 81.47 7.68 18.52 6.05 67.75 6.83
1996 4.09 6.60 8.40 3.06 18.65 21.07 19.98 80.02 7.82 18.29 6.19 67.70 -0.32
1997 3.39 6.80 9.00 3.26 19.13 22.56 18.33 81.67 7.70 18.58 6.48 67.24 5.25
1998 0.36 6.90 8.90 3.31 18.55 27.87 20.95 79.05 7.92 18.51 6.52 67.05 0.14
1999 0.49 9.40 10.80 4.42 17.01 40.12 14.24 85.76 8.12 18.46 6.27 67.14 0.60
2000 4.38 10.00 11.70 5.00 18.33 35.92 13.88 86.12 8.09 19.72 6.10 66.09 12.03
2001 1.28 12.20 13.50 5.72 18.44 40.38 15.22 84.78 8.36 19.50 5.88 66.26 0.41
2002 3.08 14.10 12.60 6.31 17.96 44.74 18.34 81.66 8.78 19.31 5.60 66.32 1.88
2003 1.22 15.00 12.10 6.87 16.67 42.03 15.32 84.68 9.22 19.36 5.39 66.03 0.63
2004 5.66 16.40 12.50 7.77 17.38 32.88 14.94 85.06 8.98 20.25 5.44 65.32 10.45
2005 3.15 15.10 11.50 7.12 17.22 21.06 15.09 84.91 8.98 19.88 5.38 65.76 1.03
2006 4.00 14.40 11.50 6.43 17.26 18.01 16.93 83.07 9.06 19.44 5.43 66.08 1.50
2007 6.01 13.40 11.80 5.62 18.07 17.23 15.61 84.39 8.94 19.48 5.38 66.21 6.03
2008 5.02 13.70 13.50 5.37 19.49 15.54 18.14 81.86 9.06 18.84 5.58 66.51 0.57
2009 -0.23 11.00 11.10 3.73 19.21 16.62 21.05 78.95 8.82 17.49 5.56 68.13 -7.51
2010 7.57 10.87 11.90 3.49 20.59 15.92 22.24 77.76 8.92 17.96 5.79 67.33 10.10
2011 3.92 11.89 12.62 3.53 20.64 15.46 18.50 81.50 9.03 17.55 5.86 67.56 0.10
2012 1.76 12.59 14.03 3.78 20.21 18.27 20.22 79.78 - - - - -
2013 2.74 12.55 15.04 3.92 20.53 20.32 20.75 79.25 - - - - -
Source: Elaborated by the author, using data from The Brazilian Central Bank, Brazilian Institute of Geography and Statistics (IBGE), Ipeadata, World Bank – World
Development Indicators, The Conference Board and Groningen Growth and Development Centre and UNCTAD.
Note: Data for GDP Growth and Gross Fixed Capital Formation (GFCF) in current Prices. Sectoral composition calculated from data at constant 2005 national prices (in
millions).
98
Appendix 2.2 - South Korea’s Macroeconomic Indicators, 1950-2013 (%)
Manuf. GFCF Sectoral Composition
GDP Export/ Import/ GFCF/ External
Year Exp/ Primary Manuf.
Growth GDP GDP GDP Debt/GDP Public Private Manuf. Construction Services
GDP Sector Growth

1950 - - - - - - - - - - - - -
1951 - - - - - - - - - - - - -
1952 - - - - - - - - - - - -
1953 - - - 14.70 - - - 49.07 8.41 1.78 40.74 -
1954 7.20 - - - 11.00 - - - 41.69 10.84 2.17 45.30 16.01
1955 5.80 - - - 11.70 - - - 46.88 10.65 2.52 39.95 18.48
1956 0.70 - - - 8.30 - - - 49.15 10.75 2.41 37.70 15.44
1957 9.20 - - - 14.30 - - - 46.39 10.24 2.84 40.54 3.69
1958 6.50 - - - 11.50 - - - 42.22 11.51 2.59 43.68 5.68
1959 5.40 - - - 10.10 - - - 35.85 12.76 2.75 48.64 7.80
1960 2.30 3 12.60 - 9.70 - - - 38.17 12.30 2.63 46.90 5.04
1961 6.90 5 14.94 - 11.70 3.52 - - 41.09 12.09 2.56 44.26 1.01
1962 3.80 5 16.50 0.37 13.10 3.24 - - 37.90 13.17 2.65 46.29 8.29
1963 9.20 5 15.70 1.01 17.80 4.06 - - 43.66 13.84 2.43 40.07 9.62
1964 9.50 6 13.35 1.64 14.70 5.27 - - 48.11 14.91 2.37 34.61 0.64
1965 7.20 8 15.92 3.40 14.00 6.83 - - 41.20 16.95 2.84 39.01 13.40
1966 12.00 10 20.13 3.99 20.60 10.30 - - 37.51 17.39 3.02 42.07 7.13
1967 9.10 11 21.92 4.55 21.10 13.72 - - 33.49 17.64 3.19 45.68 14.90
1968 13.20 12 25.30 5.67 25.90 20.13 - - 29.72 18.96 3.99 47.33 15.32
1969 14.50 13 25.15 6.34 28.80 24.08 - - 30.05 19.35 4.75 45.85 8.85
1970 10.00 13 22.53 6.80 25.70 23.86 23.73 76.27 29.36 20.19 4.73 45.72 10.24
(Continue)

99
 Manuf. GFCF Sectoral Composition
GDP Export/ Import/ GFCF/ External
Year Exp/ Primary Manuf.
Growth GDP GDP GDP Debt/GDP Public Private Manuf. Construction Services
GDP Sector Growth
1971 10.50 14 24.19 8.36 25.20 28.05 24.26 75.74 29.64 20.10 4.07 46.18 9.77
1972 7.20 18 22.85 11.95 21.20 31.57 22.60 77.40 28.90 21.20 3.75 46.15 9.78
1973 14.80 27 29.99 18.61 25.70 29.28 16.94 83.06 27.25 23.85 4.05 44.85 17.25
1974 9.50 25 35.67 18.49 32.30 29.05 15.51 84.49 27.27 24.68 4.15 43.90 9.92
1975 7.90 25 33.33 17.65 29.10 37.03 19.14 80.86 27.64 24.90 4.36 43.11 6.64
1976 13.10 28 30.17 21.51 27.50 33.55 19.26 80.74 26.07 26.57 4.21 43.15 12.67
1977 12.30 29 29.41 21.17 30.60 31.42 18.55 81.45 25.44 26.27 5.17 43.12 5.08
1978 10.80 27 30.29 20.68 34.20 27.31 18.53 81.47 23.75 26.62 7.24 42.39 11.64
1979 8.60 25 31.64 19.18 38.00 29.09 18.20 81.80 22.09 27.62 7.96 42.32 3.04
1980 -1.70 30 37.63 23.13 34.60 40.07 21.66 78.34 17.75 28.27 7.82 46.17 0.41
1981 7.20 32 37.19 25.11 32.70 42.54 24.32 75.68 18.57 28.42 6.75 46.26 3.39
1982 8.30 31 33.38 24.40 32.30 45.40 19.93 80.07 17.27 28.07 7.26 47.40 -1.45
1983 13.20 31 31.81 24.55 32.90 44.60 19.19 80.81 15.90 29.19 7.61 47.31 3.83
1984 10.40 31 31.36 26.68 32.50 43.07 18.32 81.68 15.03 29.85 7.42 47.69 7.91
1985 7.70 30 29.23 26.66 33.00 45.08 19.39 80.61 14.94 29.17 7.29 48.60 -0.49
1986 11.20 33 28.42 26.64 32.80 37.16 17.88 82.12 13.47 30.69 6.70 49.14 7.98
1987 12.50 35 28.93 28.92 33.10 32.18 16.18 83.82 12.21 31.51 7.01 49.27 7.56
1988 11.90 34 27.13 27.92 34.60 21.68 16.06 83.94 12.07 32.17 7.34 48.43 2.21
1989 7.00 29 26.62 23.30 37.10 17.08 15.08 84.92 11.20 30.76 8.57 49.48 -2.56
1990 9.80 26 26.89 21.35 39.50 16.43 14.15 85.85 10.17 29.34 11.20 49.29 0.36
1991 10.40 24 26.88 20.08 41.40 16.40 15.39 84.61 9.22 29.61 12.40 48.77 1.17
(Continue)

100
 Manuf. GFCF Sectoral Composition
GDP Export/ Import/ GFCF/ External
Year Exp/ Primary Manuf.
Growth GDP GDP GDP Debt/GDP Public Private Manuf. Construction Services
GDP Sector Growth
1992 6.20 25 25.70 19.98 38.50 16.57 16.50 83.50 8.79 29.35 11.89 49.97 0.43
1993 6.80 25 24.15 19.54 37.50 16.83 16.18 83.82 8.00 29.71 11.97 50.32 0.40
1994 9.20 25 25.27 19.54 38.60 17.61 15.61 84.39 7.91 30.19 11.30 50.60 3.59
1995 9.60 27 27.66 20.86 39.20 19.48 14.97 85.03 7.49 30.69 11.41 50.41 3.19
1996 7.60 26 28.96 19.86 39.90 24.00 14.87 85.13 7.07 29.90 11.69 51.34 0.44
1997 5.90 30 30.39 22.10 37.60 28.84 16.22 83.78 6.49 29.50 11.83 52.17 0.04
1998 -5.50 42 30.55 32.10 27.90 40.26 19.17 80.83 6.02 30.50 10.07 53.41 -2.97
1999 11.30 36 29.65 27.02 31.20 28.75 17.88 82.12 6.14 31.45 8.83 53.58 12.60
2000 8.90 35 32.94 27.83 33.20 24.07 17.24 82.76 5.60 32.38 7.87 54.15 8.86
2001 4.50 33 31.18 25.60 31.90 21.77 18.02 81.98 5.29 30.64 8.14 55.93 -2.70
2002 7.40 31 29.33 24.59 31.20 21.13 17.01 82.99 4.92 30.70 8.29 56.09 2.38
2003 2.90 33 30.69 26.39 32.30 20.33 18.34 81.66 4.58 30.18 9.29 55.95 1.74
2004 4.90 38 34.47 30.59 32.30 19.39 18.28 81.72 4.56 32.19 8.95 54.30 4.44
2005 3.90 37 34.37 28.79 32.50 18.03 17.58 82.42 4.14 32.10 8.81 54.95 0.17
2006 5.20 37 36.39 28.78 33.00 22.66 16.73 83.27 3.92 31.71 8.70 55.66 2.55
2007 5.50 39 38.06 29.52 32.80 30.17 16.47 83.53 3.61 31.99 8.61 55.79 0.45
2008 2.80 50 49.97 36.60 33.00 31.52 16.88 83.12 3.42 32.81 8.17 55.60 -1.54
2009 0.70 48 42.86 36.11 28.60 38.21 20.75 79.25 3.27 32.50 8.03 56.20 0.66
2010 6.50 49 46.23 37.91 32.10 32.52 17.11 82.89 3.10 35.36 7.25 54.29 7.53
2011 3.70 56 54.25 39.68 32.90 33.27 16.31 83.69 3.15 36.35 6.77 53.73 7.53
2012 2.30 56 53.55 38.12 30.80 33.44 16.09 83.91 - - -
2013 3.00 54 48.86 36.99 28.80 32.46 15.98 84.02 - - -
Source: Elaborated by the author, using data from The Bank of Korea (BOK), Collins and Park (1989), World Bank, OECD and The Groningen Growth and Development
Centre.
Note: Data for GDP Growth and Gross Fixed Capital Formation (GFCF) in current Prices. Sectoral composition calculated from data at constant 2005 national prices (in
millions).
101
Appendix 2.3 – Classification of manufacturing industries by technology group

International Standard Abbreviation used International Technology

Industrial Classification full in this chapter Standard group
description Industrial
Classification
code Revision 3
Food and beverages Food and beverages 15 Low-tech
Tobacco products Tobacco 16 Low-tech
Textiles Textiles 17 Low-tech
Wearing apparel, fur, leather Wearing apparel 18 and 19 Low-tech
products and footwear
Wood products (excluding Wood products 20 Low-tech
furniture)
Paper and paper products Paper 21 Low-tech
Printing and publishing Printing and 22 Low-tech
publishing
Furniture; manufacturing, not Furniture, not 36 Low-tech
elsewhere classified elsewhere classified
Coke, refined petroleum products Coke and refined 23 Medium-tech
and nuclear fuel petroleum
Rubber and plastic products Rubber and plastic 25 Medium-tech
Non-metallic mineral products Non-metallic 26 Medium-tech
minerals
Basic metals Basic metals 27 Medium-tech
Fabricated metal products Fabricated metals 28 Medium-tech
Chemicals and chemical products Chemicals 24 High-tech
Machinery and equipment, not Machinery and 29 and 30 High-tech
elsewhere classified; office, equipment
accounting and computing
machinery
Electrical machinery and Electrical 31 and 32 High-tech
apparatus; radio, television and machinery and
communication equipment apparatus
Medical, precision and optical Precision 33 High-tech
instruments instruments
Motor vehicles, trailers, semi- Motor vehicles 34 and 35 High-tech
trailers and other transport
equipment
Source: Unido (2013).

102
CHAPTER 3 - PRODUCTIVE DIVERSIFICATION,
STRUCTURAL TRANSITION AND DENSITY OF
INDUSTRIAL CHAINS: ARE BRAZIL AND SOUTH
KOREA VERY DIFFERENT?

Introduction120

Since the outbreak of the debt crisis in 1982, Brazil and South Korea have shown distinct
economic performances. During the 1990s in both Brazil and South Korea, although to
different extents, neoliberal reforms were intensified under the assumption that economies
could adjust effectively to a tough international environment with a hastily devised trade and
financial liberalisation. In Brazil, more specifically after the Real Plan (in 1994), the
economic environment began to be characterised by an overvalued currency, high interest
rates and increasing participation of foreign capital in the domestic market. Under the
fundamentalist process of economic reforms, the role of the State become secondary since the
main goal of the proponents of the neoliberal view was strong market-friendly policies to
define the allocation of resources. Therefore, as a result this pro-market environment, the
neoliberal agenda has resulted in poor economic growth for several decades.

Despite of the Brazilian economic boom from 2002 until the global financial crisis of
2008, the neoliberal shock of the 1990s produced deleterious economic changes. The country
faced a productive disarticulation and the absence of sustained economic growth. In a “stop-
and-go” performance, the manufacturing sector has gradually shrunk. In this process,
domestic industrial chains have shrunk with the increase of imported inputs121. In other words,
imported inputs played a substitutive role in the productive chain. Moreover, the economy has
experienced a specialisation in less dynamic sectors such as services and commodities.
Furthermore, in terms of trade, the economy has increased its dependence on exports of
primary products. For instance, the share of agricultural and mineral commodities in total

120
I would like to thank Guilherme Magacho for his helpful comments on this chapter.
121
For an explanation of the industrial dynamics of the Brazilian economy resulting from the economic reforms,
see chapter 2.

103
exports rose from 6.27% and 6.43% in 1995 to 11.38% and 13.11% respectively in 2009,
negatively affecting the Brazilian trade pattern.

In South Korea, during the same period, reforms regarding trade and financial
liberalisation were also adopted. However, the economic reforms were totally different to the
full-scale overhaul seen in Brazil. In fact, economic reforms in South Korea were much more
pragmatic in order to continue and strengthen the existing ambitious strategy of
industrialisation. Even during the Asian crisis in 1997, in spite of the re-articulation between
state and private sector, the role of the state remained fundamental. To overcome the crisis
there was a reorganisation of the corporate structure with processes of mergers and
acquisitions gaining momentum. However, it is important to note that at no time was the
manufacturing sector neglected. Therefore, the South Korean state remained active promoting
the consolidation and expansion of industrial structure and exports. In this process, imports of
inputs had fundamentally a complementary role in the productive process. Moreover, in a
strategy of productive and trade upgrading, South Korea took advantage of the special
properties of manufacturing as an engine of growth. Not by coincidence, the South Korean
economic growth has been higher and much more sustainable when compared to the Brazilian
economy.

Therefore, inserted in this short background, this chapter seeks to analyse the
structural transition of Brazil and South Korea. Through the input-output framework, from a
comparative perspective, the chapter makes an effort to investigate these economies regarding
their pattern of productive diversification, structural transition and density of industrial chains.
Therefore, the contributions of this chapter to the existing literature are threefold. First, the
chapter adopts the methodology proposed by Imbs & Wacziarg (2003) in an inter- and intra-
sectoral perspective to identify the path of economic diversification undertaken by Brazil and
South Korea over the decades. Second, in a complementary analysis, the input-output
framework is used to identify and quantify factors that contributed to this pattern of
specialisation. The method used was based on the Structural Decomposition Analysis (SDA)
inspired particularly by Chenery’s studies. Third, the chapter advances in the input-output
methodology employed by Rocha (2011) to estimate imported input coefficients for Brazil
and South Korea through the World Input-Output Database (WIOD).

The chapter is organised as follows. The first section discusses the long-lasting debate
on specialisation or diversification and applies the method proposed by Imbs & Wacziarg

104
(2003) to investigate the stages of sectoral diversification presented by Brazil and South
Korea over the decades. The methodology is also described in this section, including database
sources and the level of sectoral aggregation. The second section focuses on the sectoral
composition of manufacturing in these economies and their economic dynamism in a global
perspective. The third section, through the so-called structural decomposition analysis (SDA),
investigates distinct factors, i.e. demand, trade and technological change, which contributed to
the movement of structural change observed in these economies. In the fourth section, the
chapter consolidates the understanding of the structural economic dynamics via movements
related to the density of industrial chains. The fifth section concludes the chapter.

1. Structural change and economic dynamic: Specialisation vs. Diversification

The neoclassical view stresses that specialisation based on comparative advantages,

whatever its nature, was a superior solution for economic growth. In this thinking, through the
process of specialisation, resources become more efficiently allocated and consequently
countries benefit mutually in terms of welfare (Krugman & Obstfeld, 2006)122. Influential
neoclassical visions such as Lipton (1968), Davis, (1995) and Mikesell (1997) grounded their
analyses on the neoclassical theory of comparative advantage and recommended that late
development countries (LDC) should specialise in the production and exports of primary
commodities and import manufactured products from industrialised economies. Although the
idea of comparative advantage has received further developments in the current mainstream
approach, particularly in the New Structural Economics (NSE) framework123, specialisation
according to the country’s factor endowment structure continues to be an integral part of
policy advice. As pointed out by Rodrik (2004, p. 6) many economists that “associate
underdevelopment with inadequate exposure to international markets generally imply –
although this is often left unstated – that specialisation according to comparative advantage is

122
The comparative advantage theory was developed by David Ricardo in his Principles of Political Economy in
1817. Ricardo uses trade between England and Portugal in cloth and wine respectively to advocate in favour of
free trade policies. In this example, these countries would benefit mutually because they would be specialising in
the good that has lower opportunity cost (Ricardo, 1817). Despite the simplicity of Ricardo’s theory, it has been
interpreted in several different ways, and continues to have a remarkable impact on the current mainstream
theory.
123
According to the former chief economist of the World Bank Justin Yifu Lin, “the economic structure of an
economy is endogenous to its factor endowment structure and that sustained economic development is driven by
changes in factor endowments and continuous technological innovation”. In other words, “the new structural
economics argues that the best way to upgrade a country’s endowment structure is to develop its industries at
any specific time according to the comparative advantages determined by its given endowment structure at that
time”. (Lin, 2012, p. 5).

105
an essential ingredient of development”. Currently this understanding is also supported by the
World Trade Organization (WTO), which advocates in favour of market-friendly policies.
The WTO (2011) states that “countries prosper first by taking advantage of their assets in
order to concentrate on what they can produce best, and then by trading these products for
products that other countries produce best”. Moreover, specialisation on comparative
advantages through liberal policies would sharp competition, motivate innovation and breed
success (ibid, 2009).

From a critical perspective, the Structuralist-Kaldorian view points out the limitations of
a strategy of economic development based on comparative advantages. Therefore, based on
the argument that diversification in manufactured products is the main engine of growth,
authors such as Rosenstein-Rodan (1943), Prebisch (1949), Lewis (1954), Rostow (1952),
Furtado (1961), Hirschman (1958), Kaldor (1966, 1967) and Chenery (1960, 1979) belong to
the handful of economic thinkers that have emphasised the importance of diversification to
economic development. According to them, development is essentially a process of structural
change. In other words, sustained economic growth is associated with the capacity to diversify
the structure of domestic production, i.e. generate new activities to expand possibilities of
production, linkages and higher value-added goods by providing incentives to manufacturing.
According to Kaldor (1966), economic growth is brought about by shifting productive sectors
with constant or decreasing returns to those with increasing returns to scale. This shift creates
dynamic economies of scale and triggers a process of cumulative causation in the economy.
The author states that manufactured goods is the sector with the greatest capacity to do so and
therefore its expansion plays a key role in promoting sustainable growth in the long term and
in the consequent modernisation and diversification of the production structure124.

1.1. The U-shaped curve for Brazil and South Korea

Recently, this long-lasting debate regarding economic specialisation or diversification

has been approached by empirical studies that tried to model the trajectory of economic
diversification undertaken by a vast range of economies. In an important paper, Imbs &
Wacziarg (2003) investigated the stages of sectoral diversification in a large cross-section of
countries. Using OECD, ILO and UNIDO databases, this study employed a non-parametric
methodology based on locally robust weighted scatter plot smoothing (Lowess) to identify

124
For a detailed discussion on the special growth-enhancing properties of manufacturing, see chapter 1.

106
regularities regarding the process of productive diversification. Thus, the authors detected a
non-linear relationship between specialisation and GDP per capita, described as a U-shaped
curve.

In this U-shaped curve, low-income countries present a very specialised productive

structure. As countries move from low to medium levels of income, diversification takes place
in the economy. This process goes on until relatively late in the process of economic
development. On average, the diversification takes place until the country reaches a turning
point around $ 9,000 GDP per capita. Thereafter, the sectoral distribution exhibits re-
specialisation with the curve presenting an ascendant trajectory. From this perspective, as
pointed out by Rodrik (2004), Imbs and Wacziarg’s findings suggest that specialisation
according to comparative advantage as conventionally understood cannot be the driving force
of economic development. In other words, countries grow and reach a high level of income
per capita during the phase of diversification and not specialisation.

Therefore, seeking to evaluate the movement of structural change undergone by Brazil

and South Korea, from a comparative perspective, this section employed the methodology of
Imbs and Wacziarg to investigate these economies from both an intersectoral perspective, i.e.
between primary sector, manufacturing, civil construction and services, as well as from an
intra-sectoral perspective (among manufacturing subsectors)125. In order to calculate each
economy’s inter-sectoral curve of specialisation, the GGDC 10-Sector Database and the IBGE
database were used for South Korea and Brazil respectively. For the intra-sectoral analysis of
South Korea, the manufacturing value-added of 23 subsectors from the UNIDO database was
used. In the case of Brazil, due to lack of long-term data, this study used the ECLAC-PADI
database with 28 manufacturing subsectors. Since this chapter seeks to compare Brazil and
South Korea, both databases were aggregated on a common level involving 14 manufacturing
subsectors126. In addition to using the same aggregation level, Figures 3.1 and 3.2 present two
distinct graphics. The first graph plots the Gini-Hirschman index (GHI) against GDP per
capita, along with a non-parametric Lowess curve. The second shows the sectoral share
between the primary sector, manufacturing, civil construction and services. The graphs are
presented together to investigate particularities regarding the process of economic
diversification and structural change undergone by these economies.

125
Technical details about methodology are summarised in Appendix 3.1.
126
For details about the aggregation see Appendix 3.2.

107
 From an inter-sectoral view, based on a developmental premise that industrialisation in
general and manufacturing specifically is a central condition to economic development, South
Korea consistently raised its share of manufacturing followed by a substantial decrease of the
primary sector. From 1953 to 2010, the former rose from 8% to 35% while the latter fell from
49% to 3%. Moreover, a joint analysis between data from the curve of diversification and the
sectoral share of the economy reveals that the movement of intersectoral structural change
pursued by South Korea followed two stages of manufacturing expansion. In the first stage,
from 1953 to 1980, South Korea moved to a strategy of economic diversification expanding
remarkably not only the manufacturing sector but also civil construction (and strongly
services during the 1950s). During this period the manufacturing share rose by 236.6%, while
civil construction increased by 337.9% and services by 13.3%. Thus, by the end of the 1980s,
the South Korean economy achieved a considerable degree of diversification.

Figure 3.1 - The inter-sectoral U-shaped curve and sectoral share for South Korea

68 100%

90%
66 08
80% 41 39
64
05 47 46 43 46 49 49
50 54
55 54
00 70%
62 53 95 60% 2 3
60
(GHI)

65 8 4
60 50% 3 5 8
70 90 17 7
12 11 11
75 85 40% 8 9 7
58 20 25
80 28
30% 29
56 49 29 31
32 32 35
20% 38 41
29 28
54 10% 18 15
10 7 6 4 3
52 0%
0 5,000 10,000 15,000 20,000 25,000
GDP per capita
Primary Sector Manuf CC Services

Source: Own calculations based on data from GGDC 10-Sector Database.

Note: Data calculated in current US dollars.

In the second stage, from 1980 to 2010, although the pace of structural change
decreased, with the share of manufacturing and services growing respectively by 25.1% and
17.6%, the economy engendered a process of economic specialisation. The changes in the
shares of services and manufacturing from the beginning to the end of this stage show the
magnitude of this trajectory. While in 1980 manufacturing and services together accounted for
approximately 74% of the economy value-added, in 2010 this number jumped sharply to 90%.

108
It was only possible because the expansion of both sectors took place mainly as a result of the
reduction of the primary sector. Even during the decades of 1990s and 2000s, the clear rise of
services was not followed by a loss in manufacturing. Therefore, the Figure 3.1 illustrates
clearly that despite of the large share of services in the economy, South Korea has showed a
pattern of specialisation based on the expansion of the manufacturing sector.

In a manner distinct from South Korea, Brazil did not experience the same process of
structural change and productive diversification for the period analysed. As a matter of fact,
the country smoothed data did not fit well to the U-shaped curve. However, two important
issues must be considered. Firstly, an important period of economic diversification not
captured by the data was engendered by the Brazilian economy during the 1930s and 1940s.
In other words, as widely documented in the Brazilian economic literature, the process of
economic diversification was initiated in the 1930s after the great world economic crisis in
1929. At that time, Brazil moved from a primary export model to a strategy of Import
Substitution Industrialisation (ISI)127. Secondly, despite an upward sloping, the country
maintained a diversified inter-sectoral productive structure until the first half of the 1970s. As
a matter of fact, from 1950 to practically 1980, the share of manufacturing and civil
construction grew respectively by 67.6% and 66.6%, while the primary sector and services
dropped by 54.7% and 3.5%. Moreover the GDP per capita expanded continually.

However, with the outbreak of the debt crisis (in 1982) and the rise of neoliberal
policies, Brazil embarked on a regressive trajectory128. From 1980 to 2010, not only
manufacturing and civil construction declined sharply in terms of GDP share (from 31.29% to
16.23% and from 6.81% to 5.65% in each sector respectively) but also income per capita
regressed in many periods such as 1981–1983, 1988–1992 and 1998–1999. In this sense, in
contrast to South Korea, Brazil did not experience the same long-term dynamism in
manufacturing and GDP per capita growth. Additionally, one of the most notable effects of
market-friendly policies, which intensified during the second half of the 1990s, was the
collapse of manufacturing share in the total GDP. The economy gradually lost the
manufacturing sector to the point where it represented a lower share than in 1950. Figure 3.2
illustrates these processes.

127
For details on the Import Substitution Industrialisation (ISI) in Brazil, see chapter 2.
128
For details on the Brazilian economic dynamic after the 1980s see chapter 2.

