The Dragon From The Mountains

The Dragon from the Mountains
China has agreed to invest more than $60 billion in Pakistan, in roads, rail,
energy and a deep-water port at Gwadar. This is unprecedented relative to
decades of minimal foreign direct investment (FDI) entering Pakistan. This is the
China–Pakistan Economic Corridor (CPEC). Support for the CPEC in Pakistan
is widespread and encompasses much of academia, the military, the mainstream
political leadership and civil society. Supporters argue that the CPEC offers the
potential to transform Pakistan and support rapid, equitable and sustainable
economic growth. Detractors of the CPEC argue that it will more likely tip
Pakistan into a dependent debt-relationship with China and that it will facilitate
more Chinese imports into Pakistan, posing a threat to Pakistan’s industrial base.
The Dragon from the Mountains utilises an in-depth understanding of economic
change in contemporary China and Pakistan, and economic theory and studies of
big infrastructure projects from the contemporary and historical world to evaluate
these contrasting views about the CPEC.
Matthew McCartney is Associate Professor in the Political Economy and Human

Development of South Asia at the University of Oxford (2011–21). He was also
the Director of South Asian Studies (2011–18) at the university. His two most
recent books are The Indian Economy, 1947–2017 (2019) and New Perspectives on
Pakistan’s Political Economy: State, Class and Social Change (co-edited, 2019).
The Dragon from the
Mountains
The CPEC from Kashgar to Gwadar
Matthew McCartney
University Printing House, Cambridge CB2 8BS, United Kingdom
One Liberty Plaza, 20th Floor, New York, NY 10006, USA
477 Williamstown Road, Port Melbourne, vic 3207, Australia
314 to 321, 3rd Floor, Plot No.3, Splendor Forum, Jasola District Centre, New Delhi 110025,
India
103 Penang Road, #05–06/07, Visioncrest Commercial, Singapore 238467
Cambridge University Press is part of the University of Cambridge.

It furthers the University’s mission by disseminating knowledge in the pursuit of
education, learning and research at the highest international levels of excellence.
www.cambridge.org
Information on this title: www.cambridge.org/9781108834155
© Matthew McCartney 2021
This publication is in copyright. Subject to statutory exception
and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.
First published 2021
Printed in India
A catalogue record for this publication is available from the British Library
ISBN 978-1-108-83415-5 Hardback

Cambridge University Press has no responsibility for the persistence or accuracy
of URLs for external or third-party internet websites referred to in this publication,
and does not guarantee that any content on such websites is, or will remain,
accurate or appropriate.
To Ayesha and Ali,
Thanks for the stomach-filling Lahori food and
the pen-inspiring research insights
Contents
List of Maps and Figures ix

List of Tables xi
Preface xiii
Acknowledgements xvii
1. Introduction 1
2. Big Infrastructure: Big Problems or Big Benefits? 27
3. CPEC Spillovers Rippling Outwards 49
4. Through the Eyes of Who? Evaluating the Success of the CPEC 78
5. The Dragon Uncoils: Special Economic Zones (SEZs) from

Shenzhen to Africa 114
6. The Dragon’s Embrace: Pakistan–China Trade Policy 145
7. The Will of the Dragon: The Importance of an Industrial Policy 161
8. Conclusion: The Way of the Dragon or the Way of the Falcon? 219
Bibliography226
Index256
vii
Maps and Figures
Maps
1.1 The Major Projects of the CPEC 2
1.2 The Chinese Belt and Road Initiative (BRI) 17
Figures
1.1 GDP growth trends, 1960–2016 12
3.1 Pakistan GDP growth (annual %) 57
3.2 Pakistan gross capital formation (% of GDP) 58
3.3 A comparison with India 58
4.1 Tax revenue as a share of GDP in Pakistan, 1970–2017 110
6.1 Pakistan: imports from and exports to China, 2003–17 153
7.1 Pakistan, foreign direct investment, net inflows (current US$) 174
7.2 Savings and investment in Pakistan, 1988–2018 199
7.3 Government consumption and tax revenue in Pakistan,
1990–2015 199
7.4 Household final consumption expenditure as a share of GDP 204
ix
Tables
1.1 GDP growth, 1990–2011 11

1.2 GDP growth stability analysis 13
2.1 World Bank overall logistics performance index 31