109
 Figure 3.2 - The inter-sectoral U-shaped curve and sectoral share for Brazil

82 100%

90%
77
80%
08
53 52 51 54 55 51 51 49
70% 62
72 69 70 69 70
05
60%
(GHI)

00
67 95 50%
80 4 5 5
4 5 6 7
90 5
85 40%
62 75 19 20 7
30% 26 25 32
55 60 65 70 5 5 6
47 50 27 31 31 6
20% 23
57 19 17 18 16
10% 25 24
19 17
12 12 11 14 8 8 8
7 7
52 0%
0 2,000 4,000 6,000

GDP per capita

Primary Sector Manuf CC Services

Source: Own calculations based on data IBGE.

Note: Data calculated in current US dollars.

As previously mentioned, the intra-sectoral movement inside the manufacturing sector

was also analysed. As it is possible to see in Figure 3.3, both Brazil and South Korea
experienced a U-shaped trajectory. However, from a comparative perspective, the U-shaped
pattern also presented three important differences. First, the phase of economic diversification
was more intense in South Korea. Second, in terms of value-added, the South Korean
economy showed the point of maximum diversification in 1988 with $7,621 GDP per
capita129, while in Brazil it occurred in 1975 with $4,187 GDP per capita. After these stages,
both economies engendered an ascendant and constant trend of re-specialisation.
Nevertheless, as showed by the Figure 3.3, after 1988 South Korea specialised its
manufacturing in subsectors that enabled a substantial increase in terms of GDP per capita.
Thus, from 1988 to 2008 the GDP per capita increased by 168.8%, reaching $ 20,482 in 2008.
In the case of Brazil, the premature trajectory of re-specialisation culminated in a less
dynamic productive structure that trapped the economy into a 'low-growth landscape'. From
1975 to 2008, the GDP per capita increased only by 54.7% (on average an annual increase of
1.42%) and the economy showed a chronic dynamic of ‘stop and go’. The following section
analyses manufacturing subsectors and their evolution over time.

129
It is important to note that the turning point of a trajectory of re-specialisation is significantly lower than those
observed in advanced countries. For more details see Imbs & Wacziarg (2003).

110
Figure 3.3 - The intra-sectoral U-shaped curve

South Korea (1963-2008) Brazil (1970-2008)

33
42

32.5 08
40 08

05 32
63 05
38

(GHI)
(GHI)

70
00 31.5 00
36
95
70 80
95 31
34 80 75 90
85 85
90
32 30.5

30 30
0 5,000 10,000 15,000 20,000 25,000 2,000 3,000 4,000 5,000 6,000 7,000

GDP per capita GDP per capita

Source: Own calculations based on data from UNIDO and ECLAC-PADI.

Note: Data calculated in constant US dollars for Brazil and in current prices for Korea due the lack of data.

1.2. Sectoral composition and economic dynamism

Another dimension that deserves a close analysis to understand the path of

specialisation undertaken by the Brazilian and South Korean manufacturing industries in
relation to the GDP per capita is the evolution of this process over time. In this context, by
simply taking into account the share of each manufacturing subsector in the total value-added
and specialisation indexes over decades it is possible to trace the pattern of specialisation
under way in these economies. Moreover, through this analysis is also possible to investigate
the pattern of specialisation in terms of technology intensity. Therefore, starting in 1990,
Table 3.1 shows the evolution of sectoral composition of the manufacturing value-added in
Brazil and South Korea, at an interval of five years in each decade.

The trajectory of intra-sectoral specialisation in South Korea was marked by a

significant drop of low-tech sectors compensated by a strong expansion in high-tech sectors.
From 1990 to 2008, the share of the former in the total manufacturing slumped from 33.2% to
13.9% while the latter increased sharply from 43.4% to 59.3%. In this trend, the sharp

111
increase of the ‘Machinery’ subsector draws attention particularly due to its substantial rise in
the total value-added of the economy, i.e. by 10.1 percentage points. Recognised as a
knowledge-intensive subsector of manufacturing, the Structuralist-Schumpeterian literature
has stressed the growth-inducing effects of the machinery subsector particularly regarding the
introduction and diffusion of technological change130. Another sector that contributed to
expanding the high-tech group was ‘Transport Equipment’, which rose from 10.5% to 16.8%.
Additionally, the increase of medium-tech subsectors, although to a lesser extent, also
contributed to productive specialisation (and economic dynamism), increasing by 3.4
percentage points. In short, the South Korea's pattern of intra-sectoral specialisation was
marked by a gradual share concentration in medium- and high-tech sectors that together
reached 86.1% of the total value-added in 2008.

Table 3.1 – Sectoral composition of the manufacturing sector (%), 1990-2008

South Korea Brazil
Subsectors 90 95 00 05 08 90 95 00 05 08
Food, Beverages and Tobacco 10.7 8.4 8.3 6.6 6.1 15.1 13.4 14.5 13.6 15.1
Textiles and Textile Products 10.3 8.8 7.1 4.8 3.6 9.7 7.9 7.0 5.5 4.9
Leather, Leather and Footwear 3.6 1.4 0.9 0.5 0.4 2.3 1.4 1.0 0.8 0.7
Wood and Products of Wood and Cork 0.9 0.8 0.6 0.5 0.4 1.5 1.1 1.2 1.1 0.9
Pulp, Paper, Paper , Printing and
Publishing 4.9 4.9 4.8 4.0 2.1 6.5 7.8 7.7 7.8 7.1
Coke, Refined Petroleum and Nuclear
Fuel 3.0 3.4 2.6 3.5 4.6 6.1 7.6 8.6 7.8 6.9
Chemicals and Chemical Products 9.2 9.3 9.5 8.6 8.7 12.7 11.1 11.3 10.0 9.0
Rubber and Plastics 3.9 3.9 4.0 4.7 4.3 4.6 3.3 3.4 2.9 2.6
Other Non-Metallic Mineral 5.1 4.8 3.8 3.4 3.3 2.6 2.3 2.4 2.1 2.1
Basic Metals and Fabricated Metal 11.4 11.5 10.4 13.2 14.6 13.3 13.9 13.4 13.6 13.0
Machinery 19.7 24.8 28.2 30.2 29.8 8.1 10.4 9.8 12.0 12.8
Electrical and Optical Equipment 4.0 4.3 4.8 4.5 4.0 8.2 10.5 10.9 11.8 11.4
Transport Equipment 10.5 11.3 13.1 13.8 16.8 7.1 7.8 7.6 10.0 12.4
Manufacturing, Nec; Recycling 2.9 2.3 1.8 1.6 1.3 2.2 1.6 1.3 1.1 0.9
Total manufacturing 100 100 100 100 100 100 100 100 100 100
Low-Tech 33.2 26.6 23.4 18.0 13.9 37.4 33.2 32.7 30.0 29.7
Medium-Tech 23.4 23.7 20.9 24.9 26.8 26.5 27.0 27.7 26.2 24.7
High-Tech 43.4 49.7 55.7 57.2 59.3 36.1 39.8 39.6 43.7 45.7
GHI 32.3 34.8 37.1 38.8 39.8 31.0 31.3 31.5 31.9 32.3
Source: Own calculations based on data from UNIDO and ECLAC-PADI.
Note: Data calculated in 2008 constant US dollars for Brazil and in current prices for Korea due the lack of data.
Nec = Not elsewhere classified.

In Brazil, the pattern of productive specialisation was also marked by the expansion of
high-tech subsectors such as ‘Machinery’, ‘Electrical and Optical Equipment’ and ‘Transport

130
For a detailed explanation about the growth-enhancing particularities of the machinery subsector, see chapter
1.

112
Equipment’. However, in spite of the substantial increase of 9.6 percentage points in this
group, the pace and proportion of this expansion was clearly lower than in South Korea.
Moreover, although the low-tech sector showed a clear drop of 7.7 percentage points, the
Brazilian economy remained dependent on less dynamic sectors such as ‘Food, Beverages and
Tobacco’, which represents 15.1% of the total value-added in the economy. Furthermore, the
medium-tech group presented a slight descendent trajectory varying from 26.5% in 1990 to
24.7% in 2008. In this group, the clearest loss was in ‘Rubber and Plastics’ which dropped by
2 percentage points. In brief, Brazil presents a manufacturing sector more diversified than
South Korea but in less dynamic sectors. From the perspective of comparative economic
development, this dichotomy stimulates different processes of cumulative causation and,
consequently, results in divergent growth rates.

The inter- and intra-sectoral composition of Brazil and South Korea culminated in
distinct economic performances. From a comparative perspective it can also be illustrated in
relation to the global economy. Thus, figure 3.4 shows the evolution of their shares in both
global manufacturing and GDP, elucidating the South Korean thriving economic dynamism
vis-à-vis the poor performance of Brazil. Therefore, over the decades, South Korea showed a
sustained economic growth taking advantage of dynamic economies of scale, increasing
returns and spillover effects that set in motion processes of cumulative causation. From 1950
to 1980, the share of South Korea in global GDP expanded by 133.1% and manufacturing
rose by 210.7% (from 1970 to 1980). Even after the turbulent 1980s and the rise of neoliberal
policies in the developing world particularly in the 1990s, the economy kept the impetus of
growth, increasing its share in the global GDP and manufacturing. Brazil, by contrast,
although with a considerable increase of these variables from 1950 until 1980, collapsed with
the outbreak of the debt crisis. During the 1980s, Brazil was trapped in an economic dynamic
that sentenced the country to a poor economic performance. From the 1990s, with the advent
of neoliberal reforms the economy not only remained stagnated in the evolution of the global
GDP share and but also lost a significant share in global manufacturing.

113
Figure 3.4 - Shares in global manufacturing and GDP

Share in Global GDP Share in Global Manufacturing

3.50 3.50 3.27

3.12
3.00 3.00
2.59 2.68
2.49 2.47 2.55
2.50 2.50
2.06 2.09
1.92 1.90 1.91 1.98
2.00 2.00 1.76
1.65
1.54
1.43
1.50 1.25 1.50

1.00 0.76 1.00

0.59
0.49
0.50 0.33 0.35 0.50
0.19

0.00 0.00
1950 1960 1970 1980 1990 2000 2010 1970 1980 1990 2000 2010

Brazil South Korea

Source: Elaborated by the author, using data from The Groningen Growth and Development Centre and
UNCTAD.
Note: Sectoral shares calculated at 2005 constant national prices.

2. The structural decomposition analysis

Seeking to understand the results found in the previous section, this section tries to
find factors that directed the path of structural change observed in Brazil and South Korea in
light of liberalising policies implemented in these economies during the second half of the
1990s. Therefore, based on the pioneer analysis developed by Leontief (1936, 1941), the
theoretical approach is based on the input-output method. Over the years, this methodology
has been widely used in structural analyses to study effects of distinct economic policies over
the structure of production. Thus, from an analytical input-output framework, the so-called
structural decomposition analysis (SDA) inspired by Chenery (1960), Chenery et al. (1962),
Chenery & Syrquin (1980) and Chenery et al. (1986), has been applied for many countries in
vast range of contexts and periods to identify and quantify factors that contribute to a given
change in the sectoral structure. This method is usually adopted to investigate distinct factors,

114
i.e. demand, trade and technological change, which contributed to a certain change in the
industrial structure131.

In contrast to the framework of neoclassical models, which consider economic growth

as a process limited by the supply side, the SDA is a method of growth accounting on the
demand side. Furthermore, it takes account of inter-relationships between various sectors (and
subsectors) of the economy, rather than sectoral trends in isolation from one another. Using
this methodology, it is possible to empirically investigate the economic role of a productive
sector without restricting the analysis to its “direct effects” on the economy regarding
generating production, employment, value-added products, tax revenue, and exports. With
this method, it is also possible to investigate the “indirect effects”, i.e. the effects that a sector
can exert on other sectors through channels established by input-output transactions between
different economic sectors.

Over the years, the SDA methodology was improved as analytical tool and different
methods were developed. In this sense, Feldman et al. (1987), Skolka (1989), Rose & Casler
(1996) and Dietzenbacher & Los (1998) were particularly relevant. More recently, many
studies analysed the economic structure through this line of methods. Kupfer el al. (2003)
decomposed the output and employment variation of the Brazilian economy between 1990
and 2001. Tregenna (2012) applied this method to South Africa from 2000 to 2007. In
addition, Zakariah & Ahmad (1999) used the factor decomposition approach to Malaysia from
1978 to 1987. However, none employed the SDA methodology from a comparative
perspective between Brazil and South Korea through a common database. Therefore, the data
available at the World Input-Output Database (WIOD) was used to analyse the output
decomposition between two periods: 1995-2002 and 2003-2008. These breakdowns were
applied more as a result of data availability than because of distinct economic policies. In
addition, input-output tables were estimated in constant prices using the database available in
both current and previous years’ prices132. Therefore, analogously to Chenery et al. (1986)
this chapter decomposed output growth (∆Q) in manufacturing into the following four
components133:

131
These explanatory factors are commonly used in the debate on industrialisation/deindustrialisation to
understand a given pattern of specialisation.
132
Previous analysis showed that for this type of methodology, data for 2009 was strongly affected by the global
financial crisis. Therefore, exclusively in this chapter, this year was excluded in the input-output analysis.
133
Technical details about methodology are summarised in Appendix 3.1.

115
 • Domestic demand (∆R): Refers to the direct and indirect effects of the variation in the
domestic final demand on the sector (or subsector) output. Domestic demand includes
household consumption expenditure, government consumption expenditure and gross
capital formation.
• Export (∆S): Indicates the direct and indirect effects of the variation of exports on
each sectoral (or subsector) output. It is important to note that this component is
affected by intersectoral relationships. In other words, it is affected not only by exports
of the sector under analysis, but also by exports of other sectors with which this sector
is linked.
• Import substitution (∆T): Refers to the direct and indirect effects of the variation of
sectoral output associated with the substitution of imports (intermediates and final
goods) by domestic production. Analogously to export expansion, this component is
affected by intersectoral relationships and consequently reflects not only import
substitution in a specific sector, but also import substitution in other sectors with
which a sector is linked.
• Technological change (∆U): Indicates the variation on sector (or subsector) output
associated with changes in the production processes which affect the economy
technical coefficients (input-output coefficients). Thus, this component is obtained
through changes in the table of technical coefficients, which shows the flow of
intermediate inputs (domestically and imported) into the production of all goods and
services of the economy.

Tables 3.2 and 3.3 show the results for both countries from 1995 to 2002 and 2003 to
2008. For each sector (or subsector) the contributions of the four components (domestic
demand, export, import substitution and technological change) to economic growth are
presented. Inter- and intra-sectoral results are presented to provide subsidies to understand the
process of specialisation in Brazil and South Korea. In order to maintain a robust intra-
sectoral analysis, results of the SDA were presented as close as possible to the aggregation of
the last section. Thus, manufacturing was aggregated in 15 subsectors. Additionally,
manufacturing was divided by technology intensity (low-, medium- and high-tech) according
to UNIDO classification.

116
 3. Interpreting results in light of distinct economic policies

During the 1990s, Brazil and South Korea experienced distinct economic policies. In
Brazil, after a turbulent and critical passage through the 1980s marked by economic
stagnation and high inflation rates, the Real Plan (Plano Real) was adopted in 1994, aiming to
stabilise the economy. The economic policy of the period was essentially based on liberalising
reforms that sought to stabilise the economy through high interest rates, currency
overvaluation and wage squeezes. Measures such as trade liberalisation (the reduction of tariff
and non-tariff barriers), privatisation of state-owned enterprises and capital account openness
culminated in a disastrous economic environment for the national industry. Moreover, from
1998 to 2003 the Brazilian economy experienced a period of balance-of-payment crisis and
currency depreciation that was associated with a restrictive macroeconomic policy based on
the inflation target regime which undermined any attempt of significant recovery in the
manufacturing sector.

Therefore, as shown in Table 3.2, from 1995 to 2002 manufacturing output increased
by only 9.1% despite exports increasing by 15.9%. Additionally, it is important to note that
during this period, with the exception of exports, all components (domestic demand, import
substitution and technological change) of the Brazilian manufacturing sector registered
negative variations. In an intra-sectoral perspective, the results of the structural decomposition
analysis reveal that many manufacturing subsectors affected were those belonging to the
medium- and high-tech group. Remarkable examples are the subsectors of ‘rubber and
plastics’, ‘non-metallic mineral’ as well as ‘electrical and optical equipment’. In contrast the
primary sector output based on the Brazilian comparative advantage grew by 43.0%, mostly
driven by exports which represented 26.0% of this expansion. Furthermore, during this period
services output expanded by 18.0% probably as a result of the privatisation process of the
decade, particularly in subsectors of infrastructure such as energy and telecommunications134.

In a distinct economic dynamic, South Korea did not embark on the same type of
‘fundamentalist’ economic reforms. As a matter of fact, although financial and trade
liberalisation marked South Korea during the 1990s, it was much more pragmatic in order to
continue and strengthen the existing ambitious strategy of industrialisation. Over the decades,
the South Korean history of economic development has been notable for the special properties

134
For details on the process of privatisation see chapter 2.

117
 of manufacturing as a key sector to economic growth. Therefore economic reforms were
implemented as a pragmatic mechanism to help lift specific pressing economic and financial
constraints in order to expand the manufacturing sector. Even during the Asian crisis in 1997,
in spite of the re-articulation between the state and private sector, the role of the state was
central in maintaining industrial dynamism. To overcome the crisis there was a reorganisation
of the corporate structure with processes of mergers and acquisitions gaining momentum.
Additionally, the South Korean state remained active both through long-term programmes for
the development of high-tech sectors and by ensuring important long-term funding sources
through public banks.

Table 3.2 - Inter and Intra-sectoral decomposition of output growth (%), 1995-2002
South Korea Brazil
Sectors (∆U) (∆T) (∆R) (∆S) (∆Q) (∆U) (∆T) (∆R) (∆S) (∆Q)
Primary Sector 21.0 -16.8 -8.7 4.5 0.0 9.6 1.0 6.3 26.0 43.0
Manufacturing -9.7 6.3 11.0 50.7 58.3 -2.7 -0.8 -3.3 15.9 9.1
Food, Beverages and Tobacco 14.7 3.7 -4.9 4.8 18.3 -3.3 1.0 5.4 16.3 19.3
Textiles and Textile Products -2.1 9.2 5.6 8.1 20.8 -4.5 1.9 -17.9 6.3 -14.1
Leather, Leather and Footwear -14.3 2.4 -1.4 -15.7 -29.0 -5.4 0.6 -19.1 17.2 -6.7
Wood and Products of Wood
and Cork -28.6 18.7 11.3 12.4 13.8 -27.4 -0.1 -2.2 32.7 3.0
Pulp, Paper, Paper , Printing
and Publishing -14.6 11.9 21.9 30.3 49.5 -1.9 3.3 6.7 12.5 20.6
Coke, Refined Petroleum and
Nuclear Fuel -38.6 4.9 13.9 47.8 28.1 7.2 -0.5 -2.9 13.7 17.5
Chemicals and Chemical
Products -2.9 14.0 6.3 57.8 75.1 7.9 -2.5 -1.8 11.1 14.7
Rubber and Plastics -10.9 1.5 10.4 50.7 51.7 -8.5 -0.6 -4.8 14.8 0.9
Other Non-Metallic Mineral -23.5 -1.8 18.5 29.3 22.7 -1.3 0.0 -4.6 10.0 4.0
Basic Metals and Fabricated
Metal -22.2 5.2 9.2 45.7 37.9 -8.7 -1.4 -1.4 19.4 7.9
Machinery -9.3 8.3 18.0 50.6 67.7 -4.1 -1.6 3.7 16.6 14.5
Electrical and Optical
Equipment 24.5 7.5 63.1 222.1 317.2 -1.2 -5.7 -21.0 14.6 -13.4
Transport Equipment 8.9 4.2 3.5 69.7 86.4 -8.7 -2.4 -6.8 28.7 10.7
Manufacturing, Nec;
Recycling -4.0 3.2 -1.2 16.6 14.6 -5.0 -0.1 -4.0 7.9 -1.2
Services -4.0 0.4 36.8 11.7 44.9 3.3 0.0 10.5 4.2 18.0
Source: Own calculations based on data from WIOD.
Note: Data calculated in 2008 US constant dollars.

Indeed, from an inter- and intra-sectoral perspective, Table 3.2 highlights the
dichotomy between Brazil and South Korea. From 1995 to 2002 manufacturing production in
South Korea increased by 58.3% of which 50.7% was explained by increase in exports.
Moreover the booming expansion of medium- and high-tech industries composed one of the
main sources of economic dynamism. During the 1980s and 1990s, the South Korean

118
government engendered a massive effort to constitute a solid strategy based on technology
development enhancing the industrial structure toward dynamic manufacturing industries. For
example, the output in ‘chemicals and chemical products’, ‘rubber and plastics’, ‘basic metals
and fabricated metal’, ‘machinery’, ‘electrical and optical equipment’ and ‘transport
equipment’ expanded significantly, mainly as a result of exports. ‘Electrical and optical
equipment’ and ‘transport equipment’ increased their outputs by 317.2% and 86.4%
respectively where exports corresponded to 222.1% and 69.7%. Not by coincidence, the
South Korean economic growth has been higher and much more sustainable when compared
to the Brazilian economy.

From 2003 until the outbreak of the global financial crisis in 2008, the pattern of
economic development for these two economies showed interesting movements. According to
Table 3.3, in South Korea exports of the manufacturing sector were maintained as an engine
of economic growth. However, the ‘flying geese’ movement of regional productive
fragmentation associated with the economic policy based on a paradigm of global value
chains negatively affected import substitution135. Although the country showed a higher
output expansion when compared to the Brazilian economy, the productive reallocation of
components and labour-intensive assembly-end parts to other countries within the Asian block
such as the 'third-tier' NICs (particularly China) slowed the pace of output growth during the
period.

As shown by Table 3.3, although the output growth rose by 39.1%, mainly pushed by
exports, the component of import substitution negatively affected the manufacturing sector by
a factor of 8.6%. From an intra-sectoral view, despite the consolidation of medium- and high-
tech manufacturing subsectors as engines of economic dynamism, the same industries were
badly affected with a negative variation of import substitution. In this aspect, notable
examples were ‘Coke, Refined Petroleum and Nuclear Fuel’, ‘Chemicals and Chemical
Products’, ‘Rubber and Plastics’, ‘Other Non-Metallic Mineral’, ‘Basic Metals and Fabricated
Metal’, ‘Machinery’, and ‘Transport Equipment’. In short, the economic dynamism of the
South Korean economy was not higher because of deleterious effects of imports on domestic
production.

135
For a detailed explanation about the ‘flying geese’ paradigm, see chapter 2. See also Ozawa (1991, 1995).

119
 Table 3.3 - Inter and Intra-sectoral decomposition of output growth (%), 2003-2008
South Korea Brazil
Sectors (∆U) (∆T) (∆R) (∆S) (∆Q) (∆U) (∆T) (∆R) (∆S) (∆Q)
Primary Sector 117.5 -118.5 11.7 4.3 15.0 9.2 -11.8 13.7 10.2 21.3
Manufacturing 4.8 -8.6 3.9 39.0 39.1 3.1 -8.2 26.8 0.5 22.2
Food, Beverages and Tobacco 3.9 -3.3 3.0 3.9 7.4 1.6 -1.6 13.7 1.2 14.9
Textiles and Textile Products 2.7 0.9 13.5 -13.8 3.3 -6.9 -5.0 29.5 -3.3 14.3
Leather, Leather and Footwear -1.3 0.7 -19.0 -10.3 -29.9 -1.3 -0.5 14.3 -18.0 -5.6
Wood and Products of Wood
and Cork 2.3 -1.9 8.1 9.3 17.7 -2.8 -2.1 16.1 -24.1 -12.9
Pulp, Paper, Paper , Printing
and Publishing -9.8 -4.0 3.9 15.2 5.4 -1.0 -5.1 25.4 0.4 19.7
Coke, Refined Petroleum and
Nuclear Fuel 27.3 -20.8 1.3 28.1 35.9 2.9 -11.6 15.4 6.6 13.4
Chemicals and Chemical
Products -11.2 -9.6 4.9 32.8 16.9 6.0 -16.6 21.5 -0.4 10.5
Rubber and Plastics 3.7 -6.6 5.0 37.2 39.4 -0.6 -9.8 27.4 2.4 19.5
Other Non-Metallic Mineral 5.7 -7.2 6.5 19.4 24.4 3.4 -4.9 30.1 -0.8 27.8
Basic Metals and Fabricated
Metal 8.1 -18.3 4.7 33.5 28.1 0.5 -11.3 26.8 -1.3 14.6
Machinery 17.5 -7.7 5.6 49.2 64.5 3.4 -6.6 49.6 3.2 49.5
Electrical and Optical
Equipment -0.6 3.9 5.5 88.3 97.1 13.2 -16.5 44.4 -0.3 40.8
Transport Equipment 0.2 -4.9 -2.1 59.3 52.4 11.4 -6.9 56.5 6.0 67.0
Manufacturing, Nec;
Recycling 19.6 -2.3 14.8 2.4 34.4 -1.9 -1.8 27.2 -3.4 20.1
Services 4.8 -5.2 12.4 8.3 20.2 4.8 -2.8 24.1 1.0 27.1
Source: Own calculations based on data from WIOD.
Note: Data calculated in 2008 US constant dollars. Nec = Not elsewhere classified.

In Brazil, inter and intra-sectoral results indicates a different pattern of economic

growth but also some common trends. While in South Korea the pattern of economic
development was in both periods (before and after 2003) driven by exports of manufacturing
products, in Brazil the domestic demand from 2003 to 2008 composed the key variable of
economic growth. In this period, the so-called commodity prices boom and the credit-led
consumption-boom boosted the economy. The former affected the economy promoting the
exports of primary products, while the latter pushed the domestic demand as a whole.
However, even in this context, once again manufacturing performed worse when compared to
the services sector, which showed the most dynamic output growth. Not even the Brazilian
National Development Bank (BNDES) providing many benefits for national producers, such
as funding with very low interest rates and certain benefits to stimulate exports (especially for
producers that use domestic inputs), was capable of reversing this trend. In an extreme
adverse macroeconomic policy marked by high and volatile interest and exchange rates, the
industrial policy became ineffective.

120
 Moreover, a comparative intra-sectoral analysis between Brazil and South Korea
reveals that low-tech manufacturing subsectors such as ‘leather, leather and footwear’ and
‘wood and products of wood and cork’ (in the case of Brazil) showed a negative output
variation while ‘machinery’, ‘electrical and optical equipment’ and ‘transport equipment’
expanded significantly in both economies. Finally, another common trend between Brazil and
South Korea was the clear negative variation of import substitution in the total output. In both
economies, the demand has been met by imports of either capital goods or intermediate
inputs. Seeking to evaluate this common trend in detail next, the section turns to this issue.

4. The density of industrial chains

Data regarding import input coefficients for Brazil and South constitutes an important
indicator to analyse the density of domestic industrial chains. Thus, through data available at
the World Input-Output Database (WIOD) and some input-output techniques, it is possible to
measure the evolution of such coefficients over time136. In this way, Tables 3.4 and 3.5
elucidate that in each economy these coefficients followed a distinct trend marked by
contrasting economic policies and strategies of development137. In South Korea, over the
years the process of industrialisation was characterised by a constant expansion of
manufacturing and gradual density increase of the domestic industrial chains, i.e., relative
increase of domestic inputs in the productive process vis-à-vis imported ones. However, as
shown by Table 3.4, from 1995 to 2002 this process was accompanied by an increasing share
of imported inputs in the domestic manufacturing sector that were consolidated in the
subsequent period.

As a matter of fact, from 2003 to 2008, the South Korean economy significantly
increased the share of imported inputs not only in the manufacturing sector but also in the
economy as a whole. The reasons for that are twofold. The first relates to the new productive
environment marked by global value chains, where the productive structure has been
gradually fragmented over the years. From this perspective, productive fragmentation is seen
as an important source of firm efficiency through improvements in competitiveness both in
domestic and international markets and by cost savings. The second reason relates to the

136
Technical details about methodology are summarised in Appendix 3.1.
137
Rocha (2011) analysed the participation of imported input coefficients only in the Brazilian industrial chains.
Additionally, in this study the analysis adopted the IBGE database.

121
 ‘flying geese’ process of industrialisation, where the productivity capacity at a certain stage of
the economic development is transferred to the less-developed country (particularly Nics-2
and China) in a process of regional fragmentation of the productive chain. Although both
reasons are associated with the concept of productive fragmentation, they might be seen as
different sides of the same coin. It is because the first reason is related with cost savings per
se and a premature process of de-industrialisation in most developing countries, while the
second one relates to a regional strategy of development and the expansion of manufacturing
in the economy. Since South Korea expanded not only the use of imported inputs in its
productive structure but also the share of manufacturing, the second reason becomes seems
likely.

Table 3.4 - Import input coefficients for Korea, 1995-2008

South Korea
Sectors 95 96 97 98 99 00 01 02 03 04 05 06 07 08
Primary Sector 14.6 14.4 13.6 11.5 11.8 13.1 12.0 11.5 12.0 12.9 14.7 16.1 16.7 19.5
Manufacturing 31.2 32.5 31.8 27.0 27.9 30.9 27.8 26.9 29.1 32.7 35.7 38.5 39.4 44.2
Food, Beverages and
Tobacco 25.4 25.7 23.2 19.2 19.8 20.3 18.8 18.3 19.0 20.6 22.3 23.5 24.6 29.1
Textiles and Textile
Products 40.2 39.7 38.4 35.2 33.8 35.0 31.6 28.3 29.3 30.2 31.2 31.4 31.0 30.4
Leather, Leather and
Footwear 30.3 28.7 29.4 24.9 23.8 26.7 27.2 24.9 23.9 25.9 26.9 28.3 29.2 29.0
Wood and Products of
Wood and Cork 45.9 44.8 43.1 30.4 33.2 34.3 31.1 32.0 31.2 33.6 34.9 36.3 38.4 39.2
Pulp, Paper, Paper ,
Printing and Publishing 22.7 22.6 21.7 20.9 19.8 23.5 19.7 18.3 19.1 21.4 22.8 23.5 24.5 28.0
Coke, Refined
Petroleum and Nuclear
Fuel 28.1 31.9 31.0 22.9 25.7 37.6 32.9 30.8 36.3 40.7 51.0 61.2 63.8 82.0
Chemicals and
Chemical Products 29.7 28.8 27.7 23.3 23.5 28.0 25.4 23.1 26.3 31.8 35.8 39.5 41.4 48.1
Rubber and Plastics 28.7 29.3 28.0 22.9 23.7 25.5 23.7 23.4 25.3 28.5 32.3 34.3 34.9 37.4
Other Non-Metallic
Mineral 31.0 31.5 29.9 25.0 24.7 25.4 24.1 24.2 25.5 26.7 29.3 30.7 29.8 32.5
Basic Metals and
Fabricated Metal 26.4 27.9 27.8 24.4 23.8 23.9 21.7 22.0 25.0 31.5 33.5 38.0 40.8 48.8
Machinery 34.4 36.3 35.5 29.7 30.9 29.5 27.5 27.8 29.5 32.1 35.1 36.8 37.2 36.8
Electrical and Optical
Equipment 83.2 87.4 86.7 67.7 62.1 60.6 52.7 48.1 47.0 46.3 46.5 43.8 41.7 41.1
Transport Equipment 25.3 26.4 27.3 24.4 25.3 25.6 23.9 24.5 25.0 27.0 29.2 31.3 32.6 34.4
Manufacturing, Nec;
Recycling 31.7 32.6 30.4 24.9 26.6 28.3 26.0 26.2 26.9 29.8 32.0 34.2 34.5 32.2
Services 11.5 12.1 12.2 9.6 10.2 12.6 11.3 10.8 11.8 13.4 15.3 16.8 17.5 19.2
Source: Own calculations based on data from WIOD.
Note: Data calculated in 2008 US constant dollars. Nec = Not elsewhere classified.

In the Brazilian economy, unlike in South Korea, the manufacturing sector fell apart
during the so-called lost decade in the 1980s. Moreover, the restrictive neoliberal economic

122
 policy grounded in market-friendly mechanisms led to a process of premature de-
industrialisation without an end in sight. In this dynamic, industrial restructuring involved
mainly defensive adjustments in a context where investments were kept to a minimum level.
Thus, the productive fragmentation seeking cost savings was widely adopted without any
long-term economic planning. Additionally, under high and volatile interest and exchange
rates, in periods of demand expansion (as during the 2000s), substitution of national inputs for
imported ones constituted the easiest way to attend the demand. By analysing the import input
coefficients, it is possible to realise the dependence of the Brazilian domestic industry in
relation to the international supply and the progressive loss of density in its industrial chains.
A gradual shift to free-market policies intensified the presence of imported inputs for the
domestic industry.

Table 3.5 - Import input coefficients for Brazil, 1995-2008

Brazil
Sectors 95 96 97 98 99 00 01 02 03 04 05 06 07 08
Primary Sector 3.7 4.0 4.1 4.1 3.9 4.5 4.6 4.4 4.6 5.3 5.6 5.9 7.1 9.4
Manufacturing 6.9 7.4 7.6 7.4 6.9 8.5 8.5 7.5 7.8 9.0 9.9 11.1 12.7 16.1
Food, Beverages and
Tobacco 4.7 5.7 5.9 5.5 4.9 5.6 5.1 4.6 5.0 5.0 5.5 6.0 7.2 9.6
Textiles and Textile
Products 6.5 5.7 6.2 6.2 5.6 6.4 6.2 5.7 5.5 6.0 6.3 7.6 9.1 10.5
Leather, Leather and
Footwear 7.1 6.8 7.3 7.1 6.5 7.7 7.5 6.3 6.1 7.2 6.8 7.6 9.2 10.9
Wood and Products of
Wood and Cork 2.7 2.9 3.1 3.1 2.9 3.5 3.5 3.5 3.7 4.7 4.9 5.2 6.3 8.2
Pulp, Paper, Paper ,
Printing and Publishing 7.6 7.6 8.1 8.3 7.4 8.8 8.2 7.1 6.5 7.0 7.5 7.8 9.3 10.8
Coke, Refined
Petroleum and Nuclear
Fuel 10.0 11.7 10.5 8.2 8.3 11.5 11.2 9.8 10.3 14.1 15.9 18.9 21.6 26.5
Chemicals and
Chemical Products 6.1 7.0 7.2 6.9 6.8 9.4 9.3 8.5 9.4 11.3 10.9 11.4 13.9 18.9
Rubber and Plastics 8.4 8.7 9.1 9.2 9.1 11.1 10.6 9.2 10.3 11.6 12.3 13.1 15.3 18.5
Other Non-Metallic
Mineral 6.9 6.6 6.9 7.2 6.3 6.9 7.8 6.0 6.5 6.5 6.8 8.0 9.5 12.7
Basic Metals and
Fabricated Metal 6.6 6.1 6.5 6.7 6.1 6.5 6.7 6.5 6.3 7.2 8.8 10.6 12.2 15.6
Machinery 7.2 7.1 7.8 8.2 7.4 8.7 9.4 8.2 8.3 8.8 10.1 11.5 12.8 15.2
Electrical and Optical
Equipment 8.7 9.2 9.8 10.0 9.9 12.9 12.7 12.1 11.4 13.3 15.3 17.2 16.2 21.8
Transport Equipment 8.7 8.9 9.9 10.5 10.0 11.8 13.2 11.5 11.5 12.5 14.0 15.2 16.6 19.8
Manufacturing, Nec;
Recycling 5.9 5.9 6.3 6.6 6.1 7.0 6.7 5.7 5.9 6.5 6.7 7.3 8.3 10.6
Services 3.0 3.2 3.4 3.5 3.1 3.8 3.8 3.3 2.9 3.1 3.6 4.0 4.5 5.4
Source: Own calculations based on data from WIOD.
Note: Data calculated in 2008 US constant dollars. Nec = Not elsewhere classified.

123
 From 1995 to 1999, the rising trend of import input coefficients appeared blurred.
However, from this period onwards these coefficients raised almost constantly. Furthermore,
particularly after 2002, this process gained speed essentially due the economic recovery
(pushed by the commodity boom) and the structural mismatch between the structures of
supply and demand. The most representative examples of subsectors that showed a substantial
rising trajectory of imported inputs were ‘coke, refined petroleum and nuclear fuel’,
‘chemicals and chemical products’, ‘rubber and plastics’, ‘electrical and optical equipment’
and ‘transport equipment’. Additionally, it is important to note that the most affected
manufacturing subsectors were those which formed part of the medium- and the high-tech
groups. In this sense, it is not by chance that the Brazilian economy has not structured the
basis for a long-term economic growth. In fact, the only achievement of the Brazilian
economic policy has been the establishment of successive short cycles of ‘stop and go’,
arising from a decreasing share of manufacturing in the value-added and the disarticulation of
domestic industrial chains especially in medium- and high-tech subsectors.

5. Concluding remarks

This empirical chapter sought to provide an understanding of the structural transition

undertaken by Brazil and South Korea from a comparative perspective. Therefore, the first
section, through the methodology of Imbs and Wacziarg, analysed the pattern of inter and
intra-sectoral diversification presented by these economies before and after the 1980s. Firstly
the data showed that Brazil was affected by a process of premature specialisation at the inter-
and the intra-sectoral levels. Secondly, the intra-sectoral data of manufacturing revealed that
during the specialisation stage South Korea increased sharply the share of medium- and high-
sectors in the economy, while Brazil, although more diversified, remained dependent on less
dynamic sectors. From a comparative perspective of economic development, this dichotomy
propelled different processes of cumulative causation and, consequently, impacted the
countries’ growth rates distinctly. Therefore, it is not by coincidence that the constant drop of
the Brazilian share in the global manufacturing was followed by a regressive evolution in the
global GDP. Consequently, Brazil remained trapped in a middle-income landscape resulting
in an unsustainable economic performance.

In the second and third section, the analysis turned to the understanding of factors
(demand, trade and technological change) that contributed to the formation of these

124
productive structures in light of liberalising policies implemented during the second half of
1990s. Thus, from 1995 to 2002 during the height of neoliberal reforms, Brazil increased its
manufacturing output by only 9.1%. In this period, with the exception of exports, all
components registered negative variations. In a distinct economic dynamism, the South
Korean manufacturing production increased by 58.3% of which 50.7% was explained by
exports. Furthermore, this expansion was composed mainly by medium- and high-tech
industries. In the subsequent period, from 2003 to 2008, South Korea maintained
manufacturing as an engine of economic growth but with an significant increase of imports
mainly due to the ‘flying geese’ movement of regional productive fragmentation marked
specially by the rise of China as the world's factory. While the South Korean pattern of
economic development was in both periods (before and after 2003) driven by exports of
manufacturing products, in Brazil from 2003 to 2008 economic growth was mainly propelled
by domestic demand fuelled by the credit-led consumption of the period. In this dynamic, the
so-called commodity prices boom also played a central role to boost the economy. However,
even in this context, once again manufacturing performed worse when compared to services,
which presented the most dynamic output growth.

In the fourth section, the density of domestic industrial chains through imported input
coefficients was measured. In South Korea, from 1995 to 2002, the productive structure
sketched movements related to an increasing share of imported inputs in the domestic
manufacturing sector that were consolidated in the subsequent period. From 2003 to 2008, the
South Korean economy significantly increased the share of imported inputs not only in the
manufacturing sector but also in the economy as a whole. This finding is consistent with the
aforementioned ‘flying geese’ process of industrialisation where the productivity capacity at a
certain stage of the economic development is transferred to the less developed country (in this
case particularly China) in a process of regional fragmentation of the productive chain. In
contrast, Brazil did not follow this pattern of economic development. As a matter of fact, the
manufacturing sector fell apart during the 1980s and 1990s in a process of premature de-
industrialisation without an end in sight. In this productive disarticulation, many
manufacturing industries kept investments to a minimum level. Furthermore, the productive
fragmentation seeking cost savings per se was widely adopted without any long-term
economic planning. Additionally, under high and volatile interest and exchange rates, in
periods of demand expansion (as during the 2000s), substitution of national inputs for
imported ones constituted the easiest way to meet the demand. This fact not only revealed the

125
dependence of the Brazilian domestic industry in relation to international supply but also the
absence of important intra-sectoral linkages.

126
Appendix 3.1 – Technical Details

Lowess Methodology and the U-shaped curve

Analogously to the procedure adopted in Imbs & Wacziarg (2003), the study employs a non-
parametric methodology based on locally robust weighted scatter plot smoothing
(LOWESS)138. In this procedure, each estimate  is a function of the G ’s in the
neighbourhood to be estimated, with the closest observations receiving more weight than the
more distant observations. The LOWESS smoother requires specification of a span, which is
the percentage or total number of observations to be included in the estimated regression at
observation V. In this study the span adopted is 0.60 which means that 60 of total observations
are used in each local regression. Moreover, the LOWESS uses a tricubic weighting function
with the following formula:

]
[2[\ ]
wW = X1 − Z ([)
Z ^ (3.1)

Where  is the smoothed response, G are the nearest neighbours of  as defined by the
span, and _() is the distance along the abscissa from  to the most distant predictor value
within the span. In the present study, as well as in Imbs & Wacziarg (2003), the response
variable ` corresponds to a measure of sectoral specialisation, while the independent variable
 is the GDP per capita of each country (in 2005 dollars). The observations of  and ` are the
annual values of each indicator for the country under consideration. In this sense, the  values
estimated by the nonparametric local regression will give form to a smoothed curve (U-
curved) that connects  to ` and represents the relationship between specialisation and GDP
per capita. The measure of sectoral specialisation employed is the Gini-Hirschman
index (GHI) which is obtained by taking the square root of the Hirschman-Herfindahl Index
(HHI) and multiplying it by a factor of one hundred. These indexes are defined as follows:

[ h
HHI = ∑iWj; d [f\e g (3.2)

[ h
GHI = 100. m∑iWj; d [f\e g (3.3)

138
This method was first proposed by Cleveland (1979) and further developed by Cleveland & Devlin (1988).

127
 In these equations, xij is the value-added of sector i and x is the total value-added of the
country j and n is the number of sectors. Therefore, the higher the GHI is, the more
specialised the economy is. The GHI is within a certain limit value. The maximum value of
the index is 100 and in this case there is only one sector in the economy.

Decomposition of output growth

Analogously to the method used in Chenery (1960) and Chenery et al. (1962), the so-
called Structural Decomposition Analysis (SDA) is applied following Miller and Blair’s
(2009) approach. The total output is defined as:

Q = (n − o )2; (R + S) (3.4)

Where, o is the matrix of domestic inputs, R is the domestic demand of national goods and
S is the exports. Considering the basic Leontief model for two periods p and +ℎ , the vector
of gross output () can be written as follows:

 4@ = r4@ (R4@ + S 4@ ) and   = r (R + S  ) (3.5)

Where the matrix r(ss) represents the Leontief matrix of direct and indirect production
coefficients defined as r = (n − o )2; . Thus, the change in gross output between two periods
p and +ℎ , can be rewritten as:

t =  4@ −   = r4@ (R4@ + S 4@ ) − r (R  + S  ) (3.6)

Then through basic matrix algebraic manipulation, changes in r and D + E, and

consequently effects on ∆ are represented as:

t = r4@ (R  + S  + tR + tS ) − (r4@ − tr)(R  + S  ) (3.7)

t = (r + tr)(R 4@ + S 4@ ) − r4@ (R4@ + S 4@ − tR − tS) (3.8)

Additionally, the equations are summarised according to the average approach. As

argued by Dietzenbacher and Los (1998) this method enhances the SDA methodology. Thus,
summing equations (3.7) and (3.8), and applying the average approach, the following
equation is obtained:

; ; ;
t = h (tr)(R + R 4@ + S  + S 4@ ) + h (r + r4@ )(tR) + h (r + r4@ )(tS) (3.9)

128
Where the first term indicates the effects of the change in the Leontief coefficients over the
change in gross output, the second term refers to the effects of the change in domestic
demand, and the third, exports.

Assuming o = o − o, , where o is the matrix of total inputs and o, is the matrix of

imported inputs, equation 3.9 can be re-written as:

; ; ;
t = (r + r4@ )(tR ) + (r + r4@ )(tS ) + wr4@ (−to, )r ](R  + R 4@ + S  +
h h h
;
S 4@ ) + h wr; (to)r ](R  + R4@ + S  + S 4@ ) (3.10)

The four components of total output change for any sector or subsector over the period
ℎ are thus calculated as follows:

;
Domestic demand (∆R) = h (r + r4@ )(tR + tS )

;
Export (∆S) = h (r + r4@ )(tS )

;
Import substitution (∆T) = wr4@ (−to, )r ](R + R 4@ + S  + S 4@ )
h

;
Technological change (∆U) = h wr4@ (to)rx ](R + R 4@ + S  + S 4@ )

The density of industrial chains

Through the input-output framework, Rocha (2011) decomposed import coefficients

between inputs and final goods. In this analysis, import input coefficients by sector are
defined as nny and the total import input coefficient for the economy as TIIC.

Firstly define o, (D|D) as the matrix of direct import input coefficients:

}~
o, = |
(3.11)

where D|;is the vector of gross output and D|D

€
is the matrix of intermediate consumption.

Considering the Leontief matrix of direct and indirect production coefficients defined
as r = (n − o )2; and multiplying it by y € the following equation is obtained:

nny € = o, (n − o )2; (3.12)

129
 nny€ = ∑Dj; ynnG€ (3.12)

|
nnyB)‚
€
= ∑Dj; dynn€ . |ƒ g (3.12)

where V and „ indicate each economic sector (seller and buyer, respectively) and s is the total
of these sectors.

130
 Appendix 3.2 – Map of sectoral aggregation (UNIDO)
Aggregation – 14 Subsectors
Food, Beverages and Tobacco
Food, Beverages and Tobacco
Textiles and Textile Products
Textiles and Textile Products
Leather, leather products and footwear
Wood and Products of Wood and Cork
Pulp, Paper, Paper , Printing and Publishing
Pulp, Paper, Paper , Printing and Publishing
Coke, Refined Petroleum and Nuclear Fuel
Chemicals and Chemical Products
Rubber and Plastics
Other Non-Metallic Mineral
Basic Metals and Fabricated Metal
Basic Metals and Fabricated Metal
Machinery
Machinery
Electrical and Optical Equipment
Machinery
Electrical and Optical Equipment
Transport equipment
Transport equipment
Manufacturing, Nec; Recycling
Manufacturing, Nec; Recycling
Total manufacturing
Source: Elaborated by the author.
Note: Nec = Not elsewhere classified.
UNIDO Database – 23 subsectors
Food and beverages
Tobacco products
Textiles
Wearing apparel, fur
Leather, leather products and footwear
Wood products (excl. furniture)
Paper and paper products
Printing and publishing
Coke, refined petroleum products, nuclear fuel
Chemicals and chemical products
Rubber and plastics products
Non-metallic mineral products
Basic metals
Fabricated metal products
Machinery and equipment n.e.c.
Office, accounting and computing machinery
Electrical machinery and apparatus
Radio, television and communication equipment
Medical, precision and optical instruments
Motor vehicles, trailers, semi-trailers
Other transport equipment
Furniture; manufacturing n.e.c.
Recycling
Total manufacturing
131
Appendix 3.3 – Map of sectoral aggregation (ECLAC-PADI)

Aggregation – 14 Subsectors ECLAC-PADI Database – 28 subsectors

Food, Beverages and Tobacco Food
Food, Beverages and Tobacco Beverages
Food, Beverages and Tobacco Tobacco
Textiles and Textile Products Textiles
Textiles and Textile Products Wearing apparel
Leather, Leather and Footwear Leather products
Leather, Leather and Footwear Footwear
Wood and Products of Wood and Cork Wood products
Wood and Products of Wood and Cork Furniture
Pulp, Paper, Paper , Printing and Publishing Paper and cellulose
Pulp, Paper, Paper , Printing and Publishing Printing and publishing
Chemicals and Chemical Products Chemical industry
Chemicals and Chemical Products Other chemicals
Coke, Refined Petroleum and Nuclear Fuel Refined petroleum
Coke, Refined Petroleum and Nuclear Fuel Petroleum and carbon products
Rubber and Plastics Rubber products
Rubber and Plastics Plastic products
Other Non-Metallic Mineral Ceramic
Other Non-Metallic Mineral Glass
Other Non-Metallic Mineral Other non-metallic minerals
Basic Metals and Fabricated Metal Iron and Steel
Basic Metals and Fabricated Metal Nonferrous metals
Basic Metals and Fabricated Metal Metal products
Machinery Nonelectric Machinery
Electrical and Optical Equipment Electric Machinery
Transport Equipment Transport Equipment
Electrical and Optical Equipment Professional and scientific instruments
Manufacturing, Nec; Recycling Other Manufacturing
Total manufacturing Total Manufacturing
Source: Elaborated by the author.
Note: Nec = Not elsewhere classified.

132
 Appendix 3.4 – Classification of manufacturing subsectors by technology group

Aggregation – 14 Subsectors Technology group (UNIDO)

Food, Beverages and Tobacco Low-Tech
Textiles and Textile Products Low-Tech
Leather, Leather and Footwear Low-Tech
Wood and Products of Wood and Cork Low-Tech
Pulp, Paper, Paper , Printing and Publishing Low-Tech
Coke, Refined Petroleum and Nuclear Fuel Medium-Tech
Chemicals and Chemical Products High-Tech
Rubber and Plastics Medium-Tech
Other Non-Metallic Mineral Medium-Tech
Basic Metals and Fabricated Metal Medium-Tech
Machinery High-Tech
Electrical and Optical Equipment High-Tech
Transport Equipment High-Tech
Manufacturing, Nec; Recycling Low-Tech
Source: Elaborated by the Author based on Unido (2013).
Note: Nec = Not elsewhere classified.

133
CHAPTER 4 – ENGINES OF GROWTH AND
SECTORAL INTERDEPENDENCE IN BRAZIL
AND SOUTH KOREA

“Ordinarily economists have been content with general

references to the advantages of external economies, complementarities,
cumulative causation, etc. But no systematic effort has been made to
describe how the development path ought to be modified so as to
maximize these advantages even though the existence of input-output
statistics supplies us with a few tools of an analysis of this kind"
(Hirschman 1958, p. 100).

Introduction139

During the neoliberal decades, economic growth between Brazil and South Korea has been
remarkably distinct. In Brazil, although it peaked from 2002 until the outbreak of the global
financial crisis, the economy has been marked by successive “stop and go” cycles. In this
dynamic, the performance of manufacturing has been poor, while services have been
increasing as a share of GDP. The former decreased to 16% and the latter increased to 70% in
2010. From a comparative perspective, South Korea has engendered an economic growth
much more sustained than Brazil. Except for the year of 1998, due to the Asian financial
crisis, South Korea has shown a consistent trajectory of economic growth. In this process,
manufacturing expanded continually reaching 35% of GDP in 2010140. These divergent
dynamics of economic growth and sectoral structures are likely to be closely related to each
other141.

In the developmental literature, manufacturing is traditionally regarded to play a

special role as engine of economic growth. In this sense, poor manufacturing performance
would imply in deleterious effects in sustainable economic growth. The Structuralist-
Kaldorian rationale behind this dynamic is the existence of increasing returns to scale that is
at the heart of cumulative causation processes. Additionally, learning by doing, spill-over
effects and intra-sectoral linkages are considered particularly stronger in manufacturing. In a
139
I would like to thank Guilherme Magacho and Joaquim Guilhoto for their helpful comments on this chapter.
140
Data calculated in current US dollars.
141
The loss of the Brazilian economic dynamism is associated to the process of negative de-industrialisation. For
a detailed distinction between the process of natural and the so-called negative de-industrialisation, see chapter 1
(section 2.2).

134
Schumpeterian sense, manufacturing has been also recognised as the main engine of
technological dynamism and specifically a locus of innovation. From this perspective, it is
argued that technological change is mainly triggered and diffused from the manufacturing
sector.

Nevertheless, recently the capacity of services to act as an engine of economic growth

has attracted considerable interest in academic and policy circles. According to this type of
narrative, services and manufacturing have been increasingly combined through a synergistic
and symbiotic relationship that raised the share of services in the economy (Pilat & Wolfl,
2005; Pilat et al. 2006; Lodefalk, 2010; Nordås & Kim, 2013). Additionally, many studies
have shown that manufacturing has been using more intermediate services over the years. In
this sense, the share of services in manufacturing value-added has been increasing steadily,
surpassing 60% in middle-income countries and 65% in high-income countries (CNI, 2014).
Although these inter and intra-industrial relationships reveal an increasing importance of
services in the economy, their role is still more limited than those of the manufacturing sector
(Park & Chan, 1989).

Therefore, in this context, the chapter uses principally input-output methodologies to

compare the Brazilian and South Korean productive structures in terms of productive
dynamisms and leading sectors. Moreover, aspects regarding the interdependence of the
manufacturing and services are also investigated. The contributions of this chapter are
twofold. First, the analysis employs the input-output methodology to identify key sectors in
both economies. This analysis provides a useful framework to understand specificities and
regularities of each economy and methodological referential for further empirical studies
about sectoral specificity of growth. Second, this chapter estimates the interdependence
between services and manufacturing through a new input-output methodology. Therefore, this
study advances some findings presented by the literature and opens an interesting window for
future research.

The chapter is organised as follows. Section 1 provides a brief discussion about the
relevance of industrial linkages for economic growth, based mainly on Hirschman’s theory.
Section 2, from a comparative analysis between Brazil and South Korea, measures the output
multipliers, Hirschman-Rasmussen backward and forward indices, as well as pure normalised
indices comprising all productive subsectors of manufacturing, services and the primary

135
sector. Section 3 analyses the interdependence between manufacturing and services.
Concluding remarks are provided in Section 4.

1. The relevance of industrial linkages: a Hirschmanian perspective

Albert Hirschman through the study The Strategy of Economic Development (1958)
provided the first solid critique of the ‘balanced growth’ approach proposed by Nurkse and
Rosenstein-Rodan142. According to him, the ‘balanced growth’ theory is not only impractical
but also does not capture the nature of the development process. Therefore, as stated by
Hirschman (1958), economic growth is essentially an unbalanced dynamic process, in which
successive disequilibria produce the conditions for development in different sectors. In this
sense, growth proceeds as a ‘series of uneven advances of one sector followed by the
catching-up of other sectors’ (ibid, p.63).

As pointed out by Toner (1999), Hirschman made a number of important contributions

to the theory of economic development. First, Hirschman detailed the industrial structure to
explain the development process in an intra-sectoral analysis. Second, input-output methods
were adopted to analyse sectors and subsectors. Third, input-output multipliers were seen as
providing a means of quantifying pecuniary externalities arising from
demand and supply interactions. Last but not least, the orientation of final demand by the state
was viewed as the best way to alter the industrial structure by creating strong inducements to
domestic investments in selected manufacturing subsectors. The criterion for such decisions
was based on the capacity of a given subsector to generate externalities which act to induce
investments in other subsectors in a circular and cumulative chain. In this dynamic,
manufacturing plays a central whole since the primary sector has very limited linkages with
other sectors.

In Hirschman’s thinking, the state is a central agent to direct investments toward

dynamic manufacturing subsectors, which posses the strongest linkages with the rest of the
economy. In this framework, linkages are responsible for providing necessary means for the
expansion of domestic demand and supply. Thus, Hirschman defines an underdeveloped
economy by its absence of linkages or inter- and intra-sectoral interdependences.
Consequently, an underdeveloped economy is characterised by a very large leakage of

142
For a detailed explanation about these theoretical approaches see chapter 1.

136
demand for both intermediate and final goods into imports. Moreover, in an underdeveloped
industrial structure, cumulative expansion in demand and supply resulting from investments
in intermediate and capital goods are restricted due to the lack of dynamic industrial linkages.
Therefore, the main aim of the economic growth process is to develop an interdependent
industrial structure.

The extent of backward and forward linkages in a productive structure determines the
size of the stimulus to an economy though a given expansion in final demand. Hirschman, on
the one hand, explains that backward linkages are related to derived demand, i.e. the provision
of input for a given sector. According to him, every non-primary economic sector will induce
attempts to supply through domestic production the inputs needed in that sector. On the other
hand, forward linkages will induce attempts to utilise outputs as inputs in some new activities
(Hirschman, 1958, p.100). Tregenna (2009, p.143), argues that forward linkages can raise
overall economic growth, increasing the capacity utilisation, technological upgrading and
productivity in downstream sectors. However, Hirschman states that backward linkage is the
dominant form of linkage. In his words, “forward linkage could never occur in pure form. It
must always be accompanied by backward linkage which is the result of the pressure of
demand. In other words, the existence or anticipation of demand is a condition for forward
linkage effects to manifest themselves”. Unlike backward linkages, forward linkages cannot
therefore be regarded as an independent inducement mechanism (Hirschman, 1958, p.116-
117).

In this framework, manufacturing is referred to as a sector with special properties in

terms of strong backward and forward linkages. Hirschman states that the significance of
manufacturing for development is that there is a close correlation between income per capita
and the proportion of people occupied in manufacturing. Additionally, in his view, from a
growth perspective manufacturing is superior to the primary sector not as a result of higher
productivity but rather because linkages that agriculture and mining have to the rest of the
economy are lower than those that manufacturing has. In this way, the author points out that,
for instance, mining in underdeveloped countries usually culminates in an enclave problem
since “exports slip out the country without leaving much of a trace in the rest of the
economy143” (ibid, p.110).

143
Hirschman does not mention the services sector in this analysis.

137
 Based on this theoretical background, the following section presents the forward and
backward linkages for the Brazilian and South Korean economies as a whole, i.e. covering the
primary sector, manufacturing and services. The study used input-output analyses to
investigate the capacity of sectors and subsectors to promote economic growth. This
methodology is useful because these models incorporate inter-relationships between various
sectors in terms of linkages. Using this method, it is possible to empirically investigate the
economic role of a productive sector without restricting the analysis to its “direct effects” on
the economy regarding generating production, employment, value-added, tax revenue, and
exports. This analysis, also considers “indirect effects”, i.e. effects that a given sector exerts
over other sectors through channels established by transactions of inputs and outputs between
different economic sectors.

2. Output Multipliers and Hirschman-Rasmussen Linkages144

2.1.Output Multiplier

The analysis of engines of growth in Brazil and South Korea is assessed through the
use of input-output methods. Initially, simple output multipliers and Hirschman-Rasmussen
indices are calculated and analysed. The output multiplier indicates how much is produced for
each monetary unit spent on final demand. These multipliers incorporate direct and indirect
effects to measure the impact of a demand shock on the economy. For the purposes of this
analysis, multipliers are used only considering linkage effects restricted to demand for
intermediate inputs, that is, without making household demand endogenous to the
model145.Moreover, the data was compiled in two periods, namely 1995-2002 and 2003-2009,
for both countries.

As shown by Table 4.1, the output multipliers are significantly different among
countries both in terms of absolute and relative magnitudes. From 1995 to 2002, South Korea
presented higher output multipliers for the majority of subsectors, particularly in
manufacturing. The exceptions were ‘Mining’, Petroleum and Nuclear Fuel’ and a few
services subsectors such as ‘Financial Services’, ‘Business Services’, ‘Public Admin’, ‘Health
and Education’, ‘Utilities’ and ‘Construction’. In the subsequent period, i.e. from 2003 to
144
Technical details on methodology are summarised in Appendix 4.1.
145
This type of multiplier is recognised as type I. Models that present household demand endogenised in the
system and induced effects provide type II multipliers (Guilhoto, 2009).

138
2009 South Korea increased its output multipliers with only four sectors with lower
multipliers than Brazil. In this sense, only ‘Mining’, Petroleum and Nuclear Fuel’, ‘Utilities’
and ‘Transport Services’ had a higher output multiplier in Brazil.

Table 4.1 - Output Multipliers

South Korea Brazil
1995-2002 2003-2009 1995-2002 2003-2009
Subsectors Mult Rank Mult Rank Mult Rank Mult Rank
Agriculture 1.67 18 1.71 16 1.59 19 1.67 17
Mining 1.44 23 1.55 22 1.50 23 1.66 19
Food and Beverages 2.21 8 2.26 5 2.13 2 2.26 2
Textiles 2.38 5 2.18 8 1.94 9 1.93 9
Leather and Footwear 2.45 3 2.33 4 2.15 1 2.14 4
Wood and Products of Wood 2.05 12 2.04 13 1.66 18 1.88 14
Paper , Printing and Publishing 2.20 9 2.15 10 2.01 7 1.92 11
Petroleum and Nuclear Fuel 1.17 25 1.21 25 2.12 3 2.27 1
Chemicals 2.09 11 2.08 12 1.82 13 2.05 7
Rubber and Plastics 2.20 10 2.20 6 1.94 10 2.06 6
Other Non-Metallic Mineral 2.30 7 2.11 11 1.88 12 1.93 10
Basic Metals and Fabricated Metal 2.02 13 2.18 7 1.72 14 1.90 13
Machinery 2.52 2 2.39 3 2.04 6 2.10 5
Electrical and Optical Equipment 3.07 1 2.16 9 2.05 5 1.98 8
Transport Equipment 2.44 4 2.53 1 2.11 4 2.24 3
Manufacturing, Nec 2.36 6 2.40 2 1.97 8 1.91 12
Utilities 1.59 20 1.57 21 1.69 16 1.70 16
Construction 1.90 14 2.00 14 1.92 11 1.75 15
Sales 1.66 19 1.65 17 1.35 25 1.40 25
Traditional Services 1.73 15 1.74 15 1.51 22 1.51 23
Transport Services 1.70 16 1.62 19 1.55 20 1.66 18
Financial Services 1.67 17 1.64 18 1.69 15 1.56 21
Business Services 1.53 21 1.61 20 1.66 17 1.57 20
Public Admin 1.40 24 1.48 24 1.48 24 1.48 24
Health and Education 1.46 22 1.54 23 1.53 21 1.51 22
Source: Author's own elaboration based on WIOD tables.
Note: Data calculated in 2008 US constant dollars. ‘Mult’ = Output Multipliers and “Rank’ = Ranking. Nec =
Not elsewhere classified.

The aforementioned ‘landscape’ indicates that the South Korean subsectors are more
interrelated to each other than the Brazilian ones are and, consequently, an increase in final
demand in South Korea has a higher impact on the domestic economy than that in Brazil.
Additionally, as shown by Table 4.1, except for ‘Petroleum and Nuclear Fuel’, all
manufacturing subsectors in South Korea have a multiplier higher than 2.00. This numbers

139
indicates that for both periods, the South Korean manufacturing sector is able to more than
double its output given a unitary increase in demand. Distinctly, Brazil showed this dynamism
in only a few manufacturing sectors and with some shifts between 1995-2002 and 2003-2009.
Therefore, from 1995 to 2002, the subsectors with highest capacity (with a multiplier higher
than 2.00) to increase output in face of a demand rise were ‘Leather and Footwear’, ‘Food and
Beverages’, ‘Petroleum and Nuclear Fuel’, ‘Transport Equipment’, ‘Electrical and Optical
Equipment’, ‘Machinery’ and ‘Paper, Printing and Publishing’. From 2003 to 2009, the
subsectors of ‘Paper, Printing and Publishing’ and ‘Electrical and Optical Equipment’ left this
group while ‘Rubber and Plastics’ and ‘Chemicals’ showed output multipliers of 2.06 and
2.05 respectively.

2.2.Hirschman-Rasmussen Linkages

The results of the previous analysis of output multipliers can be complemented by the
Hirschman-Rasmussen Linkages (Figure 4.1). The seminal works of Hirschman (1958) and
Rasmussen (1956) gave rise to a wide range of studies attempting to investigate
interrelationships between sectors, i.e. backward and forward linkage effects in a given
economic structure146. The Backward Linkage (BL) index indicates the extent to which a
given sector demands from other sectors. In other words, it measures the capacity of a
hypothetical sector to stimulate the output of other sectors. The Forward Linkage (FL) index,
in turn, measures the extension to which a given sector is demanded by other sectors, i.e. the
importance of a hypothetical sector as an input supplier. In this sense, these indices allow the
investigation of productive structures in detail and identify sectors that depend on inter- and
intra-sectoral supply and inter- and intra-sectoral demand or those that are relatively
independent from other sectors147.

In this way, as shown by Figure 4.1, South Korean and Brazilian subsectors are
classified into the following four groups depending on their indices values: (i) located in the
lower-left are subsectors that expose both BL and FL indices lower than 1, and consequently

146
These indices were first presented by Hirschman (1958) with a particular interest in Latin American
economies. It was based on his experiences as an official advisor and private consultant in Colombia in the first
half of the 1950s (Hirschman, 1984).
147
The backward and forward linkages for Brazil and South Korea are presented in terms of averages from 1995
to 2009 since the focus of this chapter is the comparison between sectors/subsectors and not the tendency over
time. Moreover, subsectors did not expose significant changes in terms of quadrants. Tables for backward and
forward linkages for each year are presented in the appendices 4.2, 4.3, 4.4 and 4.5.

140
are highly independent from other subsectors; (ii) located in the upper-right are subsectors
with both linkages higher than 1, denoting subsectors that are strongly dependent on other
sectors in terms of both supply and demand of inputs; (iii) located in the lower-right are
subsectors that presents only BL indices greater than 1 and are strongly dependent on inter-
and intra-sectoral supply (or stimulate production in other sectors); and (iv) located in the
upper-left are subsectors that present only FL greater than 1 and are strongly dependent on
inter- and intra-sectoral demand (or dependent on the production of other sectors).

Figure 4.1 – Hirschman-Rasmussen BL and FL indices (Average 1995-2009)

South Korea Brazil

1.70 FL 1.70
Min FL

1.50 1.50
Min
Bm Ppp
Rp
1.30 1.30 Ut
Ut Wp Nmm Ppp
Nmm Bm
Bs
Ag Ch Wp Ch PnF
Rp 1.10 Ag
1.10 Fs Bs
Pnf BL Sal Trs BL
Fs
Tex
Te Fb
0.90 0.90 Eoe
Sal Ts
Man Ts Te Fb
Trs Mac Mac
Tex Eoe He Man
0.70 Lf 0.70
Con Lf
Con Pa
He
0.50 Pa 0.50

0.30 0.30
0.60 0.70 0.80 0.90 1.00 1.10 1.20 1.30 1.40 0.60 0.70 0.80 0.90 1.00 1.10 1.20 1.30 1.40

Source: Author's own elaboration based on WIOD tables.

Note: Data calculated in 2008 US constant dollars. Colours: Black dots = manufacturing, green dots = primary
sector and, orange dots = services. Ag = ‘Agriculture’, Min = ‘Mining’, Fb = ‘Food and Beverages’, Tex =
‘Textiles’, Lf = ‘Leather and Footwear’, Wp = ‘Wood and Products of Wood’, Ppp = ‘Paper, Printing and
Publishing’, Pnf = ‘Petroleum and Nuclear Fuel’, Ch = ‘Chemicals’, Rp = ‘Rubber and Plastics’, Nmm = ‘Other
Non-Metallic Mineral’, Bm = ‘Basic Metals and Fabricated Metal’, Mac = ‘Machinery’, Eoe = ‘Electrical and
Optical Equipment’, Te = ‘Transport Equipment’, Man = ‘Manufacturing, Nec’, Ut = ‘Utilities’, Con =
‘Construction’, Sal = ‘Sales’, Ts = ‘Traditional Services’, Trs = ‘Transport Services’, Fs = ‘Financial Services’,
Bs = ‘Business Services’, Pa = ‘Public Admin’, and He = ‘Health and Education’.

The inter- and intra-sectoral dependence both in terms of supply and demand varies
between countries. In the case of South Korea, located in the upper-right quadrant, there are
six manufacturing subsectors characterised by a strong dependence on other sectors in terms
of both supply and demand of inputs. These ones are ‘Basic Metals and Fabricated Metal’,

141
‘Paper, Printing and Publishing’, ‘Other Non-Metallic Mineral’, ‘Rubber and Plastics’,
‘Chemicals’ and ‘Wood and Products of Wood’. By contrast, Brazil does not present the same
composition. Thus, as seen in Figure 4.1, ‘Basic Metals and Fabricated Metal’ and ‘Wood and
Products of Wood’ are not situated in this quadrant. Moreover, ‘Petroleum and Nuclear Fuel’,
which is situated in this quadrant, showed one of the highest BL linkages of the Brazilian
economy148. Distinct to mining and agriculture, a large portion of petroleum is consumed
domestically supplying inputs for manufacturing149. For instance, the main industrialised
products of the ‘Chemicals’, ‘Rubber and Plastics’ and ‘Textiles’ subsectors demand inputs
originating in the ‘Petroleum’ subsector. Additionally, this sector demands inputs
significantly from the machinery industry for the refining process.

Regarding the lower-right quadrant, except for ‘Construction’ in the Brazilian

economy150, both countries had the same manufacturing subsectors, i.e. ‘Transport
Equipment’, ‘Textiles’, ‘Food and Beverages’, ‘Manufacturing, Nec’, ‘Machinery’, ‘Leather
and Footwear’ and ‘Electrical and Optical Equipment’. Although Hirschman had anticipated
the special growth-enhancing properties of manufacturing, the figure 4.1 shows that the
majority of medium- and high-tech subsectors are presented in the lower-right quadrant. It
indicates not only that technologically advanced subsectors are very dependent on the inter-
and intra-sectoral supply, but also that they stimulate production in other sectors.
Additionally, the biggest difference between the Brazilian and South Korean linkages refers to
their levels. As a matter of fact, South Korea presented an economy with higher backward and
forward linkages indicating a more dynamic productive structure.

In the lower-left quadrant there are no manufacturing subsectors (backward and

forward linkages lower than the unity). This result indicates that manufacturing subsectors are
generally more inter-related to other sectors, particularly in pulling the domestic productive
chain, than the primary sector or services. In fact, in both countries this area is filled
exclusively by services. In South Korea and Brazil it comprised respectively five and three
subsectors. The former was represented by ‘Sales’, ‘Traditional Services’, ‘Transport

148
The highest increase in linkage indices in this sector was registered from 2004-2006, during which Brazil
became self-sufficient in oil production. For details on this series, see appendices 4.2, 4.3, 4.4 and 4.5.
149
As it is possible to realise by analysing the national accounts, Brazilian petroleum is not exported as a raw
material, as are mineral commodities. According to the national accounts, Brazilian exports of petroleum and
natural gas were 21.3% of the total production in 2009, whereas in the case of iron, 90.2% of the total production
was exported.
150
Construction presented a BL linkage of exactly 1.00. Therefore it is not clear which quadrant this subsector
belongs to.

142
Services’, ‘Health and Education’, and ‘Public Admin’. The latter included the same
subsectors, with the exception of ‘Sales’ and ‘Transport Services’. In this manner, as a general
picture, services have showed a high independence from other subsectors both in terms of
supply and demand. In the upper-left quadrant, subsectors characterised by high dependence
on the production of other sectors are presented. These subsectors usually are characterised by
the primary sector such as ‘Mining and ‘Agriculture’ and other subsectors marked by a lower
processing level. In South Korea this quadrant revealed that sectors such as ‘Mining’ and
‘Petroleum and Nuclear Fuel’ are less interconnected with the rest of the economy than in
Brazil.

2.3.Pure Normalised Backward and Forward Linkages

Although Hirschman-Rasmussen indices have been widely used in input-output

analysis, they do not consider the size of sectors in the economy. Therefore, an important
aspect to analyse the productive structure in Brazil and South Korea is the index
normalisation according to the sector size. Based on Guilhoto, Sonis & Hewings (1996),
Table 4.2 and 4.3 were compiled in two periods and ranking positions presented according to
the index size. With reference to the Pure Normalised Backward Linkage (PNBL) index, most
evident dichotomies between Brazil and South Korea are twofold. First, South Korea, despite
the presence of the ‘Construction’ subsector, presents important manufacturing sectors ranked
as the top six. Second, in this group, high-tech manufacturing subsectors such as ‘Electrical
and Optical Equipment’, ‘Transport Equipment’, and ‘Machinery’ figure with higher pure
backward linkages than those in the Brazilian economy. Hence, it shows that in South Korea,
dynamic manufacturing sectors have been used to stimulate the production of inputs of other
sectors and create a cumulative productive environment.

In Brazil, ‘Food and Beverages’ is ranked among the sectors with the highest PNBL.
Nonetheless, although this sector demands substantially from other subsectors of the
economy, it is classified into the low-tech group. Moreover, taking into account the top six
subsectors, it is possible to see that the economy is also dependent on services such as
‘Traditional Services’, ‘Public Admin’ and ‘Health and Education’ to drive the economy151.
Additionally, an interesting movement has been undertaken by ‘Agriculture’ and ‘Mining’.
From 1995–2002 to 2003 –2009, both subsectors increased their PNBL and improved ranking

151
Services represent nearly two thirds of the Brazilian value-added. For more details, see chapters 2 and 3.

143
 positions (Table 4.2). The explanations for this phenomenon are twofold. First, it is a
consequence of a significant share of the primary sectors in the Brazilian GDP. Second, it is a
result of an economy articulated and still dependent on its natural resources endowments. In
short, the Brazilian productive structure indicates a significant dependence on services
subsectors to stimulate the economy and also an increasing dependence on the primary sector.

Table 4.2 – Pure normalised backward linkages

South Korea Brazil
1995-2002 2003-2009 1995-2002 2003-2009
Subsectors PNBL Rank PNBL Rank PNBL Rank PNBL Rank
Agriculture 0.30 16 0.22 18 0.62 14 0.78 12
Mining 0.00 25 0.00 25 0.19 22 0.40 18
Food and Beverages 1.48 7 1.07 8 3.22 1 3.52 1
Textiles 1.02 10 0.56 12 0.46 17 0.38 19
Leather and Footwear 0.24 18 0.10 22 0.38 19 0.31 20
Wood and Products of Wood 0.01 24 0.00 24 0.06 25 0.08 24
Paper , Printing and Publishing 0.14 22 0.12 21 0.28 21 0.29 22
Petroleum and Nuclear Fuel 0.18 20 0.21 19 0.97 7 0.96 8
Chemicals 0.66 12 0.75 11 0.85 12 0.93 10
Rubber and Plastics 0.26 17 0.26 16 0.14 23 0.12 23
Other Non-Metallic Mineral 0.05 23 0.05 23 0.08 24 0.08 25
Basic Metals and Fabricated Metal 0.43 13 0.49 13 0.45 18 0.59 15
Machinery 1.23 8 1.40 6 0.88 11 1.13 7
Electrical and Optical Equipment 2.44 2 2.82 3 0.89 10 0.77 13
Transport Equipment 2.37 3 3.15 2 1.51 6 1.92 5
Manufacturing, Nec 0.36 14 0.24 17 0.50 16 0.42 17
Utilities 0.15 21 0.13 20 0.29 20 0.29 21
Construction 5.04 1 4.68 1 2.67 3 1.85 6
Sales 1.66 6 1.39 7 0.90 9 0.94 9
Traditional Services 2.29 4 2.28 5 2.15 5 2.15 4
Transport Services 1.05 9 0.83 10 0.56 15 0.60 14
Financial Services 0.36 15 0.32 14 0.72 13 0.58 16
Business Services 0.20 19 0.30 15 0.93 8 0.79 11
Public Admin 1.02 11 1.07 9 2.42 4 2.46 3
Health and Education 2.05 5 2.54 4 2.89 2 2.68 2
Source: Author's own elaboration based on WIOD tables.
Note: Data calculated in 2008 US constant dollars. ‘PNBL’ = Pure Normalised Backward Linkage and “Rank’ =
Ranking. Nec = Not elsewhere classified.

The Table 4.3 shows the results of pure normalised forward linkages (PNFL). As
expected, except for ‘Chemicals’, in both countries subsectors that produce products with a
lower degree of processing in relation to other subsectors exhibited relatively higher forward
linkages. In this manner, the top six subsectors demanded by other subsectors in the South
144
 Korean economy during both periods (1995-2002 and 2003-2009) were ‘Basic Metals and
Fabricated Metal’, ‘Petroleum and Nuclear Fuel’, ‘Business Services’, ‘Chemicals’,
‘Traditional Services’ and ‘Financial Services’. In the case of Brazil, ‘Sales’, ‘Agriculture’,
‘Basic Metals and Fabricated Metal’, ‘Chemicals’, ‘Traditional Services’ and ‘Business
Services’ ranked as the top six subsectors. Additionally, it is important to note that, although
‘Mining’ is characterised by a low degree of processing, it is ranked only 8th in the last period.
This position is mainly due to the low significance of this sector as a domestic supplier. In
other words, a massive output share is directed to exports and consequently not processed
domestically.

Table 4.3 – Pure normalised forward linkages

South Korea Brazil
1995-2002 2003-2009 1995-2002 2003-2009
Subsectors PNFL Rank PNFL Rank PNFL Rank PNFL Rank
Agriculture 1.06 8 0.84 12 1.91 4 2.04 2
Mining 0.18 20 0.12 23 1.41 7 1.46 8
Food and Beverages 0.76 11 0.76 14 0.94 13 0.84 13
Textiles 0.13 23 0.21 20 0.26 22 0.21 21
Leather and Footwear 0.04 25 0.03 25 0.01 25 0.01 25
Wood and Products of Wood 0.15 21 0.14 22 0.30 19 0.21 22
Paper , Printing and Publishing 0.74 14 0.75 15 0.71 15 0.67 15
Petroleum and Nuclear Fuel 2.45 2 1.89 5 1.39 8 1.37 9
Chemicals 2.09 4 1.92 4 1.64 6 1.53 6
Rubber and Plastics 0.78 10 0.93 10 0.78 14 0.68 14
Other Non-Metallic Mineral 0.75 13 0.72 16 0.63 16 0.54 16
Basic Metals and Fabricated Metal 4.54 1 4.52 1 1.95 2 1.77 5
Machinery 0.60 16 0.85 11 0.28 20 0.23 20
Electrical and Optical Equipment 0.32 18 0.82 13 0.43 17 0.41 18
Transport Equipment 0.32 17 0.45 17 0.27 21 0.30 19
Manufacturing, Nec 0.14 22 0.20 21 0.14 23 0.11 24
Utilities 1.01 9 1.16 8 1.15 11 1.28 10
Construction 0.60 15 0.32 18 0.40 18 0.43 17
Sales 1.36 7 1.38 7 2.87 1 2.96 1
Traditional Services 2.05 5 1.97 3 1.67 5 1.92 4
Transport Services 0.76 12 0.98 9 1.37 10 1.26 11
Financial Services 1.73 6 1.55 6 1.38 9 1.51 7
Business Services 2.12 3 2.13 2 1.91 3 2.02 3
Public Admin 0.07 24 0.07 24 0.14 24 0.14 23
Health and Education 0.24 19 0.29 19 1.07 12 1.12 12
Source: Author's own elaboration based on WIOD tables.
Note: Data calculated in 2008 US constant dollars. ‘PNFL’ = Pure Normalised Forward Linkage and “Rank’ =
Ranking. Nec = Not elsewhere classified.

145
3. The interdependence between manufacturing and services

The increasing share of services in middle-income countries has been studied in the
economic development literature. The analysis made by Rowthorn & Wells (1987), and more
recently by Palma (2005), Dasgupta & Singh (2006) and Tregenna (2009), addresses this
phenomenon through the classical de-industrialisation approach. In this line of thinking,
countries follow a broadly similar trajectory of economic development. As development gets
under way, in low-income countries, the share of the primary sector in the economy’s total
value-added and employment fall and there is a rapid increase in the share of manufacturing.
This process takes place until the economy reaches a certain level of per capita income152.
From this stage, manufacturing stabilises and then starts to fall back, taking the form of an
‘inverted-U’ curve. This trend is also followed by a corresponding increase in the share of
services in national employment and value-added and is often described as a natural process
of de-industrialisation. Nevertheless, some developing economies seem to present an
anomaly, i.e. a process of premature de-industrialisation153. For instance, on the one hand, the
Brazilian share of services in GDP of nearly 70% is similar to that of countries with much
higher incomes per capita. On the other hand, In South Korea, whose income per capita is at
least 2.5 times larger than in Brazil, services account for 54% of the GDP154.

This rising trend of services in GDP has stimulated not only a number of hypotheses to
explain the fall in manufacturing155, but particularly also studies that sought to investigate the
interdependence between manufacturing and services. According to this type of study,
services and manufacturing have been increasingly combined through a synergistic and
symbiotic relationship. In this way, Pisano & Shih (2009) state that manufacturing and
services share a collective pool of resources (industrial commons) that sustains innovative
process. Therefore, it is not mere coincidence that many studies have shown that
manufacturing has been using more intermediate services over the years. In this new
productive landscape, the share of services in manufacturing value-added has been increasing

152
According to Rowthorn & Wells (1987), considering a sample of 70 countries, this level would be around
US$12,000 in 1991.
153
For details on the process of de-industrialisation, see chapter 1.
154
Data calculated for 2010 in current US dollars.
155
For a detailed explanation about the three better-known hypotheses about de-industrialisation, see chapter 1.

146
steadily, surpassing 60% in middle-income countries and 65% in high-income countries (CNI,
2014)156.

Figure 4.2 compares the intermediate consumption of services in manufacturing value-

added for two periods in order to indentify an increasing evolution. As it is clear, although
these countries showed a distinct trend in terms of sectoral composition over the decades, both
presented an increasing trend of services in manufacturing value-added. From a comparative
perspective, Brazil presented a higher share of services in its manufacturing value-added than
South Korea. In numbers, comparing the averages for 1995 to 2009, Brazil and South Korea
showed respectively 40.37% and 38.51% of intermediate services in the manufacturing value-
added. Through this data it is also possible to identify a distinct evolution over time. During
the first period (1995 – 2002), South Korea presented a ratio of intermediate consumption of
services to manufacturing value-added of 37.19% and Brazil 35.78%. In the subsequent
period (2003-2009), not only did both countries expose a rising trend, but Brazil also
surpassed South Korea considerably. The former reached 45.62% and the latter 40.03%.

Figure 4.2 – Intermediate consumption of services in manufacturing value-added (%)

South Korea Brazil

50 50
48 48
46 46
44 44 45.62

42 42
40 40
40.03 40.37
38 38
38.51
36 37.19 36
34 35.78
34
32 32
30 30
1995-2002 2003-2009 1995-2009 1995-2002 2003-2009 1995-2009

Source: Author's own elaboration based on WIOD tables.

3.1. Opening intersectoral linkages

As mentioned previously, distinctions in terms of growth-enhancing effects between

manufacturing and services have blurred. In this context, although Brazil and South Korea
undertook distinct sectoral evolutions in GDP over the years, both countries have increased
156
See also Arbache (2015).

147
the ratio of services in manufacturing value-added. This information is in line with pure
normalised linkages presented in the previous section that showed an increasing relevance of
services in the Brazilian and South Korean productive structures. Therefore, in order to
clarify this supposed new trend, this section adopted once again the input-output methodology
to understand the intersectoral interdependence between manufacturing and services. For this
purpose, in a new methodological approach, each sectoral forward and backward linkage
index was decomposed. The idea is to understand the extension to which a given linkage is
composed by the primary sector and principally manufacturing and services. Both backward
and forward linkages were decomposed in terms of percentage shares.

3.1.1. Forward Linkage decomposition

As mentioned in Section 2.2, the forward linkage measures the extension to which a
given sector is demanded by other sectors, i.e. the importance of a hypothetical sector as an
input supplier. In Figure 4.3 it is possible to see the forward linkage decomposition for two
periods, 1995–2002 and 2003 –2009. The main findings are twofold. First, the linkage
composition, i.e. sectoral shares, did not vary significantly between the periods analysed.
Second, the data illustrated that in both countries, despite a considerable share of services in
GDP, its role is still more limited than those of the manufacturing sector157. In South Korea,
where services represented more than 50% of the GDP, about 50% of the services forward
linkage is represented by manufacturing. In other words, 50% of services supply is destined to
manufacturing. This relationship is distinct in the manufacturing sector. As showed clearly by
the manufacturing forward linkage, only approximately 25% of the manufacturing
intersectoral supply is destined to services.

In the Brazilian case, this distinct inter-sectoral dependence is even clearer. In Brazil,
even in a context where services represent 70% of the total value-added, services are more
dependent on manufacturing than vice versa. Data shows that the services forward linkage is
made up of manufacturing by approximately 40%. This means that 40% of services supply is
directed to manufacturing. In contrast, following the same trend observed in South Korea, a
smaller share of manufacturing is directed to services. Only about 30% of manufacturing
products, either inputs or outputs, are destined to services. This finding is consistent with
Walker (1985), who pointed that many services, particularly knowledge-intensive activities,

157
This finding is in line with Park & Chan (1989).

148
can only be performed with physical goods, i.e., manufacturing products, serving as inputs or
outputs. The most remarkable example is the service subsector of Information and
Communications Technology (ICT) that needs manufacturing products primarily to fulfil or
enable the function of information processing and communication by electronic means,
including transmission and display. Complementarily, the primary sector also showed a
higher dependence on manufacturing than on services. As a matter of fact, in both countries,
more than 70% of the primary sector forward linkage is represented by manufacturing.

Figure 4.3 – Forward Linkage decomposition (%)

South Korea Brazil

FL (Manuf) FL (Serv) FL (Prim) FL (Manuf) FL (Serv) FL (Prim)

100 100

90 16.56 20.72 90 17.77 17.99

23.56 23.58 29.81 29.49
80 80
47.92 49.17 52.21
70 55.53
70
60 60

50 50
73.39 71.75
78.57 74.99
40 74.42 74.72 40 64.86
66.00
30 30
50.22 49.23 41.11
39.69
20 20

10 10
5.64 6.68 8.85 10.26
4.88 4.29 0 4.19 4.78
0 2.02 1.71 1.86 1.60

Primary Sector Manufacturing Services

Source: Author's own elaboration based on WIOD tables.

Note: Manuf = Manufacturing, Serv = Services, Prim = Primary Sector.

3.1.2. Backward Linkage decomposition

The backward linkage indicates the extent to which a given sector demands from other
sectors. The backward linkage decomposition adopted in this subsection is analogous to the
method previously used to the forward linkage. Therefore, this section sought to identify the
sectoral composition of the backward linkage for each sector, i.e. manufacturing, services and
the primary sector. In this analysis, two movements stand out. First, the backward linkage

149
intersectoral change of services in South Korea was more evident than in Brazil. This means
that although in both countries manufacturing is demanding more services over the periods
described, in South Korea this movement was clearer. Second, while for South Korea the
backward linkage of manufacturing has a lower share of services than the backward linkage
of services has of manufacturing, the opposite holds true for Brazil.

Figure 4.4 – Backward Linkage decomposition (%)

South Korea Brazil

BL (Manuf) BL (Serv) BL (Prim) BL (Manuf) BL (Serv) BL (Prim)

100 100

90 90
26.87 27.37
37.41 33.85 36.82
80 39.69 80 41.16 44.34
53.92 57.94 70
70 63.11 64.07
60 60

50 50

40 67.16 67.59 40 54.36 50.58 41.91

53.53 51.97 38.85
30 30
43.59 39.61 31.18 29.35
20 20

10 10 16.93 16.81
9.05 11.79 12.60
5.97 5.03 8.34 5.71 6.59
0 2.50 2.46 0

Primary Sector Manufacturing Services

Source: Author's own elaboration based on WIOD tables.

Note: Manuf = Manufacturing, Serv = Services, Prim = Primary Sector.

This overall picture can be detailed by the numbers shown in the Figure 4.4. In South
Korea, during the first and second period, respectively 43.59% and 39.61% of the services
backward linkage was made up of manufacturing. Nonetheless, the backward linkage of
manufacturing is represented approximately by only 27% of services, illustrating the distinct
interdependence between these sectors. In Brazil, data shows that services backward linkage
not only experienced a drop in terms of manufacturing share for the periods analysed, but also
manufacturing makes up approximately 30%. It means that the magnitude of services demand
from manufacturing is approximately only 30%. In the case of the manufacturing backward
linkage, the share of services increased over the periods. Moreover, as indicated before, the

150
share of services in the manufacturing backward linkage is greater than the share of
manufacturing in the services backward linkage. Additionally, the composition of the primary
sector backward linkage in both countries showed an increase of services. Nevertheless, in
Brazil it surpassed the manufacturing share during 2003–2009.

4. Concluding Remarks

This chapter sough to analyse the productive composition of the Brazilian and South
Korean economies through an input-output analysis. In this way, the first section discussed
briefly Hirschman’s theory particularly about the relevance of industrial linkages for
economic growth. In this approach, an underdeveloped economy is marked by the absence of
linkages or inter- and intra-sectoral interdependences that affect cumulative effects on demand
and supply. Moreover, in an underdeveloped industrial structure, investments in intermediate
and capital goods are restricted due to the lack of dynamic industrial linkages. Therefore, a
central condition in the economic growth process is an interdependent industrial structure. For
this purpose, the state is a central agent to direct investments particularly to dynamic
manufacturing subsectors which present the strongest capacity to articulate strong linkages to
the rest of the economy.

In this context, the chapter advanced to an applied analysis that outlined key sectors
and productive specificities of each economy. The first section initially measured output
multipliers and Hirschman-Rasmussen indices. The output multipliers indicated that South
Korean subsectors are more interrelated than the Brazilian ones. From 1995 to 2002, South
Korea presented higher output multipliers for the majority of subsectors, particularly in
manufacturing. In the subsequent period, i.e. from 2003 to 2009, South Korea expanded its
output multiplier showing only four sectors with lower multipliers than Brazil. Consequently,
an increase in the South Korean final demand has a more dynamic effect on the domestic
economy than in Brazil.

In terms of Hirschman-Rasmussen backward and forward linkages, as a general

picture, in both countries manufacturing showed high backward indices, while services and
the primary sector presented high forward linkages. The most evident exception was the
‘Petroleum’ subsector that in Brazil showed one of the highest backward linkages, while in
South Korea presented a very low one. Although Hirschman had anticipated the central role

151
of manufacturing, this analysis showed specifically that medium- and high-tech subsectors are
very dependent on the inter- and intra-sectoral supply, and consequently have the highest
potential to stimulate the production in other sectors. Furthermore, the main difference
between the Brazilian and South Korean linkages refers to the level of the linkages. From a
comparative perspective, South Korea presented an economy with higher backward and
forward linkages than Brazil.

In a complementary way, seeking to refine the analysis of the productive structure, the
Hirschman-Rasmussen indices were normalised according to the sector size in each economy.
Therefore, in a cross-country analysis, the different productive structures of Brazil and South
Korea become more evident. In terms of backward linkages, the main dichotomies between
Brazil and South Korea were threefold. First, South Korea had important manufacturing
sectors ranked as the top six. Second, in this group, high-tech manufacturing figured with
higher pure normalised backward linkages than those in Brazil. Thus, it illustrates that in
South Korea, dynamic manufacturing sectors have been used to stimulate the output of other
sectors and create an environment of cumulative productive causation. Third, despite a
general increase of the share of services in GDP in both economies, the Brazilian productive
structure indicated a significant dependence on services subsectors to stimulate the economy
and also an increasing dependence on the primary sector. With regard to forward linkages, in
both countries, except for ‘Chemicals’, subsectors that produce products with a lower degree
of processing in relation to other subsectors exhibited relatively higher forward linkages.

In the face of the increasing role of services in these indices and the rising trend of
intermediate consumption of services in manufacturing value-added, the third section
examined essentially the interdependence between services and manufacturing. Therefore, the
forward and backward linkages for each sector were decomposed between manufacturing,
services and the primary sector through a new input-output method. First, the forward linkage
decomposition showed that: (i) the sectoral share of each linkage did not vary significantly
between the periods analysed; and (ii), in both countries despite a considerable share of
services in GDP, their role in terms of forward linkage composition is still more limited than
those of manufacturing. Second, the backward linkage decomposition showed that: (i) the
backward linkage of all sectors are using more services over the periods (moreover, this trend
is more evident in South Korea than in Brazil); however (ii), in South Korea, the backward
linkage of manufacturing has a lower share of services than the backward linkage of services
has of manufacturing, while the opposite holds true for Brazil. Additionally, compared to

152
South Korea, the smaller share of the Brazilian manufacturing sector in both forward and
backward linkages illustrates pervasive effects of neoliberal reforms. After trade
liberalisation, the share of imported manufacturing inputs in final manufacturing production
in Brazil increased significantly. In other words, production chains were broken and
manufacturing moved more to a final output assembly type of production. As many services
are non-tradable, the impact was different in this sector.

153
Appendix 4.1 – Technical Details

Output Multipliers and Hirschman-Rasmussen Linkages

The indices calculated in this chapter through input-output methods used data from the
World Input-Output Database (WIOD) for the period 1995-2009 at the level of 34 sectors.
Additionally, this session was elaborated on the models described by Guilhoto (2009) and
Guilhoto et al. (2010)158. Therefore, the basic Leontief model can be represented as:

X = I − A2; Y (4.1)

Where Q represents the economy’s total production, and n − A2; = L is the inverse of
Leontief’s and ˆ is the final demand. From equation (4.1), it is possible to calculate a
number of important indices, such as generators, multipliers and linkages. Therefore, firstly
define ‰ as the direct coefficient:

‹e
kf = (4.2)
Œe

Where K f is the variable (e.g., employment, value-added goods, and wages) of the
corresponding sector Xf . Once ‰ is calculated, along with Leontief’s inverse matrix r, it is
possible to calculate the amount of K directly and indirectly generated for each monetary unit
produced for the final demand for each sector. This value is referred to as the generator,
which relates production for final demand to a given variable of the economy. Thus, the
generator of a variable Ž for each sector can be calculated by summing each column of matrix
GK as follows:

Ž = ∑DGj; ‰G . rG (4.3)

Dividing the generator, GK, by the respective direct coefficient, it is possible to obtain the
multiplier of variable K that associates the direct effect of a variable regarding its total (direct
and indirect) effect on the economy as follows:

I‘ƒ
TŽ = ’“
(4.4)

158
For more details, see also Miller & Blair (2009), Guilhoto (2011), Haddad et al. (2007) and Borghi (2013).

154
 In this way, multipliers for employment and production can be obtained. In addition,
the input-output methodology allows other indicators of economic importance to be
calculated. The seminal works of Hirschman (1958) and Rasmussen (1956) led to the
development of backward and forward linkage indices that become a basic instrument to
identify key sectors in the economy. The Hirschman-Rasmussen backward linkage (”r) index
measures the demand of a given sector for other sectors, while the forward linkage (•r) index
measures the degree to which this sector is demanded by other sectors. To calculate the
Hirschman-Rasmussen ”r index, lWf is defined as the elements of matrix r, L∗ is the average
of all elements of r and L∗f is the sum of a column of r. The ”r is expressed as follows:

˜∗ƒ ⁄D
”r = ˜∗
(4.5)

Defining F as being the matrix of the row coefficients obtained from the intermediate
consumption matrix, G as the Ghosh matrix obtained from  = n − •2; ,  ∗ as the average
value of all elements of , and G∗ as the sum of the elements in each row, then the forward
indices (FL) are represented as follows:

I“∗ ⁄D
•r = (4.6)
I∗

Sectors can be classified into the following four groups depending on their index
values: (i) independent from (or not highly related to) other sectors if both linkage indices are
less than 1; (ii) dependent on (or strongly related to) other sectors if both linkage indices are
greater than 1, denoting sectors that play a key role in the economy; (ii) dependent on
intersectoral supply (or stimulates production in other sectors) only if the BL index is greater
than 1; and (iv) dependent on intersectoral demand (or dependent on the production of other
sectors) only if the FL index is greater than 1. However, as observed by Cella (1984) and
Clements (1990), these indices do not take into account the different production levels in each
sector of the economy.

Therefore, to correct and refine the solutions presented by Cella (1984) and Clements
(1990), Guilhoto et al. (1994) elaborated the first version of the so-called pure linkage index,
which later became known as the GHS methodology. Guilhoto, Sonis & Hewings (1996)
present decompositions of Leontief’s inverse matrix that incorporate the main methods used

155
in input-output analysis to decompose and differentiate the impact of an economic sector on
its various components. The consolidated GHS methodology is based on a block matrix of
technical coefficients (A), represented as follows:

o o3 o o3 0 0

o=š ›=š ›+ š › = o + o3 (4.7)
o3 o33 o3 0 0 o33

where Aff and A œœ represent square matrices of the direct technical coefficients of sector „ and
the rest of the economy (the entire economy minus sector „), respectively. Afœ and
A œfrepresent rectangular matrices of direct inputs purchased by sector „ from the rest of the
economy and direct inputs purchased by the rest of the economy from sector „, respectively.
Additionally, o refers to the sector „ isolated from the rest of the economy; and o3 represents
the rest of the economy.

Based on matrix A in (4.7), a triple multiplicative decomposition of Leontief’s inverse

matrix can be expressed as follows:
r r„$ ∆„„ 0 ∆„ 0 n o„$ ∆$
r = n − o2; = šr r$$
›=š ›š ›š › (4.8)
$„ 0 ∆$$ 0 ∆$ o$„ ∆„ n

where,

2;
∆ = n − o " (4.9)

∆3 = n − o33 2; (4.10)

2;
∆ = n − ∆ o3 ∆3 o3 " (4.11)

2;
∆ = n − ∆3 o3 ∆ o3 " (4.12)

From equation (4.8) it is possible to reveal the process of production in an economy as

well as derive the Pure Backward Linkage (PBL) and the Pure Forward Linkage (PFL). The
indices are defined as:

”r = ∆3 o3 ∆ ˆ (4.13)

156
 •r = ∆ o3 ∆3 ˆ3 (4.14)

In Equation (4.13), the index indicates the impact of the value of the total output of sector j on
the remainder of the economy, minus demand for inputs that sector j produces for itself and
the returns of the remainder of the economy for sector j and vice versa. The PFL in Equation
(19) indicates the impact of the value of the total production of the remainder of the economy
on sector j. The PBL and PFL are summed to calculate the pure total linkage (PTL) index for
each sector of the economy, expressed in current values:

Ur = ”r + •r (4.15)

However, because these indices do not consider the size of the sectors, which is an
important aspect for identifying key sectors of the economy, a “normalisation” procedure
should be applied to these indices based on the approach of normalised pure linkage indices.
For this purpose, the pure indices of each sector are divided by the average of pure indices for
the economy as a whole. Thus, the normalised PBL (PBLN) index, the normalised PFL
(PFLN) index and the normalised PTL (PTLN) index can be represented as follows:

”rG = ”rG ⁄∑DGj; ”rG ⁄s (4.16)

•rG = •rG ⁄∑DGj; •rG ⁄s (4.17)

UrG = UrG ⁄∑DGj; UrG ⁄s (4.18)

The interdependence between manufacturing and services

As previously defined, the Hirschman-Rasmussen backward and forwards linkages

measure the magnitude that a given sector demands and supplies other sectors in the economy
as a whole. In this sense, to identify and quantify the interdependence between sectors,
backward and forward linkages for each sector were broken in primary sector, manufacturing
and services. Defining rž = r − n to exclude the direct impact on the sector itself, equation
4.5 was broken in the three aforementioned sectors:

157
 (˜ž ƒ ⁄D)
”r′K = ˜¡∗
(4.19)

(˜ž¢ƒ ⁄D)
”r′C = ˜¡∗
(4.20)

(˜ž£ƒ ⁄D)
”r′? = ˜¡∗
(4.21)

where r′K , r′C and r′? is the summation of the rows corresponding to the sector analysed
(primary sector, manufacturing or services, respectively) for each column (sector „), and

(˜ž∗ƒ ⁄D)
”r′ = (4.22)
˜¡∗

is the backward linkage excluding the direct impact on the sector itself.

In the case of forward linkages, an analogous method was applied redefining equation
4.6 as:

(Iž“ ⁄D)
•r′G6 = (4.23)
I ¡∗

(Iž“¢⁄D)
•r′G, = (4.24)
Iž∗

(Iž“£ ⁄D)
•r′G¤ = (4.25)
Iž∗

where ′G6 , ′G, and ′G¤ is the summation of the lines corresponding to the sector analysed
(primary sector, manufacturing or services, respectively) for each line (sector V), and

I ¡∗ƒ ⁄D"
•rž = I ¡∗
(4.26)

is the forward linkage excluding the direct impact on the sector itself.

158
 Appendix 4.1 – Map of sectoral abbreviation and aggregation

WIOD input-output table – 34 subsectors Abbreviation – 34 subsectors

Agriculture, Hunting, Forestry and Fishing Agriculture
Mining and Quarrying Mining
Food, Beverages and Tobacco Food and Beverages
Textiles and Textile Products Textiles
Leather, Leather and Footwear Leather and Footwear
Wood and Products of Wood and Cork Wood and Products of Wood
Pulp, Paper, Paper , Printing and Publishing Paper , Printing and Publishing
Coke, Refined Petroleum and Nuclear Fuel Petroleum and Nuclear Fuel
Chemicals and Chemical Products Chemicals
Rubber and Plastics Rubber and Plastics
Other Non-Metallic Mineral Other Non-Metallic Mineral
Basic Metals and Fabricated Metal Basic Metals and Fabricated Metal
Machinery, Nec Machinery
Electrical and Optical Equipment Electrical and Optical Equipment
Transport Equipment Transport Equipment
Manufacturing, Nec; Recycling Manufacturing, Nec
Electricity, Gas and Water Supply Utilities
Construction Construction
Sale, Maintenance and Repair of Motor Vehicles and
Motorcycles; Retail Sale of Fuel Sales
Wholesale Trade and Commission Trade, Except of Motor
Vehicles and Motorcycles Sales
Retail Trade, Except of Motor Vehicles and Motorcycles;
Repair of Household Goods Sales
Hotels and Restaurants Traditional services
Inland Transport Transport services
Water Transport Transport services
Air Transport Transport services
Other Supporting and Auxiliary Transport Activities;
Activities of Travel Agencies Traditional services
Post and Telecommunications Traditional services
Financial Intermediation Financial services
Real Estate Activities Traditional services
Renting of M&Eq and Other Business Activities Business services
Public Admin and Defence; Compulsory Social Security Public Admin
Education Health and education
Health and Social Work Health and education
Other Community, Social and Personal Services Health and education
Source: Author’s own elaboration
Note: Nec = Not elsewhere classified.

159
Appendix 4.2 - Hirschman-Rasmussen Backward Linkage index (Brazil)

Subsectors 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09
Agriculture 0.87 0.88 0.88 0.88 0.90 0.90 0.88 0.89 0.90 0.92 0.92 0.89 0.89 0.92 0.91
Mining 0.82 0.82 0.82 0.81 0.82 0.82 0.87 0.90 0.89 0.91 0.93 0.92 0.92 0.92 0.82
Food and Beverages 1.17 1.19 1.19 1.18 1.19 1.20 1.17 1.16 1.19 1.21 1.23 1.22 1.22 1.26 1.25
Textiles 1.07 1.08 1.08 1.06 1.09 1.06 1.08 1.11 1.08 1.08 1.05 1.05 1.03 1.03 1.02
Leather and Footwear 1.23 1.21 1.21 1.18 1.17 1.17 1.17 1.19 1.19 1.20 1.16 1.14 1.17 1.15 1.13
Wood and Products of Wood 0.92 0.92 0.91 0.91 0.92 0.92 0.93 0.95 0.98 1.03 1.04 1.01 1.03 1.04 1.03
Paper , Printing and Publishing 1.11 1.13 1.13 1.13 1.13 1.13 1.10 1.09 1.08 1.06 1.06 1.03 1.06 1.03 1.00
Petroleum and Nuclear Fuel 1.14 1.12 1.16 1.21 1.21 1.20 1.22 1.18 1.17 1.16 1.24 1.30 1.24 1.22 1.29
Chemicals 0.95 0.99 0.99 0.98 1.00 1.08 1.05 1.05 1.11 1.15 1.11 1.09 1.10 1.13 1.10
Rubber and Plastics 1.08 1.08 1.07 1.07 1.09 1.12 1.07 1.05 1.12 1.13 1.14 1.12 1.10 1.09 1.13
Other Non-Metallic Mineral 1.05 1.05 1.04 1.06 1.05 1.04 1.05 1.01 1.08 1.01 1.00 1.03 1.04 1.06 1.10
Basic Metals and Fabricated Metal 0.96 0.97 0.96 0.94 0.93 0.95 0.95 0.98 0.99 0.98 1.03 1.03 1.06 1.07 1.05
Machinery 1.16 1.16 1.14 1.12 1.11 1.12 1.13 1.12 1.15 1.13 1.14 1.14 1.13 1.11 1.17
Electrical and Optical Equipment 1.20 1.19 1.18 1.16 1.11 1.08 1.10 1.10 1.08 1.06 1.07 1.06 1.08 1.05 1.10
Transport Equipment 1.24 1.22 1.20 1.16 1.12 1.14 1.14 1.18 1.23 1.22 1.24 1.23 1.20 1.18 1.22
Manufacturing, Nec 1.12 1.12 1.11 1.10 1.10 1.10 1.07 1.04 1.08 1.06 1.04 1.02 1.01 1.03 1.04
Utilities 0.88 0.89 0.89 0.93 0.94 0.93 1.05 1.00 0.96 0.93 0.92 0.91 0.90 0.93 0.92
Construction 1.08 1.09 1.08 1.07 1.06 1.06 1.03 1.04 0.96 0.94 0.94 0.94 0.94 0.94 0.99
Sales 0.74 0.74 0.75 0.75 0.76 0.75 0.75 0.75 0.75 0.75 0.76 0.76 0.77 0.77 0.76
Traditional Services 0.81 0.82 0.83 0.84 0.86 0.85 0.85 0.83 0.81 0.81 0.81 0.82 0.83 0.83 0.82
Transport Services 0.86 0.84 0.84 0.85 0.87 0.86 0.88 0.89 0.90 0.89 0.89 0.90 0.91 0.92 0.91
Financial Services 0.94 0.94 0.94 0.96 0.97 0.97 0.90 0.90 0.87 0.90 0.84 0.85 0.83 0.83 0.79
Business Services 0.93 0.92 0.94 0.95 0.93 0.90 0.91 0.90 0.86 0.85 0.84 0.86 0.86 0.85 0.84
Public Admin 0.81 0.81 0.82 0.83 0.82 0.82 0.81 0.84 0.77 0.80 0.81 0.81 0.82 0.82 0.80
Health and Education 0.86 0.85 0.86 0.87 0.85 0.84 0.84 0.84 0.81 0.82 0.81 0.83 0.83 0.82 0.82
Source: Author's own elaboration based on WIOD tables.
Note: Nec = Not elsewhere classified.

160
Appendix 4.3 - Hirschman-Rasmussen Forward Linkage index (Brazil)

Subsectors 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09
Agriculture 1.10 1.09 1.09 1.10 1.11 1.10 1.09 1.08 1.07 1.09 1.11 1.11 1.11 1.08 1.06
Mining 1.46 1.46 1.49 1.46 1.45 1.47 1.46 1.42 1.43 1.40 1.39 1.38 1.36 1.33 1.40
Food and Beverages 0.82 0.81 0.82 0.82 0.83 0.83 0.82 0.81 0.80 0.81 0.82 0.83 0.83 0.83 0.80
Textiles 0.93 0.94 0.93 0.93 0.94 0.93 0.93 0.96 0.96 0.96 0.93 0.93 0.90 0.89 0.86
Leather and Footwear 0.72 0.71 0.71 0.69 0.68 0.68 0.69 0.70 0.69 0.68 0.68 0.68 0.68 0.69 0.67
Wood and Products of Wood 1.27 1.27 1.25 1.24 1.12 1.15 1.10 1.05 1.07 1.06 1.11 1.13 1.16 1.22 1.28
Paper , Printing and Publishing 1.24 1.25 1.27 1.27 1.25 1.27 1.23 1.24 1.22 1.23 1.22 1.21 1.22 1.21 1.18
Petroleum and Nuclear Fuel 1.16 1.13 1.16 1.18 1.19 1.19 1.18 1.20 1.22 1.23 1.22 1.22 1.22 1.21 1.20
Chemicals 1.15 1.15 1.15 1.15 1.16 1.18 1.18 1.19 1.22 1.25 1.23 1.19 1.17 1.19 1.18
Rubber and Plastics 1.31 1.32 1.31 1.29 1.30 1.29 1.30 1.30 1.32 1.33 1.35 1.33 1.33 1.32 1.32
Other Non-Metallic Mineral 1.21 1.22 1.20 1.21 1.22 1.22 1.22 1.23 1.25 1.22 1.21 1.21 1.22 1.25 1.28
Basic Metals and Fabricated Metal 1.23 1.25 1.25 1.23 1.18 1.18 1.18 1.18 1.17 1.15 1.18 1.18 1.19 1.19 1.30
Machinery 0.81 0.82 0.82 0.81 0.83 0.81 0.80 0.79 0.76 0.75 0.76 0.75 0.73 0.72 0.76
Electrical and Optical Equipment 0.91 0.92 0.92 0.91 0.90 0.85 0.92 0.86 0.92 0.88 0.87 0.87 0.91 0.89 0.93
Transport Equipment 0.85 0.82 0.80 0.78 0.75 0.74 0.72 0.74 0.76 0.77 0.80 0.80 0.78 0.78 0.81
Manufacturing, Nec 0.76 0.75 0.75 0.75 0.76 0.76 0.73 0.75 0.74 0.74 0.73 0.73 0.73 0.73 0.72
Utilities 1.19 1.19 1.19 1.21 1.23 1.25 1.33 1.34 1.31 1.32 1.29 1.29 1.28 1.29 1.24
Construction 0.64 0.65 0.65 0.66 0.69 0.67 0.67 0.68 0.67 0.67 0.67 0.67 0.68 0.68 0.69
Sales 1.02 1.02 1.02 1.02 1.04 1.05 1.05 1.07 1.08 1.08 1.08 1.07 1.06 1.05 1.04
Traditional Services 0.79 0.80 0.80 0.81 0.83 0.83 0.84 0.84 0.84 0.85 0.84 0.85 0.85 0.86 0.83
Transport Services 1.06 1.06 1.06 1.07 1.08 1.09 1.10 1.09 1.08 1.10 1.10 1.10 1.11 1.10 1.07
Financial Services 1.03 1.02 1.02 1.03 1.05 1.06 1.05 1.04 1.03 1.04 1.02 1.04 1.06 1.08 1.04
Business Services 1.06 1.06 1.06 1.07 1.08 1.09 1.10 1.09 1.08 1.10 1.10 1.10 1.11 1.10 1.07
Public Admin 0.57 0.57 0.57 0.57 0.59 0.59 0.59 0.59 0.58 0.58 0.57 0.58 0.58 0.59 0.56
Health and Education 0.71 0.71 0.71 0.71 0.73 0.73 0.73 0.73 0.72 0.73 0.73 0.74 0.74 0.74 0.72
Source: Author's own elaboration based on WIOD tables.
Note: Nec = Not elsewhere classified.

161
Appendix 4.4 - Hirschman-Rasmussen Backward Linkage index (South Korea)

Subsectors 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09
Agriculture 0.82 0.82 0.84 0.86 0.86 0.87 0.86 0.86 0.87 0.87 0.87 0.88 0.89 0.91 0.91
Mining 0.74 0.74 0.73 0.70 0.74 0.72 0.71 0.76 0.79 0.81 0.81 0.80 0.80 0.82 0.80
Food and Beverages 1.07 1.10 1.12 1.12 1.16 1.14 1.13 1.14 1.13 1.14 1.16 1.16 1.16 1.19 1.24
Textiles 1.19 1.19 1.20 1.28 1.23 1.21 1.19 1.17 1.19 1.16 1.14 1.12 1.10 1.08 1.10
Leather and Footwear 1.23 1.23 1.26 1.32 1.24 1.23 1.23 1.20 1.20 1.19 1.18 1.18 1.20 1.22 1.27
Wood and Products of Wood 0.99 1.03 1.05 1.15 1.05 1.03 1.01 1.01 1.02 1.05 1.05 1.06 1.05 1.05 1.10
Paper , Printing and Publishing 1.06 1.09 1.12 1.18 1.11 1.13 1.12 1.11 1.11 1.12 1.11 1.10 1.10 1.14 1.09
Petroleum and Nuclear Fuel 0.59 0.59 0.58 0.60 0.60 0.61 0.60 0.59 0.59 0.61 0.61 0.62 0.64 0.64 0.66
Chemicals 1.06 1.04 1.04 1.07 1.07 1.10 1.08 1.05 1.06 1.09 1.07 1.08 1.08 1.10 1.05
Rubber and Plastics 1.09 1.10 1.10 1.10 1.13 1.14 1.14 1.14 1.14 1.13 1.15 1.14 1.13 1.13 1.12
Other Non-Metallic Mineral 1.21 1.20 1.19 1.18 1.17 1.12 1.13 1.15 1.15 1.12 1.10 1.09 1.07 1.05 1.05
Basic Metals and Fabricated Metal 1.05 1.02 1.00 0.97 1.03 1.06 1.05 1.06 1.05 1.09 1.13 1.11 1.13 1.18 1.21
Machinery 1.31 1.30 1.27 1.23 1.31 1.26 1.28 1.29 1.27 1.24 1.27 1.25 1.23 1.19 1.22
Electrical and Optical Equipment 1.93 1.85 1.65 1.54 1.43 1.38 1.37 1.32 1.23 1.18 1.14 1.11 1.07 1.05 1.05
Transport Equipment 1.18 1.18 1.21 1.20 1.28 1.27 1.28 1.32 1.30 1.28 1.32 1.33 1.31 1.27 1.35
Manufacturing, Nec 1.19 1.21 1.17 1.19 1.21 1.20 1.21 1.23 1.24 1.26 1.27 1.27 1.26 1.19 1.21
Utilities 0.82 0.81 0.81 0.82 0.82 0.83 0.81 0.78 0.76 0.78 0.78 0.81 0.83 0.85 0.87
Construction 0.94 0.94 0.95 0.92 0.96 1.00 1.00 1.01 1.02 1.02 1.03 1.03 1.05 1.06 1.03
Sales 0.83 0.84 0.86 0.88 0.84 0.81 0.82 0.84 0.86 0.86 0.86 0.86 0.86 0.85 0.83
Traditional Services 0.89 0.88 0.89 0.85 0.86 0.89 0.88 0.89 0.90 0.90 0.89 0.90 0.91 0.92 0.89
Transport Services 0.87 0.84 0.86 0.88 0.87 0.85 0.87 0.85 0.86 0.87 0.84 0.82 0.84 0.86 0.80
Financial Services 0.80 0.83 0.85 0.83 0.85 0.89 0.88 0.86 0.85 0.83 0.81 0.85 0.87 0.89 0.84
Business Services 0.75 0.77 0.80 0.75 0.75 0.78 0.80 0.81 0.83 0.83 0.83 0.86 0.84 0.83 0.80
Public Admin 0.71 0.71 0.71 0.69 0.70 0.72 0.73 0.74 0.76 0.77 0.77 0.78 0.77 0.76 0.74
Health and Education 0.70 0.71 0.73 0.70 0.73 0.78 0.79 0.81 0.81 0.80 0.79 0.80 0.81 0.80 0.78
Source: Author's own elaboration based on WIOD tables.
Note: Nec = Not elsewhere classified.

162
Appendix 4.5 - Hirschman-Rasmussen Forward Linkage index (South Korea)

Subsectors 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09
Agriculture 0.97 1.05 1.10 1.21 1.15 1.15 1.14 1.13 1.15 1.15 1.13 1.13 1.13 1.13 1.19
Mining 1.65 1.56 1.60 2.28 1.66 1.66 1.67 1.62 1.61 1.57 1.66 1.62 1.62 1.48 2.13
Food and Beverages 0.79 0.89 0.91 0.90 0.90 0.89 0.90 0.91 0.93 0.93 0.92 0.93 0.93 0.96 0.92
Textiles 0.72 0.72 0.72 0.64 0.72 0.75 0.75 0.78 0.78 0.79 0.80 0.83 0.84 0.84 0.83
Leather and Footwear 0.76 0.78 0.76 0.73 0.73 0.71 0.71 0.72 0.75 0.74 0.74 0.76 0.78 0.84 0.85
Wood and Products of Wood 1.28 1.29 1.31 1.25 1.30 1.32 1.32 1.34 1.34 1.36 1.36 1.37 1.36 1.37 1.34
Paper , Printing and Publishing 1.37 1.40 1.40 1.38 1.39 1.41 1.40 1.40 1.39 1.37 1.35 1.36 1.38 1.37 1.31
Petroleum and Nuclear Fuel 1.21 1.14 1.12 1.15 1.12 1.07 1.08 1.04 1.02 1.06 1.10 1.08 1.06 1.04 1.05
Chemicals 1.21 1.23 1.21 1.21 1.26 1.28 1.28 1.27 1.25 1.21 1.19 1.19 1.16 1.14 1.09
Rubber and Plastics 1.20 1.19 1.17 1.10 1.18 1.19 1.18 1.18 1.18 1.16 1.18 1.19 1.18 1.15 1.14
Other Non-Metallic Mineral 1.26 1.26 1.27 1.21 1.24 1.26 1.25 1.24 1.23 1.22 1.19 1.19 1.19 1.19 1.17
Basic Metals and Fabricated Metal 1.37 1.36 1.32 1.17 1.35 1.39 1.39 1.39 1.36 1.36 1.39 1.35 1.35 1.34 1.31
Machinery 0.92 0.92 0.92 0.86 0.90 0.89 0.87 0.87 0.89 0.90 0.90 0.91 0.91 0.89 0.88
Electrical and Optical Equipment 0.80 0.80 0.74 0.73 0.76 0.76 0.78 0.79 0.78 0.77 0.77 0.77 0.78 0.75 0.75
Transport Equipment 0.75 0.77 0.79 0.72 0.79 0.77 0.77 0.79 0.80 0.79 0.81 0.82 0.80 0.75 0.76
Manufacturing, Nec 0.78 0.76 0.78 0.73 0.77 0.77 0.77 0.82 0.83 0.86 0.87 0.89 0.90 0.91 0.96
Utilities 1.35 1.31 1.28 1.26 1.26 1.25 1.24 1.22 1.23 1.26 1.25 1.25 1.26 1.30 1.25
Construction 0.59 0.58 0.59 0.59 0.59 0.59 0.58 0.56 0.56 0.55 0.53 0.53 0.53 0.54 0.50
Sales 0.79 0.81 0.82 0.82 0.83 0.83 0.84 0.84 0.85 0.85 0.84 0.84 0.84 0.85 0.80
Traditional Services 0.88 0.85 0.86 0.85 0.84 0.84 0.83 0.83 0.83 0.84 0.82 0.83 0.83 0.85 0.80
Transport Services 0.84 0.82 0.81 0.78 0.79 0.77 0.83 0.85 0.88 0.89 0.90 0.89 0.88 0.92 0.83
Financial Services 1.20 1.19 1.19 1.17 1.16 1.15 1.13 1.11 1.10 1.09 1.06 1.07 1.09 1.12 1.05
Business Services 1.24 1.25 1.25 1.20 1.23 1.22 1.21 1.21 1.20 1.19 1.15 1.15 1.15 1.17 1.10
Public Admin 0.51 0.52 0.53 0.51 0.52 0.53 0.52 0.52 0.52 0.52 0.51 0.51 0.51 0.52 0.48
Health and Education 0.56 0.55 0.56 0.55 0.55 0.56 0.56 0.56 0.56 0.56 0.55 0.54 0.54 0.56 0.51
Source: Author's own elaboration based on WIOD tables.
Note: Nec = Not elsewhere classified.

163
Appendix 4.6 - Backward and Forward Linkages Decomposition (Brazil, %)

1995-2002 2003-2009
Backward Linkage Forward Linkage Backward Linkage Forward Linkage
Subsectors Prim Manuf Serv Prim Manuf Serv Prim Manuf Serv Prim Manuf Serv
Agriculture 23.72 45.86 30.42 11.78 72.17 16.05 23.31 44.57 32.12 12.99 72.51 14.50
Mining 10.13 37.97 51.90 5.91 74.61 19.48 10.31 33.13 56.56 7.54 70.98 21.47
Food and Beverages 36.25 34.36 29.39 10.27 48.74 41.00 36.74 31.50 31.76 12.55 50.50 36.94
Textiles 5.46 61.22 33.33 2.24 74.43 23.33 7.76 54.98 37.26 3.54 72.52 23.93
Leather and Footwear 6.36 60.43 33.21 0.38 95.34 4.28 7.10 56.88 36.01 0.43 95.68 3.89
Wood and Products of Wood 16.55 55.56 27.89 2.54 85.59 11.87 18.88 50.62 30.50 3.07 85.53 11.40
Paper , Printing and Publishing 7.53 50.79 41.68 2.18 43.49 54.33 11.17 48.60 40.23 3.23 45.66 51.12
Petroleum and Nuclear Fuel 46.40 28.21 25.39 6.47 54.48 39.05 46.44 25.18 28.38 8.78 52.02 39.20
Chemicals 9.62 51.60 38.78 11.97 67.10 20.94 10.00 48.00 42.00 14.89 66.56 18.55
Rubber and Plastics 4.97 58.38 36.65 3.81 67.71 28.47 5.92 55.11 38.97 4.63 64.34 31.03
Other Non-Metallic Mineral 13.43 44.27 42.31 2.00 83.73 14.27 10.93 43.23 45.84 3.70 81.10 15.21
Basic Metals and Fabricated Metal 8.63 57.11 34.26 4.10 84.46 11.44 9.96 50.44 39.60 4.98 84.50 10.52
Machinery 3.83 64.62 31.55 7.73 74.57 17.70 4.47 59.66 35.87 10.29 72.65 17.05
Electrical and Optical Equipment 3.99 57.92 38.09 2.83 66.38 30.78 4.40 53.76 41.84 4.16 63.19 32.66
Transport Equipment 2.65 62.45 34.90 1.61 65.89 32.50 3.11 60.27 36.62 2.04 69.09 28.87
Manufacturing, Nec 4.73 65.71 29.55 2.00 43.20 54.81 5.84 60.57 33.59 2.58 41.49 55.93
Utilities 6.47 21.32 72.20 4.25 41.19 54.56 10.46 20.72 68.83 5.65 43.74 50.60
Construction 6.38 62.80 30.83 2.78 34.85 62.38 6.23 59.87 33.90 5.80 28.11 66.09
Sales 4.33 26.13 69.54 6.42 58.16 35.43 5.03 25.60 69.37 8.22 59.72 32.05
Traditional Services 11.57 44.98 43.45 5.03 34.29 60.68 11.20 40.71 48.09 7.45 35.54 57.01
Transport Services 5.78 34.82 59.40 4.87 36.99 58.15 6.69 33.54 59.78 6.90 37.67 55.43
Financial Services 2.27 22.04 75.69 3.51 35.62 60.86 2.47 19.14 78.40 5.06 38.39 56.55
Business Services 5.78 34.82 59.40 4.87 36.99 58.15 6.69 33.54 59.78 6.90 37.67 55.43
Public Admin 4.60 26.70 68.71 4.46 37.51 58.03 4.82 25.48 69.70 6.34 38.76 54.90
Health and Education 4.90 38.61 56.49 4.85 36.75 58.40 5.34 36.06 58.60 6.88 37.42 55.70
Source: Author's own elaboration based on WIOD tables
Note: Nec = Not elsewhere classified.

164
Appendix 4.6 - Backward and Forward Linkages Decomposition (South Korea, %)

1995-2002 2003-2009
Backward Linkage Forward Linkage Backward Linkage Forward Linkage
Subsectors Prim Manuf Serv Prim Manuf Serv Prim Manuf Serv Prim Manuf Serv
Agriculture 16.97 58.46 24.58 9.20 69.15 21.65 15.74 57.14 27.12 8.06 62.82 29.11
Mining 1.14 48.61 50.25 0.56 87.98 11.46 0.93 46.80 52.27 0.51 87.16 12.34
Food and Beverages 35.67 43.06 21.27 11.87 45.89 42.23 32.46 42.17 25.37 9.23 40.24 50.54
Textiles 3.38 66.93 29.68 1.16 86.58 12.26 2.27 65.25 32.48 1.99 81.13 16.88
Leather and Footwear 7.40 69.00 23.60 0.29 88.85 10.86 6.32 65.14 28.54 0.43 80.81 18.76
Wood and Products of Wood 15.49 53.20 31.30 3.03 83.36 13.61 12.06 57.48 30.46 2.99 84.90 12.11
Paper , Printing and Publishing 0.99 69.22 29.79 1.19 66.06 32.75 0.93 66.15 32.92 1.32 63.52 35.17
Petroleum and Nuclear Fuel 3.36 62.26 34.38 2.68 60.84 36.48 3.16 57.86 38.98 2.38 64.81 32.81
Chemicals 1.65 74.31 24.04 3.15 81.23 15.62 1.41 74.54 24.06 2.27 80.86 16.87
Rubber and Plastics 2.17 75.23 22.60 2.16 83.33 14.51 1.99 74.55 23.46 1.73 85.98 12.28
Other Non-Metallic Mineral 11.56 54.80 33.64 0.39 89.95 9.65 9.00 53.76 37.24 0.31 91.96 7.73
Basic Metals and Fabricated Metal 0.69 81.83 17.48 0.38 93.46 6.16 0.59 80.65 18.76 0.26 94.71 5.03
Machinery 0.63 75.04 24.33 1.22 85.26 13.53 0.53 77.55 21.93 0.78 86.45 12.77
Electrical and Optical Equipment 1.08 66.76 32.16 0.34 89.57 10.09 0.79 72.61 26.60 0.22 89.58 10.20
Transport Equipment 0.60 78.95 20.46 0.77 82.10 17.13 0.49 82.83 16.68 0.44 84.71 14.86
Manufacturing, Nec 2.34 68.27 29.39 0.74 58.79 40.47 1.64 72.09 26.27 0.43 68.38 31.19
Utilities 1.81 53.15 45.04 1.39 58.45 40.16 1.75 42.70 55.55 1.19 55.82 42.99
Construction 2.49 68.52 28.98 0.88 21.07 78.05 1.83 71.30 26.87 0.82 22.71 76.46
Sales 1.51 32.13 66.36 1.73 65.37 32.91 1.52 32.17 66.31 1.72 62.66 35.62
Traditional Services 6.70 47.16 46.14 1.41 35.97 62.62 7.79 40.48 51.73 1.36 39.88 58.76
Transport Services 1.07 56.20 42.74 1.76 59.62 38.61 1.10 55.67 43.23 1.57 63.65 34.78
Financial Services 1.11 19.82 79.08 2.23 46.91 50.86 0.86 13.75 85.39 1.39 36.22 62.39
Business Services 4.43 36.80 58.76 1.57 57.18 41.25 3.45 35.51 61.05 1.18 56.90 41.92
Public Admin 1.75 55.50 42.76 2.72 37.24 60.05 1.68 51.83 46.49 2.74 37.60 59.65
Health and Education 1.59 47.93 50.48 2.10 41.02 56.88 1.51 44.74 53.75 1.61 41.13 57.26
Source: Author's own elaboration based on WIOD tables
Note: Nec = Not elsewhere classified.

165
 CONCLUSIONS

This thesis investigated the role of the manufacturing sector in economic growth
through a comparative analysis between Brazil and South Korea. The study adopted
theoretical, historical and empirical approaches to understand the particularities of each
productive structure that conditioned their distinct economic dynamic. The first chapter
presented the theoretical core of this thesis. An effort to combine different heterodox strands
on economic development highlighting the importance of the manufacturing sector to
economic growth was made. Through contributions from Keynesian-Kaldorian, Structuralist
and Neo-Schumpeterian theories, this chapter argued that the manufacturing sector presents
some growth-enhancing properties not found in other sectors.

Both Anglo-Saxon and Latin American Structuralism strands stressed that economic
development is narrowly linked to a radical transformation in favour of the manufacturing
sector to overcome underdevelopment. Structuralist strands state that a dynamic process of
industrialisation is a necessary condition for increasing employment, productivity and income
per capita and, consequently, reducing poverty. According to this approach, the process of
economic development involves a shift of production from low productivity to high
productivity sectors where increasing returns to scale prevail.

Complementarily, the Kaldorian literature detailed that the manufacturing sector is

imbued with special properties not shared by other sectors that trigger a process of cumulative
causation. Thus, from a Kaldorian perspective, the understanding of the development process
and growth rate differences between countries requires a sectoral approach. The Kaldorian
section also explored the role of manufacturing for productivity growth, stating that the rate of
productivity growth in the manufacturing sector depends on its rate of output growth due to
static and dynamic returns to scale. Furthermore, this economic approach emphasises the
central role of manufacturing for the balance of payments equilibrium.

The importance of manufacturing was also approached in the first chapter through the
debate about de-industrialisation that identifies three main causes for this process, namely
statistical illusions, ‘servicification’ and productive fragmentation. This section showed that a
non-negligible part of the extent of de-industrialisation reflects changes in statistical
classification rather than a real decrease in the relative importance of manufacturing.
Furthermore, despite the recent rise of services, manufacturing remains the driver of

166
economic growth, as many services essentially are spin-offs from manufacturing production.
Additionally, the increasing trend of productive fragmentation over the world not only
reduced the manufacturing share in the GDP of many economies but also affected negatively
these countries through a deterioration of the so-called “industrial commons” that resulted in a
loss of high-value-added activities and the capacity to create technological innovation.

The first chapter also examined the Neo-Schumpeterian route to development,

exploring relations between innovation, economic dynamism and catching-up in a sectoral-
specific approach. The convergence between Neo-Schumpeterian literature and Structuralist
economics showed the role of manufacturing as a main locus of technological change. From
an economic development perspective, developed and developing economies could be divided
according to their capacity to innovate and imitate. In this approach, the diffusion of
technological progress to developing economies results from a process of technology
absorption through imitation of products and process from developed economies. Along the
economic development trajectory, countries first imitate and then innovate according to their
own needs. In this process, the cumulative causation is generated by the accumulation of
knowledge on productivity growth.

From the aforementioned theoretical framework that considers a thriving

manufacturing sector a necessary condition for a successful trajectory of economic growth,
the second chapter analysed the industrial development undertaken by Brazil and South Korea
and identified domestic and external factors that determined their distinct routes of economic
dynamism. This study showed that during the industrialist era (1950-1980) both countries
presented a significant economic dynamism based on an active developmental state. In South
Korea, the strategy of economic development comprised policies for technology acquisition,
incentives for R&D activities and the promotion of cartels for specific purposes, such as
standardisation, specialisation and exports. Additionally, the South Korean domestic private
capital, represented by the Chaebols, played a decisive role in the development of the national
industry with a clear outward orientation. In Brazil, by contrast, industrialisation was mainly
driven by state-owned enterprises and multinational companies, following a pattern of
industrial growth grounded in its large domestic market.

Nevertheless, during the 1970s, both economies were strongly impacted by two oil
price shocks in 1973 and 1979, and the Volcker interest rate hikes in 1979. The outbreak of
the debt crisis in 1982 and the abrupt cut-off of the international liquidity exposed the

167
vulnerability of the Brazilian economy, given its large current account deficit and external
debt. This turbulent economic environment resulted in a serious domestic crisis and caused a
deep disarticulation of the state to provide active policies for national industry as before.
Consequently, economic growth was seriously affected during the 1980s, giving rise to the
commonly used term “the lost decade”. South Korea dealt with the crisis in a much less
traumatic way, since its strategy of industrialisation was relatively less dependent on foreign
debt and the Japanese banks played a decisive role ‘recycling’ the country’s liabilities.
Therefore, distinct from the Brazilian economy, South Korea not only recovered rapidly from
the debt crisis, but also strengthened the manufacturing sector and the production of cutting-
edge technological products.

From the 1980s onwards, Brazil embarked on a ‘fundamentalist’ neoliberal agenda

which sought to reverse almost every aspect of the growth strategy previously followed by the
country. Neoliberal policies were strongly marked by an extreme move towards market
friendly policies which weakened the domestic manufacturing sector. In South Korea, the
neoliberal policies implemented were remarkably different. South Korean economic reforms
were much more pragmatic, seeking to transform the initial catching-up efforts into a dynamic
and sustainable process of capital expansion and development. After the 1980s, despite the
neoliberal discourse, the country’s experience was characterised by the maintenance of a
strong state control over socioeconomic transformations and by pragmatism in the
management of intermediate and instrumental objectives, such as economic
internationalisation and the administration of the fiscal and monetary policy.

In Brazil, the pro-market policy orientation resulted in an economic dynamic of

successive ‘stop-and-go’ cycles derived from a fragile productive structure and lack of
effective demand. From the 1980s onwards, the country’s growth rate collapsed both in terms
of GDP and manufacturing value-added growth. Moreover, from an intra-sectoral viewpoint,
the manufacturing sector showed a technology composition that stagnated over the decades.
In contrast, South Korea presented a much more sustainable economic growth. The economy
grew, on average, more than 6% per year and the manufacturing sector expanded
significantly. In terms of intra-sectoral value-added composition, the share of high-tech
industries rose sharply, reaching around 61% of the manufacturing share in 2010.

The deleterious effects of premature de-industrialisation in Brazil vis-à-vis South

Korea’s thriving manufacturing expansion clearly had an impact on the evolution of labour

168
productivity and income catching-up. Data shows that the remarkable neglect of
manufacturing since economic reforms lies at the heart of country’s productivity stagnation.
In contrast to the period from 1950 to 1980, when in both countries manufacturing and
productivity expanded considerably, from 1980 to 2010 due to the manufacturing collapse,
Brazilian productivity remained mostly stagnant. Even during the commodities prices boom
from 2002 to 2008, Brazilian economy productivity growth was extremely poor. In terms of
income catching-up, while South Korea moved ahead after the advent of neoliberal reforms,
Brazil drastically fell behind.

Taking into account the ‘Kaldorian-Structuralist’ approach, it is not by coincidence

that the persistent difficulty to overcome the middle-income trap appears to be closely related
to the absence of a dynamic manufacturing sector. Regarding their export trajectories, the
dichotomy between these economies is also very clear. On the one hand, South Korea pursued
an outward strategy of economic growth expanding exports of manufacturing products,
particularly of high-tech products. On the other hand, Brazil has been historically dependent
on its domestic market and Ricardian comparative advantages, i.e. natural resources
endowments. As a matter of fact, even in a context of a dynamic domestic market, the
Brazilian economic policy has been unable to articulate a dynamic manufacturing sector to
compete internationally.

The structural dichotomy between Brazil and South Korea in terms of productive
structure was detailed through input-output analyses in the third and fourth chapters, which
constituted the empirical part of this thesis. The third chapter complemented the previous
historical analysis, detailing the structural transition and dichotomy between Brazil and South
Korea. Based on the methodology of Imbs and Wacziarg, data showed that Brazil was
affected by a process of premature specialisation at both the inter- and the intra-sectoral
levels. Intra-sectoral data of manufacturing revealed that during the specialisation stage, South
Korea increased sharply the share of medium- and high-tech sectors in the economy while
Brazil, although more diversified, remained dependent on less dynamic sectors. This
dichotomy propelled different processes of cumulative causation and, consequently, impacted
on the countries’ growth rates distinctly.

The third chapter also applied the structural decomposition analysis (SDA) to
understand factors such as demand, trade and technological change that contributed to the
formation of these divergent productive structures. Results showed that, during the height of

169
neoliberal reforms from 1995 to 2003, the Brazilian manufacturing output exhibited a poor
performance with the majority of components registering negative variations. By contrast, the
South Korean manufacturing output increased sharply, mainly due to exports and positive
variations in other components. Furthermore, this expansion was composed mostly of
medium- and high-tech industries. In the subsequent period, from 2003 to 2008, besides the
significant output growth, South Korea registered an expressive increase of imports, mainly
due to the ‘flying geese’ movement of regional productive fragmentation marked especially
by the rise of China as the world’s factory.

In Brazil, domestic demand from 2003 to 2008 composed the key variable of
economic growth, propelled by the credit-led consumption of the period. The boom in
commodity prices also played a central role in boosting the economy in that period.
Nevertheless, once again manufacturing performed worse when compared to services, which
showed the most dynamic output growth. Evidence of the productive disarticulation of the
Brazilian industry was reinforced by the negative variation of the component related to
imports. Under high and volatile interest and exchange rates, periods of demand expansion
(such as during the 2000s) experienced substitution of national inputs for imported ones as the
easiest way to attend demand. This fact not only revealed the dependence of the Brazilian
domestic industry in relation to the international supply but also the wreaking of important
intra-sectoral linkages.

The final chapter dealt with the productive structure of the Brazilian and South Korean
economies in terms of leading sectors and sectoral interdependences. In this sense,
Hirschman’s theory was applied to explain the importance of industrial linkages for economic
growth. According to this thinking, an underdeveloped economy is marked by the absence of
inter- and intra-sectoral linkages that affect cumulative effects in demand and supply. In this
context, the chapter initially measured output multipliers and Hirschman-Rasmussen indices.
Output multipliers indicated that the South Korean subsectors are more interrelated than the
Brazilian ones. From 1995 to 2002, South Korea presented higher output multipliers for the
majority of subsectors, particularly in manufacturing. In the subsequent period, from 2003 to
2009, South Korea expanded its output multiplier showing only four sectors with lower
multipliers than Brazil. Thus, an increase in the South Korean final demand has a more
dynamic effect on the domestic economy than in Brazil.

170
 In terms of Hirschman-Rasmussen backward and forward linkages, as a general
picture, in both countries manufacturing presented high backward indices, while services and
the primary sector presented high forward linkages. The exception was the ‘Petroleum’
subsector, which in Brazil showed one of the highest backward linkages while in South Korea
presented a very low one. Although Hirschman had anticipated the central role of
manufacturing, this analysis showed particularly that the majority of medium- and high-tech
subsectors is very dependent on the inter- and intra-sectoral supply and, consequently, has the
highest potential to stimulate the production in other sectors. Furthermore, the main difference
between Brazilian and South Korean linkages refers to the level of their linkages. From a
comparative perspective, South Korea presented an economy with higher backward and
forward linkages than Brazil.

In a complementary way, seeking to refine the analysis of the productive structure, the
Hirschman-Rasmussen indices were normalised according to the sector size in each economy.
The different productive structures of Brazil and South Korea became more evident. Results
of pure normalised backward linkages showed that South Korea not only presented more
manufacturing sectors but also more high-tech subsectors with the highest linkages. These
findings illustrate that in South Korea, dynamic manufacturing sectors have been used to
stimulate the production of inputs of other sectors and to create an environment of cumulative
productive causation. Despite a general increase of services in both economies, the Brazilian
productive structure indicated a significant dependence on services subsectors to stimulate the
economy and an increasing dependence on the primary sector as well. In regard to the forward
linkage, in both countries, except for ‘Chemicals’, subsectors that produced goods with a
lower degree of processing in relation to other subsectors exhibited relatively higher forward
linkages.

In face of the increasing role of services in these indices and their ascendant trend of
intermediate consumption in manufacturing value-added, the fourth chapter also examined the
interdependence between services and manufacturing. Both forward and backward linkages
for each sector were decomposed between manufacturing, services and the primary sector
through a new input-output method. First, the forward linkage decomposition showed that (i)
the sectoral share of each linkage did not vary significantly between the periods analysed, and
(ii) in both countries, despite a considerable share of services in GDP, their role in terms of
forward linkage composition is still more limited than those of manufacturing. Second, the
backward linkage decomposition showed that (i) all sectors are using more services over both

171
periods, a trend more evident in South Korea than in Brazil, and (ii) while the South Korean
backward linkage of manufacturing has a lower share of services than the backward linkage
of services has of manufacturing, the opposite holds true for Brazil.

172
 REFERENCES

Abramovitz, M. (1986) ‘Catching-up, Forging Ahead, and Falling Behind’, Journal of

Economic History, 66(2), 385-406.

Abreu, M. P. (2003) ‘Latin America: The External Context, 1928-1982’, Unpublished

Discussion Paper 472, PUC-RJ, Departamento de Economia.

Aghion, P. & Howitt, P. (1998) Endogenous Growth Theory, Cambridge, Mass: MIT Press.

Akyuz, Y. (1996) ‘New Trends in Japanese trade and FDI: Post-Industrial Transformation and
policy changes’. UNCTAD, study nº1.

Akyüz, Y., Chang, H-J., & Kozul-Wright, R. (1998) ‘New perspectives on East Asian
development’, The Journal of Development Studies, 34(6), 4-36.

Albuquerque, E. M. (1999), ‘National system of innovation and Non-OECD countries: notes

about a rudimentary and tentative “typology”’, Brazilian Journal of Political Economy, 19(4),
35-52.

Amsden, A. H. (1989) Asia’s next giant: South Korea and late industrialisation, New York:
Oxford University Press.

Amsden, A. (2001) The e Rise of the Rest: Non-Western Economies’ Ascent in World
Markets, Oxford: Oxford University Press.

Ancochea, D. S. (2007) ‘Anglo-Saxon Structuralism vs. Latin American Structuralism in

Development Economics’, In E. Pérez and M. Vernengo, eds. Ideas, Policies and Economic
Development in the Americas, New York: Routledge.

Angeriz, A. ; McCombie, J. S. L. & Roberts, R. (2009) ‘Increasing Returns and the Growth
of Industries in the EU Regions: Paradoxes and Conundrums’, Spatial Economic Analysis,
4(2), 127-148.

Araujo, R. A. & Lima, G.T. (2007) ‘A structural economic dynamics approach to balance-of-
payments-constrained growth’, Cambridge Journal of Economics, 31(5), 755-774.

Arbache, J. (2015) ‘The Contribution of Services to Manufacturing Competitiveness in

Brazil’, Working Paper, University of Brasilia - Department of Economics.

Arrighi, G.; Ikeda, S. & Irwan, A. (1993) ‘The Rise of East Asia: One Miracle or Many?’, in
R.A. Palat, (ed)., Pacific Asia and the Future of the World-Economy. Westport, Ct:
Greenwood Press.

Arrow, K.J. (1962) ‘The Economic Implications of Learning by Doing’, Review of Economic
Studies, 29, 3, 155-173.

173
Bacha, E. & Bonelli, R. (2005) ‘Uma interpretação das causas da desaceleração econômica do
Brasil’, Revista de Economia Política, 3(99), 163-189.

Baghirathan, R.; Rada, C. & Taylor, L. (2004) ‘The Worldly Philosophers, models, and
methodology, Social Research: An International Quarterly, 71(2), 305-326.

Balassa, B. (1988) ‘The lessons of East Asian development: an overview’, Economic

Development and Cultural Change, 36 (3), S273–S290.

Barro, R. & Sala-I-Martin, X. (2004) Economic Growth, Boston: Mit Press.

Barros, J. R. M. & Goldenstein, L. (1997) ‘Reestruturação Industrial e Reordenação

Financeira: um novo balanço’. Mimeo.

Belluzzo, L. G. M. & Almeida, J. G. (2002) Depois da Queda: A economia braisleira da crise

da dívida aos impasses do real, Rio de Janeiro: Civilização Brasileira.

Bénabou, R. (1982) ‘La Corée du Sud ou l´industrialisation planifiée’, Economie Prospective

Internationale, CEP II, 10, p.12-156.

Berger, S. (2011) ‘Why Manufacturing Matters’, MIT Technology Review, 1-2.

Bernat, G. A. (1996). ‘Does manufacturing matter? A spatial econometric view of Kaldor’s

laws’, Journal of Regional Science, 36(3), 463–477.

Bielschowsky, R. (1998) ‘Cincuenta años del pensamiento de la cepal: una reseña’, in

Cincuenta años del pensamiento de la cepal: textos seleccionados, Santiago: Economic
Commission for Latin America and the Caribbean (ECLAC)/ Fondo de Cultura Económica.

Bielschowsky, R. (2009) ‘Sixty Years of ECLAC: Structuralism and neo-structuralism’, 97,

171-192.

Blankenburg, S, Palma, J. G. & Tregenna, F. (2008) ‘Structuralism’, in Steven N. Durlauf and

Lawrence E. Blume, eds, The New Palgrave Dictionary of Economics, Basingstoke: Palgrave
Macmillan.

Blankenburg, S. (2000) ‘Knowledge, economic growth and the role of policy’, Working
Paper, 185, ESRC Centre for Business Research, Department of Applied Economics.

Blankenburg, S. (2004) ‘Knowledge in the ‘New Endogenous Growth Theory’: A factor of

production, a good and an externality’, PhD thesis, Cambridge University.

BNDES (2001) ‘Carga tributária, Evolução Histórica: Uma tendência crescente’, Secretaria
para assuntos fiscais, Informe-se, nº 29.

Borghi, R. A. Z. (2013) Brazilian tax policy as response to the international economic crisis:
an assessment based on the productive structure and leading sectors. 17th conference of the
research network macroeconomics and macroeconomic policies (FMM). Berlin.

174
Britto, G (2008) ‘Economic growth in Brazil from a Kaldorian standpoint’, PhD thesis,
University of University.

Canuto, O. & Ferreira, Jr., H. M. (1989) Coreia do Sul e Taiwan: aspectos histórico-
estruturais e políticas industriais. Reestruturação industrial e competitividade internacional,
Coleção economia paulista, São Paulo: Fundação Seade/ Unicamp.

Canuto, O. (1994) Brasil e Coreia do Sul: os (des)caminhos da industrialização tardia, São

Paulo: Nobel.

Cardoso de Mello, J. M. (1982) O capitalismo tardio: Contribuição à revisão crítica da

formação e do desenvolvimento brasileiro. Coleção 30 anos de Economia. Campinas:
IE/Unicamp.

Carneiro, R. (2002) Desenvolvimento em crise: a economia brasileira no último quarto do

século XX, São Paulo: Editora UNESP, IE – UNICAMP.

Carneiro, R. (2008) ‘Impasses do desenvolvimento brasileiro: a questão produtiva’, Texto

para Discussão. IE/UNICAMP, n. 153, Campinas.

Cassiolato, J. E. & Lastres, H.M.M. (2008) ‘Discussing innovation and development:

Converging points between the Latin American school and the Innovation Systems
perspective?’, Working Paper Series, The Global Network for Economics of Learning,
Innovation, and Competence Building System, 08-02.

Castro, A. B. & Souza, F. P. (2004) A Economia Brasileira em Marcha Forçada, Third

Edition, Rio de Janeiro: Editora Paz e Terra.

Castro, A. B. (2001) ‘A Reestruturação Industrial Brasileira nos Anos 90. Uma Interpretação’,
Revista de Economia Política, vol. 21, 3 (81).

CECIMO (2011) Study on Competitiveness of the European Machine Tool Industry, Brussels:
CECIMO.

Cella, G. (1984) ‘The input-output measurement of interindustry linkages’, Oxford Bulletin of

Economic and Statistics, v. 46, 73-84.

Chang, H-J, Andreoni, A. & Kuan, M. L. (2013) ‘International Industrial Policy Experiences
and the Lessons for the UK’, in The Future of Manufacturing , UK Government Office of
Science, London: BIS.

Chang, H-J. (1993) ‘The political economy of industrial policy in Korea’, Cambridge Journal
of Economics, 17(2), 131-157.

Chang, H-J. (1994) The Political Economy of Industrial Policy, Cambridge: St. Martin Press.

Chang, H-J. (1998) ‘Korea: the misunderstood crisis’, World Development, 26(8), 1998.

175
Chang, H-J. (1999) ‘The Economic Theory of the Developmental State’, in: M. Woo-
Cumings (ed.), The Developmental State, New York: Cornell University Press, 182-199.

Chang, H-J. (2002) Kicking away the ladder: development strategy in historical perspective,
London: Anthem Press.

Chang, H-J. (2012) ‘The Manufacturing Sector and the Future of Malaysia’s Economic
Development’, Jurnal Pengurusan, 35, 3-12.

Chang, H-J. (2014), Economics: The user’s guide. London: Penguin Group.

Chang, H-J; Park, H-J & Yoo, C. G. (1998) ‘Interpreting the Korean crisis: financial
liberalisation, industrial policy and corporate governance’, Cambridge Journal of Economics,
22, 735-746.

Chang, S-J. (2003) Crisis and Transformation of Korean Business Group: The Rise and Fall
of Chaebols. New York: Cambridge University Press.

Chenery, H. (1975) ‘The Structuralist Approach to Development Policy’, American Economic

Review, 65(2), 310-16.

Chenery, H. B. & Syrquin, M. (1980) A comparative analysis of industrial growth, in R. C. O.

Matthews (Ed.) Economic Growth and Resources: Volume 2—Trends and Factors (London:
Macmillan).

Chenery, H. B. (1960) ‘Patterns of Industriai Growth’, American Economie Review, 50(4),

624-654.

Chenery, H. B. (1979), Structural Change and Development Policy, New York: Oxford
University Press.

Chenery, H. B., Shishido, S. & Watanabe, T. (1962) ‘The pattern of Japanese growth, 1914-
1954’, Econometrica, 30(1), pp. 98–139.

Chenery, H. B.; Robinson, S. & Syrquin, M. (1986), Industrialization and growth: a

comparative study. Oxford: Oxford University Press.

Cheng, T-J., Haggard, S. & Kang, D. (1998) ‘Institutions and growth in Korea and Taiwan:
The bureaucracy’, The Journal of Development Studies, 34:6, 87-111.

Cimoli, M., G. Porcile & S. Rovira (2010) Structural Change and the BOP Constraint: Why
Did Latin America Fail to Converge?, Cambridge Journal of Economics, 389-411.

Clements, B. J. (1990) ‘On the decomposition and normalization of interindustry linkages’,

Economics Letters, v. 33 (4), 337-340.

Cleveland, W.S. & Devlin, S.J. (1988) ‘Locally weighted regression: an approach to
regression analysis by local fitting’, Journal of the American Statistical Association, 83, pp.
596–610.

176
Cleveland, W.S. (1979) ‘Robust locally weighted regression and smoothing scatterplots’,
Journal of the American Statistical Association, 74, pp. 829–836.

CNI. (2014) ‘Serviços e competitividade industrial no Brasil’, Confederação Nacional da

Indústria, Brasília: CNI, 2014.

Collins, S. & Park, W-A (1989), ‘External Debt and Macroeconomic Performance in South
Korea’, in J. D. Sachs (Ed), Developing Country Debt and the World Economy, Chicago:
University of Chicago Press.

Cornawall, J. (1976) ‘Diffusion, Convergence and Kaldor`s Laws’. The Economic Journal,
86, 307-314.

Cornawall, J. (1977) Modern Capitalism. Its Growth and Transformation, New York: St.
Martin's Press.

Coutinho, L. & Reichstul, H. F. (1982) ‘Investimento Estatal 1974/80: ciclo e crise’, in

Belluzzo, L. G. M. & Coutinho, R. (Eds), Desenvolvimento capitalista no Brasil: Ensaios
sobre a crise, São Paulo: Brasiliense.

Coutinho, L. (1999), ‘Coreia do Sul e Brasil: paralelos, sucessos e desastres’, in, J. L. Fiori,
ed, Estados e moedas no desenvolvimento das nações, Rio de Janeiro: Vozes, 351-378.

Coutinho, L. (2000) ‘Overcoming crises resulting from adherence to the Washington

Consensus: lessons from the Republic of Korea and Brazil’, International Social Science
Journal, 52(166), 517–527.

Cunha, A. M. & Acioly, L. (2009) ‘China: Ascensão à condição de potencia global -

características e implicações’ in J. C. Cardoso Jr., L. Acioly & M. Matijascic, Trajetórias
recentes de desenvolvimento: estudos de experiências internacionais selecionadas, Brasília:
IPEA.

Dahmén, E. (1987) ‘Development Blocks in Industrial Economics’, Scandinavian Economic

History Review, 1, 3-14.

Dasgupta, S. & Singh, A. (2006) ‘Manufacturing, Services and Premature Deindustrialization

in Developing Countries: A Kaldorian Analysis’, Working Papers RP2006/49, World Institute
for Development Economic Research (UNU-WIDER).

Davis, G. A. (1995) ‘Learning to love the Dutch Disease: Evidence from the mineral
economies’, World Development, Vol. 23, n. 10, pp. 1765-1779.

Dietzenbacher, E. & Los, B. (1998) ‘Structural Decomposition Techniques: Sense and

Sensibility’, Economics Systems Research, 10(4)l, 307-323.

Dixon, R. & Thirlwall, A. P. (1975) ‘A model of regional growth-rate differences on

Kaldorian lines’. Oxford Economic Papers, 27 (2), 201-214.

177
Dixon, R. & Thirlwall, A. P. (1979) A model of export-led growth with a balance of payments
constraint, in J. Bowers, ed., Inflation, Development and Integration: essays in honour of A.J.
Brown, Leeds: Leeds University Press.

Dosi, G., Pavitt, K., & Soete, L. (1990) The economics of technological change and
international trade, Brighton: Wheatsheaf.

Dosi, G.; Fabiani, S & Freeman, C. (1994) ‘The process of economic development:
introducing some stylized facts and theories on technologies, firms and institutions’.
Industrial and Corporate Change, 3(1), 1-45.

European Commission (2010), An integrated industrial policy for the globalisation era.
Putting Competitiveness and Sustainability at Centre Stage, Brussels: European Commission.

Ezell, S. J. & Atkinson, R. D. (2011) The Case for a National Manufacturing Strategy’,
Information Technology and Innovation Foundation (ITIF), Available at:
http://www.itif.org/files/2011-national-manufacturing-strategy.pdf.

Fagerberg, J. (1988) ‘International Competitiveness’, The Economic Journal, 98 (391), 355-

374.

Fagerberg, J. (1988) ‘Why growth rates differ’, in Dosi, G., Freeman, C., Nelson, R.,
Silverberg, G. and Soete, L, eds., Technical Change and Economic Theory. London: Pinter.

Fagerberg, J. (1994) ‘Technology and International Differences in Growth Rates’, Journal of

Economic Literature, 32, 1147–1175.

Fagerberg, J. & Verspagen, B. (1999) ‘Modern Capitalism in the 1970s and 1980s’, in M.
Setterfield ed., Growth, Employment and Inflation, Basingstoke: MacMillan.

Fagerberg, J. & Verspagen, B. (2002) ‘Technology-Gaps, Innovation-Diffusion and

Transformation: an Evolutionary Interpretation’, Research Policy, 31, 1291-1304.

Fagerberg, J. & Verspagen, B. (2007) ‘Innovation, Growth and Economic Development:

Have the Conditions for Catch-up Changed?’, International Journal of Technological
Learning, Innovation and Development, 1,(1), 13-33.

Fajnzylber, F. (1983) La industrialización trunca de América Latina, México, D.F: Editorial

Nueva Imagen, Centro de Economía Transnacional.

Feldman, S.J.; D. McClain & K. Palmer (1987) ‘Sources of Structural Changes in the United
States: an Input-Output Perspective’, The Review of Economics and Statistics, 69(3), 503-510.

Fingleton, B. & Mccombie, J. S. L. (1998) ‘Increasing returns and economic growth: Some
evidence for manufacturing from the European Union regions, Oxford Economic Papers, 50,
89–105.

178
Fleury, A. & Fleury, M. T. L. (2009) ‘Brazilian multinationals: surfing the waves of
internationalisation’, in R. Ramamurti & J. V. Singh, (Eds.), Emerging multinationals in
emerging markets, Cambridge: Cambridge University Press, pp.200-243.

Franco, G. (1998) ‘A inserção externa e o desenvolvimento’. Revista de Economia Política,

18(3), 121-147.

Frank, C. R.; Kim, K. S. & Westphal, L. E. (1975) ‘Economic Growth in South Korea since
World War II’, in Frank, C. R; Kim, K. S. & Westphal, L. E. Eds, Foreign Trade Regimes
and Economic Development: South Korea, New York : Columbia University press.

Frankel, M. (1962) ‘The production function in allocation and growth: a synthesis’. American
Economic Review, 52, 995-1022.

Freeman, C. & Soete, L. (1997) The Economics of Industrial Innovation, Cambridge,

Massachussets: The MIT Press.

Freeman, C. (1982a), The Economics of Industrial Innovation, second edition, Cambridge

(Mass.): MIT Press.

Freeman, C. (1982b), Technological Infrastructure and International Competitiveness,

Industrial and Corporate Change, 13 (3) [2004]:541-569.

Freeman, C. (1995) ‘The National System of Innovation in Historical Perspective’,

Cambridge Journal of Economics, 19(1), 5-24.

Freeman, C. (2004) ‘The national system of innovation in historical perspective’, Revista

Brasileira de Inovação, 3(1), 9-34.

Frenkel, R, Fanelli, J. M. & Rozenwurcel, G. (1992) ‘Crecimiento y reforma estructural en

América Latina, La Situación Actual’, in J. Vial, ed, Adonde va América Latina. Santiago:
CIEPLAN, 28-42.

Frenkel, R. (1992) Crecimiento y reforma estructural en América Latina, Santiago:

CIEPLAN.

Furtado, C. (1959) Formação econômica do Brasil: edição comemorativa – 50 anos, São

Paulo: Companhia da Letras, 2009.

Furtado, C. (1961) Desenvolvimento e Subdesenvolvimento, Rio de Janeiro, RJ: Fundo de

Cultura, 1965.

Furtado, C. (1964) ‘Dialética do Desenvolvimento’, Rio de Janeiro: Fundo de Cultura.

Furtado, C. (1967) Teoria e política do desenvolvimento econômico, São Paulo: Abril Cultura,
1983.

Furtado, C. (1995) Retour à la vision globale de Perroux et prebisch. Grenoble: Presse

Universitaires Grenoble.

179
Gereffi, G. (1989), ‘Comparative National Development: Theory and Facts for the 1990s’,
Sociological Forum, 4(4), Special Issue, 505-533.

Gibson, B. (2003) An essay on late structuralism, in A. Dutt and J. Ros, eds., Development
Economics and Structuralist Macroeconomics: Essays in Honor of Lance Taylor, 52–76,
Cheltenham, UK: Edward Elgar.

Gonçalves, R. (1996) ‘Globalização produtiva, investimento direto externo e empresas

transnacionais no Brasil: Uma perspectiva histórica’, Papers presented at The Brazilian
Association for Research in Economic History.

Gottschalk, R. & Prates, D. (2006), ‘East Asia's Growing Demand for Primary Commodities -
Macroeconomic Challenges for Latin America’, G-24 Discussion Paper Series, n.º 39, United
Nations, February.

Gouvêa, R. R. & Lima, G. T. (2010) ‘Structural change, balance-of-payments constraint, and

economic growth: evidence from the multisectoral Thirlwall’s law’, Journal of Post-
Keynesian Economics, 33(1), 169-204.

Gouvêa, R. R. & Lima, G. T. (2013) ‘Balance-of-payments-constrained growth in a

multisectoral framework: a panel data investigation’, Journal of Economic Studies, 40(2),
240-254.

Grossman, G. M. & Helpman, E. (1991) Innovation and Growth in the Global Economy,
Cambridge, Mass: MIT Press.

Guerrieri, P. & Meliciani, V. (2005) ‘Technology and international competitiveness: The

interdependence between manufacturing and producer services’, Structural change and
economic dynamics, 16, 489-502.

Guilhoto, J. (2009) Análise de Insumo-Produto: teoria, fundamentos e aplicações. Book in

preparation. Department of Economics, FEA-USP.

Guilhoto, J. (2010) Input-Output Analysis: Theory and Foundations. Book in preparation.

Department of Economics, FEA-USP.

Guilhoto, J.; Polenske, K. R, & Liu, H. (2010) ‘China and Brazil Productive Structure and
Economic Growth Compared: 1980's to 2000's’. Paper presented at the 57th North American
Meetings of the regional Science Assocation International: Denver, CO, November 10-13,
2010.

Guilhoto, J.; Sonis, M. & Hewings, G. J. D. (1996) ‘Linkages and multipliers in a

multiregional framework: integrations of alternative approaches’, Discussion Paper, 8,
Illinois: REAL.

Guilhoto, J.; Sonis, M.; Hewings, G. J. D. & Martins, E. B. (1994) ‘Índices de ligações e
setores-chave na economia brasileira: 1959/80’, Pesquisa e Planejamento Econômico, v. 2
(24), 287-314.

180
Haddad, E.; Hewings, G.; Leon, F. & Santos, R. C. (2007) ‘Building-up influence: post-war
industrialization in the State of Minas Gerais, Brazil’, Brazilian Journal of Political Economy,
vol. 27, nº 2 (106), 281-300.

Hamilton, N, A. (1790) Report on Manufactures; Papers on Public Credit Commerce and

Finance, New York: Columbia University Press, 1934.

Hansen, J. D. & Zhang, J. (1996) ‘A Kaldorian approach to regional economic growth in

China”, Applied Economics, 28, 679-685.

Harberger, A. (1988) ‘Growth, industrialization and economic structure: Latin America

and East Asia compared’, in H. Hughes, eds., Achieving Industrialization in East Asia,
Cambridge: Cambridge University Press.

Harris, N. (1987), The End of the Third World: Newly Industrializing Countries and the
Decline of an Ideology, New York: Viking Penguin.

Harris, R. I. D. & Lau, E. (1998) ‘Verdoorn’s law and increasing returns to scale in the UK
regions, 1968_91: some new estimates based on the cointegration approach’, Oxford
Economic Papers, 50, 201-219.

Harrod, R. F. (1933) International Economics, Nisbet & Cambridge University Press.

Hildreth, A. (1989) ‘The ambiguity of Verdoorn’s law: A case study of the British regions’,
Journal of Post Keynesian Economics, 11(2), 279–94.

Hiratuka, C. (2003) ‘Padrões de integração comercial das filiais de empresas transnacionais’

in M. Laplane; L. Coutinho & C. Hiratuka, Internacionalização e desenvolvimento da
indústria no Brasil. São Paulo: Editora da Unesp; Campinas: IE.

Hirschman, A. O. (1958), The Strategy of Economic development, New Haven, Conn: Yale
University Press.

Hirschman, A. O. (1981) Essays in Trespassing: Economics to Politics and Beyond, New

York: Cambridge University Press.

Hirschman, A. O. (1984) ‘A Dissenter´s Confession: The Strategy of Economic Development

Revisited’, in G. M. Meier and D. Seers (ed), Pioneers in Development, Oxford University
Press.

Hirschman, A. O. (1987) ‘Linkages’, in Eatwell, J., Milgate, M., and Newman, P. (eds.), The
New Palgrave: A Dictionary of Economics, Vol. 3, Macmillan, London and Basingstoke, 206-
211.

Ho, P. S. (2004) ‘Myrdal’s Backwash and Spread Effects in Classical Economics’. Jounal of
Economics Issues, nº 2.

Hughes, A. & Singh, A. (1991) ‘The world economic slowdown and the Asian and Latin
American economies: a comparative analysis of economic structure, policy and performance’,

181
in T. Banuri, (ed.), Economic liberalization: no panacea: the experiences of Latin America
and Asia. Oxford: Clarendon Press.

Imbs, J. & Wacziarg, R. (2003) ‘Stages of diversification’, American Economic Review, pp.
63-86.

Ioris, F. R. & Ioris, A. A. R. (2013) ‘Assessing development and the idea of development in
the 1950s in Brazil’, Brazilian Journal of Political Economy, 33(3), 411-426.

Iqbal, F. (1988) ‘External Financing for Korea: The Next Phase’, World Development, vol. 16
(1), p. 137-155

Jackson, W. A. (2003) ‘Social Structure in Economic Theory’, Journal of Economic Issues,

37 (3), 727-746.

Jayme, F. G. & Resende, M. F. C. (2009) ‘Crescimento econômico e restrição externa: teoria

e a experiência brasileira’, in Michel R. and Carvalho, L, eds., Crescimento econômico: setor
externo e inflação, Rio de Janeiro: IPEA.

Jones, L. P. & Sakong, I. L. (1980) Government, business and entrepreneurship in economic

development: the Korean case, Cambridge: Harvard University Press.

Kaldor, N. (1966) ‘Causes of the Slow Rate of Economic Growth of the United Kingdom: on
Inaugural Lecture’, Cambridge: Cambridge University Press.

Kaldor, N. (1967) Strategic Factors in Economic Development, New York: Ithaca.

Kaldor, N. (1970) ‘The case for regional policies’, Scottish Journal of Political Economy, 17,
337-348.

Kaldor, N. (1972) ‘The Irrelevance of Equilibrium Economics’, The Economic Journal, 82

(328):1237-1255.

Kaldor, N. (1981) ‘The Role of Increasing Returns, Technical Progress and Cumulative
Causation in the Theory of International Trade and Economic Growth’, Economie Appliquée,
34(6), 593-617.

Kaldor, N. (1986) ‘Limits on Growth’, Oxford Economic Papers, 38.

Kay, C. (1989) Latin American Theories of development and underdevelopment, London and
New York: Routledge.

Keynes, J. M. (1926) The end of laissez-faire, (Vol. 16), London: Hogarth Press.

Keynes, J. M. (1936) The general theory of interest, employment and money, London:
Macmillan.

Kim, J. & Kim, P. H. (2014) ‘Coordination and Capacity-Building in US Aid to South Korea,
1945–1975’, in Kim, E. M & Kim, P. H. Eds, The South Korean Development Experience:
Beyond Aid. New York: Palgrave Macmillan.
182
Kim, L. (1993) ‘National System of Industrial Innovation: Dynamics of Capability Building
in Korea’, in R. R. Nelson, National innovation systems: a comparative analysis, Oxford:
Oxford University Press.

Kim, L. (1997) Imitation to innovation: the dynamics of Korea`s technological learning.

Cambridge, Mass: Harvard Business Review Press.

Kincaid, H. (2008) ‘Individualism versus Holism’, in Steven N. Durlauf and Lawrence E.

Blume, eds., The New Palgrave Dictionary of Economics, Basingstok: Palgrave Macmillan.

Knell, M. (2004) ‘Structural change and the Kaldor-Verdoorn law in the 1990s’, Revenue
d’economie industrielle, 105,71-83.

Kruger, A. (1985) ‘The experience and lessons of Asia’s super-exporters’, in: V. Corbo,
A. Krueger & F. Ossa (Eds), Export-oriented Development Strategies, Boulder, CO:
Westview Press.

Krugman, P. & Obstfeld, M. (2006) International Economics: Theory and Policy, 7. ed.,
Boston, Massachusetts: Pearson, Addison-Wesley.

Kuczynski, P. P. & Williamson, J. (2003) ‘After the Washington Consensus: restarting

growth and reform in Latin America.Washington’. Institute of International Economics.

Kupfer, D.; Freitas, F. & Young, C. E. F. (2003) ‘Decomposição estrutural da variação do

produto e do emprego entre 1990 e 2001 - uma estimativa a partir das matrizes insumo-
produto’, Relatório de pesquisa para a CEPAL/Divisão de Indústria, IE/UFRJ.

Lafer, C. (2002) JK e o Programa de Metas (1956-61): processo de planejamento e o sistema

político do Brasil, Rio de Janeiro: Editora FGV.

Lall, S. (2001) Competitiveness, Technology and Skills, Cheltenham: Edward Elgar.

Laplane, M. & Sarti, F. (1997) ‘Investimento Direto Estrangeiro e a retomada do crescimento

sustentado nos anos 90’, Economia e Sociedade, IE/Unicamp, Campinas, 8, p.143-182.

Laplane, M. & Sarti, F. (2006) ‘Prometeu Acorrentado: o Brasil na Indústria Mundial no

início do século XXI’, in R. Carneiro (Ed.), A Supremacia dos Mercados e a Política
Econômica do Governo Lula, São Paulo: Editora UNESP.

Laplane, M. F.; Ferreira, A. N. & Borghi, R. A. Z. (2013) ‘Padrões de crescimento,

investimento e processos inovadores: o caso da Coreia do Sul’, in Padrões de
desenvolvimento econômico (1950–2008): América Latina, Ásia e Rússia. Brasília: Centro de
Gestão e Estudos Estratégicos (CGEE).

Laplane, M.; Sarti, F., Hiratuka, C. & Sabbatini, R. (2001), ‘La inversión extranjera directa
en el MERCOSUR: el caso brasileño’ in D. Chudnovsky (ed), El boom de inversión
extranjera directa em el Mercosur, Buenos Aires: Siglo Veintiuno de Argentina, p.123-208.

183
Lara-Resende, A. (1985) ‘A Moeda Indexada: Uma Proposta para Eliminar a Inflação
Inercial’, Revista de Economia Política, 5(2).

Lastres, H. M. M, Cassiolato, J. E. & Maciel, M. L. (2003) ‘Systems of Innovation for

development in the knowledge era’, in Cassiolato, J. E.; Lastres, H. M. M. and Maciel, M. L.
(eds), Systems of innovation and development, Cheltenham: Elgar.

Lee, W-Y (2000) ‘The role of Science, Technology, and Innovation in Korea’s Industrial
Development’, in L. Kim & R. R. Nelson, Technology, Learning, & Innovation: Experiences
of Newly Industrializing Economie, Cambridge: Cambridge University Press.

Leontief, W. (1936) ‘Quantitative Input and Output Relations in the Economic Systems of the
United States’, Review of Economics and Statistics, 18(3),105-25.

Leontief, W. (1941) The Structure of American Economy, Cambridge: Harvard University

Press.

Lessa, C. (1977) ‘Visão crítica do II PND’, Tibiriçá, 2(6), 47-72.

Lessa, C. (1983) Quinze anos de política econômica, São Paulo: Brasiliense.

Leung-Chuen, C. (2001) ‘South Korea: government-led development and the dominance of
giant corporations’, in K. Kai-Sun; C. Leung-Chuen; F. T, Lui & Larry D.Qiu, Industrial
development in Singapore, Taiwan, and South Korea. New Jersey, London, Singapore, Hong
Kong, Bangalore: World Scientific Publishing.

Lewis, W.A. (1954) ‘Economic development with unlimited supplies of labour’, in A.N.
Agarwala & S.P. Singh, (eds), The economics of underdevelopment, Oxford: Oxford
University Press.

Libanio, G. (2006) ‘Three essays on aggregate demand and growth’, PhD Thesis, University
of Notre Dame.

Lim, W. (2010) ‘The Emergence of the Chaebol and the Origins of the Chaebol Problem’, in
S. Haggard, W. Lim & E. Kim (Eds), Economic Crisis and Corporate Restructuring in Korea
Reforming the Chaebol. Cambridge: Cambridge University Press.

Lin, J. Y. (2012) New Structural Economics: A Framework for Rethinking Development and
Policy, Washington DC: The World Bank.

Lipton, M. (1968) ‘Strategy for agriculture: urban bias and rural planning’, in: P. Streeten and
M. Lipton, eds. The crisis of Indian planning. London: Oxford University Press.

List, F. G. (1841) Sistema nacional de economia política, São Paulo: Abril Cultural. 1983.

Lodefalk, M. (2010) ‘Servicification of Manufacturing - Evidence from Swedish Firm and

Enterprise Group Level Data’ Working Paper 3/2010, Örebro University, Swedish Business
School.

184
Love, J. (1995) ‘Economic ideas and ideologies in Latin America since 1930’, in L. Bethell,
ed., The Cambridge History of Latin America, Vol. 6: 1930 to the Present, Part 1: Economy
and Society, Cambridge: Cambridge University Press.

Love, J. (1996) ‘Las fuentes del estructuralismo latinoamericano’, Desarrollo Económico,

36(141), 391-402.

Love, J. (2005) ‘The rise and decline of economic structuralism in Latin America: new
dimensions’ Latin American Research Review, 40(3).

Lucas, R. E. (1988) ‘On the Mechanics of Economic Development’, Journal of Monetary

Economics, 22, 3-42.

Lundvall, B. A. (1992) National Systems of Innovation: Towards a Theory of Innovation and

Interactive Learning, London: Pinter Publishers.

Lundvall, B. A,; Johnson, B.; Andersen, E.S. & Dalum, B. (2002) ‘National systems of
production, innovation and competence building’, Research Policy, Volume 31, Issue 2,
February 2002, pp 213- 231.

Maddison, A. (1985) Two Crises: Latin America and Asia, 1929-1938 and 1973-1983, Paris:
OECD Development Centre.

Magacho, G. R., Rocha, I. L. & Marconi, N. (2014), ‘Do Primary Sectors Promote Economic
Growth? An Input-Output Analysis Contrasting Brazil and the Other BRICs’, Paper Presented
at 40th Annual Conference of the Eastern Economic Association (EEA) - Boston, Mass,
March 6-9, USA.

Malan, P. S. & Bonelli, R. (1983), ‘Crescimento Econômico, Industrialização e Balanço de

Pagamentos: O Brasil dos anos setenta e anos oitenta’, Texto para discussão nº 60, Rio de
Janeiro: Ipea/Inpes.

Mamgain, V. (1999) ‘Are the Kaldor-Verdoorn laws applicable in the newly industrializing
countries?’ Review of Development Economics, 3 (3), 295-309.

Marshall, A. (1920) Principles of Economics, 8th edition, London: Macmillan.

Mazzucato, M. (2013) The entrepreneurial State: Debunking Public vs. Private Sector Myths,
London: Anthem Press.

Mccarthy, I. & Anagnostou, A. (2004) ‘The impact of outsourcing on the transaction costs
and boundaries of manufacturing’, International Journal of Production Economics. 88. 61-71.

Mccombie, J. S. L. & De Ridder, J. (1983) ‘Increasing returns, productivity and output

growth: the case of the United States’, Journal of Post Keynesian Economics, 5(3), 373–87.

Mccombie, J. S. L. & Thirlwall, A. P. (1994) Economic growth and the balance-of-payments

constraints, New York: ST. Martin’s Press.

185
Mckinsey (2012), ‘Manufacturing the future: the next era of global growth and innovation”,
Mckinsey Global Institute, November.

Mikesell,, R. F. (1997) Explaining the resource curse, with special reference to mineral-
exporting countries. Resources policy, v. 23, n. 4, pp. 191-199.

Miller, R. E & Blair P. D. (2009) Input-Output Analysis: Foundations and Extensions. 2nd.
Edition. Cambrige: Cambridge University Press.

Moreira, M. M. (1999) ‘A Indústria Brasileira nos Anos 90. O que já se Pode Dizer?’, in F.
Giambiagi & M. M. Moreira (Eds.), A Economia Brasileira nos anos 90, Rio de Janeiro:
BNDES.

Musacchio, A. & Lazzarini, S. G. (2014) Reinventing State Capitalism: Leviathan in

Business, Brazil and Beyond, Cambridge: Harvard University Press.

Myrdal, G. (1957) Economic Theory and Underdeveloped Regions, New York: Harper and
Row.

Nelson, R. & Pack, H. (1999) ‘The Asian Miracle and Modern Growth Theory’, The
Economic Journal, n. 457 (July), 416-436.

Nelson, R. (1993) National Innovation Systems: a comparative analysis, New York, Oxford:
Oxford University.

Nelson, R. (1996) The sources of economic growth, Cambridge, Mass: Harvard University
Press.

Nelson, R. R & Winter, S. G (2002) ‘Evolutionary Theorizing in Economics’, Journal of

Economic Perspectives, 16 (2), p. 23-46.

Nelson, R. R & Winter, S.G. (1982) An Evolutionary Theory of Economic Change,

Cambridge, Mass: Harvard University Press.

Nolan, P. (2001) China and the global economy: national champions, industrial policy and
the big business revolution: Houndsmill: Palgrave.

Nordas, H. K. & Kim, Y. (2013) ‘The Role of Services for Competitiveness in

Manufacturing’, OECD Trade Policy Papers, nº 148.

Nurkse, R. (1953) Problems of Capital Formation in Underdeveloped Countries, Oxford:

Oxford University Press.

Ocampo, J. A & Parra, M. A. (2006) ‘The commodity terms of trade and their strategic
implications for development’, in Jomo K.S. (ed.), Economic Globalization, Hegemony and
the Changing World Economy during the Long Twentieth Century, Oxford University Press,
New Delhi and Oxford.

186
OECD/WTO (2013) ‘Interconnected economies: benefiting from global value chains’.
Preliminary Version, Genebra.

Oliveira, F, H, P., Jayme, F. G. & Lemos, M. B. (2003) ‘Increasing returns to scale and
international diffucion of technology: Na empirical study for Brazil (1976-2000)’, Texto para
Discussão, nº 211. Belo Horizonte: UFMG/Cedeplar.

O'Sullivan, E. & N. Mitchell (2013) 'What are the key international approaches to
understanding the future of manufacturing?', in The Future of Manufacturing, UK
Government Office of Science, London: BIS

Ozawa, T. (1991) ‘The dynamics of Pacific Rim industrialization: how Mexico can join the
Asian flock of 'flying geese', in R. Roett, (ed.), 129-154, Mexico's External Relations in the
1990s, Boulder and London: Rienner.

Ozawa, T. (1995) ‘The "Flying-Geese" Paradigm of FDI’, Economic Development and Shifts
in Competitiveness (revised). A background paper submitted to UNCTAD.

Ozawa, T. (2001) ‘The 'hidden' side of the 'flying-geese' catch-up mode: Japan's dirigiste
institutional setup and a deepening financial morass’, Journal of Asian Economics. 12(4),
471-491.

Pack, H. & Whestphal, L. (1986) ‘Industrial strategy and technological change: theory versus
reality’, Journal of Development Economics, 22, p.87-128.

Palma, J. G. (1987) ‘Structuralism’, in Eatwell, J.; Milgate, M. and P. Newman, eds, The
New Palgrave: A Dictionary of Economics, London: Macmillan.

Palma, J. G. (1996) ‘Does it make a difference to export micro-chips rather than potato-chips?
Comparing export structures in East Asia and Latin America’, UNCTAD Working Paper.

Palma, J. G. (2000) ‘The Magical Realism of Brazilian Economics: How to create a financial
crisis by trying to avoid one’, CEPA Working Paper Series III (17), Center for Economic
Policy Analysis New School for Social Research.

Palma, J. G. (2004) ‘Latin American during the second half of the 20th Century: from the ‘age
of extremes’ to the age of ‘end-of-history’ uniformity’, in H-J Chang (Ed.), Rethinking
Development Economics, London: Anthem Press.

Palma, J. G. (2005) ‘Flying-geese and lame ducks: Regional powers and the different
capabilities of Latin America and East Asia to ‘demand-adapt’ and ‘supply-upgrade’ their
export productive capacity’, University of Cambridge, Cambridge, UK.

Palma, J. G. (2005) ‘Four sources of “de-industrialization” and a new concept of the “Dutch
Disease”’, in J. A. Ocampo, eds, Beyond Reforms, Structural Dynamics and Macroeconomic
Vulnerability, Stanford, CA: Stanford University Press.

187
Palma, J. G. (2012) ‘Was Brazil’s recent growth acceleration the world’s most overrated
boom?’, Cambridge Working Papers in Economics (CWPE 1248).

Palma, J. G. (2013) ‘How the full opening of the capital account to highly liquid and unstable
financial markets led Latin America to two and a half cycles of mania, panic and crash’, in M.
Wolfson and G.A. Epstein (eds.), The Handbook of the Political Economy of Financial
Crises, Oxford: Oxford University Press.

Park, S. (1974) ‘On Input-Output Multipliers with Errors in Input-Output Coefficients’,

Journal of Economic Theory, 6, 399-403.

Park, S. H. & Chan, K. (1989) ‘A cross-country input-output analysis of intersectoral

relationships between manufacturing and services and their employment implications’, World
Development, 17(2), 199–212.

Pasinetti, L. (1981) Structural change and economic growth, Cambridge: Cambridge

University Press.

Perala, M. (2011) ‘Early Development Theory from Sun Yat-Sen to Ragnar Nurkse’, in R.
Kattel; J. A. Kregel and E. S. Reinert, eds, Classical Development Economics and its
Relevance for Today. London-New York: Anthem Press.

Perez, C. & Soete, L. (1988) ‘Catching-up in Technology: Entry barriers and Windows of
Opportunity’, in G. Dosi (Ed.), Technical Change and Economic Theory. London: Pinter.

Perez, C. (2002) Technological Revolutions and Financial Capital. The Dynamics of Bubbles
and Golden Ages, Cheltenham: Edward Elgar.

Perroux, F. (1939) ‘Pour un approfondissement de la notion de structure’, in: Mélanges

économiques et sociaux offerts à Emile Witmeur, Librairie du Recueuil Sirey.

Perroux F. (1950) ‘The Domination Effect and Modern Economic Theory’, Social Research,
17(2): 188–206.

Perroux, F., (1955) ‘Note on the Concept of Growth Poles’, in David L. McKee; Robert D.
Dean and William H. Leahy, eds., Regional Economics: Theory and Practice. New York: The
Free Press.

Pierper, U. (2003) ‘Sectoral Regularities of Productivity Growth in Developing Countries: A

Kaldorian Interpretation’, Cambridge Journal of Economics, 27, 831-50.

Pilat, D. & Wölfl, A. (2005) ‘Measuring the interaction between manufacturing and services’,
OECD Science, Technology and Industry Working Papers, 2005/5.

Pilat, D., Cimper, A., Olsen, B.O. & Webb, C. (2006) ‘The Changing Nature of
Manufacturing in OECD Economies’, OECD Science, Technology and Industry Working
Papers, 2006/9.

188
Pinto, A. (1965) ‘Concentración del progreso técnico y de sus frutos en el desarrollo de
América Latina’, El trimestre económico, nº 125, 3-69.

Pinto, A. (1970) ‘Naturaleza e implicaciones de la ‘heterogeneidad estructural’ de la

América Latina’, El trimestre económico, 37(145), 83-100.

Pinto, A. (1971) ‘El modelo de desarrollo reciente de la América Latina. El Trimestre

Económico, 38 (150), 477-498.

Pinto, A. (1976) ‘Notas sobre los estilos de desarrollo en América Latina’, Revista de la
CEPAL, l(1),167-214.

Pisano, G. & Shih, W.C. (2009) ‘Restoring the American Competitiveness’, Harvard Business
Review, July-August, 114-125.

Prebisch, R. (1949) ‘Estudo econômico da América Latina’, in R. Bielschowsky, eds,

Cinqüenta anos de pensamento na Cepal. São Paulo: Cepal/Cofecon/ Record.

Prebisch, R. (1950) The Economic Development of Latin America and its Principal Problems.
New York: United Nations.

Prebisch, R. (1959) ‘Commercial Policy in the Underdeveloped Countries’, American

Economic Review, 49, 251-273.

Rasmussen, P. (1956) Studies in Intersectoral Relations, Amsterdam: North Holland.

Ricardo, D. (1817) On the principles of political economy and taxation, London: Murray, 3rd
edition, 1821.

Robinson, J. (1956) The Accumulation of Capital, London: Palgrave Macmillan.

Rocha, I. L. (2011) ‘Reestruturação industrial no Brasil: uma análise da dinâmica comercial e

produtiva da economia’. MA thesis in Economics. Campinas: IE/Unicamp.

Rocha, I. L; Magacho, G. R. & Marconi, N. (2014) ‘Sectoral Capabilities and Productive

Structure: An input-output analysis of the key sectors of the Brazilian economy’. Paper
Presented at the 42th Brazilian Economics Meeting (ANPEC – 2014).

Rodrik, D. (1994) Getting interventions right: how South Korea and Taiwan grew rich,
NBER (National Bureau of Economic Research), Working Paper n. 4964. Cambridge
(Massachusetts): NBER.

Rodrik, D. (2004) ‘Industrial policy for the twenty-first century’, John H. Kennedy School of
Government, Harvard University.

Rodrik, D. (2006) ‘Industrial development: stylized facts and policies’, in Industrial

Development for the 21st Century, New York: United Nations.

Rodrik, D. (2013a) ‘Unconditional Convergence in Manufacturing’, Quarterly Journal of

Economics, 128, 165-204.
189
Rodrik, D. (2013b) ‘Structural Change, Fundamentals, and Growth: An overview’, Institute
for Advanced Study, September 2013.

Romer, P. M. (1986) ‘Increasing Returns and Long-Run Growth’, Journal of Political

Economy, 94, 1002-1037.

Romer, P. M. (1987) ‘Growth Based on Increasing Returns due to Specialization’, American

Economic Review, 77, 56–62.

Romer, P. M. (1990) ‘Endogenous Technological Change’, Journal of Political Economy, 98,

71–102.

Rose, A. & Casler, S. (1996) ‘Input-Output Structural Analysis Decomposition: a critical

appraisal’, Economic Systems Research, 8(1), 33-62.

Rosenberg, N. (1963) ‘Technological Change in the Machine Tool Industry, 1840–1910’, The
Journal of Economic History, 23, 414-443.

Rosenstein-Rodan, P. (1943) ‘Problems of Industrialisation of Eastern and South-Eastern

Europe’, Economic Journal, 53(210/1) 202-11.

Rosenstein-Rodan, P. (1944) ‘The International Development of Economically Backward

Areas’, International Affairs (Royal Institute of International Affairs), 20(2), 157-165.

Rosenstein-Rodan, P. (1961) ‘Notes on the Theory of the Big Push’, in H.S. Ellis and H.C.
Wallich, (eds), Economic Development in Latin America, New York: Macmillan.

Rosenstein-Rodan, P. (1984) ‘Natura Facit Saltum: Analysis of the Disequilibrium Growth

Process’, in G. M. Meier and D. Seers, eds, Pioneers in Development. Oxford-New York:
Oxford University Press.

Rostow, W. W. (1978) The World Economy, Austin: University of Texas Press.

Rostow, W. W. (1952) The Process of Economic Growth, New York: W. W. Norton & Co.

Rowthorn R. & Wells J. (1987) De-industrialisation and Foreign Trade, Cambridge,

Cambridge University Press.

Rowthorn, R. & Coutts, K. (2004) ‘Deindustrialization and the Balance of Payments in

Advanced Economies’, Cambridge Journal of Economics, 28(5): 767-90.

Rowthorn, R & Coutts, K. (2013a) 'Update of Prospects for the UK Balance of

Payments', in The Future of Manufacturing, UK Government Office of Science, London:
BIS.

Rowthorn, R & Coutts, K. (2013b) 'De-industrialisation and the Balance of Payments in

Advanced Economies', in The Future of Manufacturing, UK Government Office of Science,
London: BIS.

190
Rowthorn, R & Coutts, K. (2013c) 'Re-industrialisation: A Commentary', in The Future of
Manufacturing, UK Government Office of Science, London: BIS.

Rowthorn, R. (1994), ‘Korea at the cross-roads’, Working Paper No. 11, Cambridge: Centre
for Business Research.

Rynn, J. (2010) Manufacturing Green Prosperity: The Power to Rebuild the American Middle
Class, Santa Barbara, CA: Praeger Press.

Sarti, F. & Laplane, M. F. (2002) ‘O investimento direto estrangeiro e a internacionalização

da economia brasileira nos anos 90’, Economia e Sociedade, v. 11, n. 1 (18), p. 129-164.

Saxena, P. K. & Sharma, A. (2014) ‘Role of Machine Tools Industry in Economic

Development’, International Journal of Enhanced Research in Science Technology &
Engineering, 3(5), 188-193.

Schumpeter, J. A. (1912) The theory of economic development: an enquiry into profits,

capital, credit, interest, and the business cycle, New Brunswick, NJ: Transaction Publishers,
2004.

Schumpeter, J. A. (1942) Capitalism, socialism and democracy, London: Routledge, 1994.

Seong, S. (2001) ‘Korea’s industrial development, globalization, and prospects for regional
and global integration’, Korea Development Institute (KDI), Working Paper 2001-03, Seoul,
Korea.

Serra, J. (1982) ‘Ciclos e mudanças estruturais na economia brasileira do pós-guerra’, in

Belluzzo, L. G. & Coutinho, R, Eds, Desenvolvimento Capitalista do Brasil: Ensaios Sobre a
Crise, vol.1, São Paulo: Brasiliense.

Silverberg, G. & Verspagen, B. (1995) ‘An evolutionary model of long-term cyclical

variations of catching up and falling behind’, Journal of Evolutionary Economics, 5(4), 209-
227.

Simonsen, M. H. (1970) Inflação: Gradualismo Versus Tratamento de Choque, Rio de

Janeiro: APEC Editora.

Singer, H. W. (1950) ‘The distribution of gains between investing and borrowing countries’,
American Economic Review, 40 (2), 473-485.

Singh, A. (1985) ‘The world economy and the comparative economic performance of large
semi-industrial countries’, Asian Regional team for Employment Promotion (mimeo),
Bangkok: International Labour Organization,

Skolka, J. (1989) ‘Input-Output Structural Decomposition for Austria’, Journal of Policy

Modelling, 11(1), 45-66.

Snowdon, B. & Vane, H. R (2005) Modern Macroeconomics: Its Origins, Development and
Current State. Cheltenham: Edward Elgar.

191
Solow R (1950) ‘Labor productivity functions in meat packing’, The Review of Economics
and Statistics, (8/1), MIT, Cambridge.

Solow, R. M. (1956) A Contribution to the Theory Economic Growth, Quarterly Journal of

Economics, 70, 65-94.

Solow, R. M. (1957) Technical Change and the Aggregate Production Function, Review of
Economics and Statistics, 39, 312–320.

Sonis, , M. & Hewings, G. J. D. (1989) ‘Error and Sensitivity Input-Output Analysis: A New
Approach’, in Miller, R. E., Polenske, K. R. & Rose, A. Z. (eds.), Frontiers of Input-Output
Analysis. New York, Oxford University Press.

Sonis, M. & Hewings, G. J. D. (1994) ‘Interpreting Spatial Economic Structure and Spatial
Multipliers: Three Perspectives’, Geographical Analysis, 26, 124-151.

Stallings, B. & Peres, W. (2000) Growth, Employment and Equity: the Impact of the
Economic Reforms in Latin America and the Caribbean, Washington, D.C: Brookings
Institution.

Stehrer, R. (2013) ‘On the role of BRIC countries in providing global manufacturing output’,
in The Future of Manufacturing, UK Government Office of Science, London: BIS

Street, J. H. e James, D. D. (1982) ‘Institutionalism, Structuralism, and Dependency in Latin

America’, Journal of Economic Issues, 16(3), 673-689.

Studart, R. (1995) Investment Finance in Economic Development, London: Routledge.

Sunkel, O. (1989) ‘Structuralism, Dependency and Institutionalism: An Exploration of

Common Ground and Disparities’ Journal of Economic Issues, 23(2), 519-533.

Suzigan, W. (1988), ‘Estado e Industrialização do Brasil’, Revista de Economia Política, 8(4),

5-16.

Swan, T. W. (1956) ‘Economic Growth and Capital Accumulation’, Economic Record, 32,
334–361.

Szirmai, A. & Verspagen, B. (2011) ‘Manufacturing and Economic Growth in Developing

Countries, 1950-2005’, UNU-MERIT Working Paper Series, 2011-069.

Szirmai, A. (2012) ‘Industrialisation as an Engine of Growth in Developing Countries, 1950-

2005’, Structural Change and Economic Dynamics, 23(4), 406-20.

Tavares, M. C. (1972) ‘Auge e declínio do processo de substituição de importações no

Brasil’, in M. C. Tavares, ed, Da substituição de importações ao capitalismo financeiro. Rio
de Janeiro: Zahar.

192
Thirlwall, A. P. (1979) ‘The Balance of Payments Constraint as an Explanation of
International Growth Rate Differences’, Banca Nazionale del Lavoro Quaterly Review, 128,
45-53.

Thirlwall, A. P. (2013) Economic growth in an open developing economy: The role of

structure and demand, Cheltenham: Edward Elgar.

Timmer, M. P. & G. J. de VRIES (2007) ‘A Cross-country Database for Sectoral Employment

and Productivity in Asia and Latin America, 1950-2005’, Research Memorandum GD-98,
Groningen Growth and Development Centre.

Timmer, M. P. & G. J. de VRIES (2009) ‘Structural change and growth accelerations in Asia
and Latin America: a new sectoral data set’, Cliometrica, 3(2), 165-190.

Timmer, M. P. & Szirmai, A. (2000) ‘Productivity growth in Asian manufacturing: the

structural bonus hypothesis examined’, Structural Change and Economic Dynamics, 11, 371-
392.

Toner, P. (1999), Main Currents in Cumulative Causation: The Dynamics of Growth and
Development, New York: St. Martin’s Press.

Trebat, T. J. (1983) Brazil's state-owned enterprises: a case study of the state as

entrepreneur. Cambridge: Cambridge University Press.

Tregenna, F. (2007) ‘Which Sectors Can be Engines of Growth and Employment in South
Africa?: An Analysis of Manufacturing and Services’, Paper presented at the UNU-WIDER
CIBS Conference on Southern Engines of Global Growth, Helsinki, 7-9 September.

Tregenna, F. (2009) ‘Essays on sectoral structure and growth’, PhD Thesis, University of
Cambridge.

Tregenna, F. (2012) ‘Sources of Subsectoral Growth in South Africa’, Oxford Development

Studies, vol. 40(2), 162-189.

UNCTAD (1996), Trade and Development Report, Geneva.

UNCTAD (2002) Trade and Development Report 2002: Developing countries in world trade.
New York and Geneva: United Nations.

UNCTAD (2013) Global Value Chains and development. Investment and Value added trade
in the global economy, New York and Geneva: United Nations.

UNIDO (2013) Sustaining Employment Growth: The role of manufacturing and structural
change, Industrial development report 2013, Vienna, Austria.

UNIDO (2013) Sustaining Employment Growth: The role of manufacturing and structural
change, Industrial development report 2013, Vienna, Austria.

193
Verdoorn, P. J. (1949) ‘On the factors determining the growth of labour productivity’, in L.
Pasinetti, ed, Italian Economic Papers, Oxford University Press, 45-53.

Verspagen, B. (2000) ‘Growth and Structural Change: Trends, Patterns and Policy Options’,
Working paper series 00.08, Eindhoven Center for Innovation Studies (ECIS).

Vittucci, G. M. (2008) ‘Input-output data and services outsourcing’ A reply to Dietrich,

McCarthy and Anagnostou, Working Papers: 621, Universita' di Bologna.

Wade, R. (1992) ‘East Asia’s Economic Success, Conflicting Perspectives, Partial Insights,
Shaky Evidence’, World Politics, 44(2), Jan, 270-320.

Walker, R. A. (1985) 'Is there a service economy? The changing capitalist division of labour',
Science and Society, Vol. 49, No. 1 (Spring, 1985), pp. 42-83.

Wells, H. & Thirwall, A. P. (2003) ‘Testing Kaldor’s laws across the countries of Africa’,
African Development Review, 15(2-3), 89-105.

Williamson, J. (1989) ‘What Washington Means by Policy Reform’, in J. Williamson (ed.),

Latin American Adjustment: How Much Has Happened? Washington, DC: Institute for
International Economics.

WTO (2011) Understanding the WTO, Geneva: WTO- Information and Media Relations
Division.

Young, A. (1928) ‘Increasing returns and economic progress’, The Economic Journal, 38,
(152), 527-42.

Zakariah, A. R. & Ahmad, E. E. (1999) ‘Sources of industrial growth using factor

decomposition approach: Malaysia, 1979-87’, The developing economies, XXXVII-2, 162-
196.

Zysman, J, & Doherty, E. (1995) ‘The Evolving Role of the State in Asian Industrialization’,
UCAIS Berkeley Roundtable on the International Economy, Working Paper 84, November
1995.

194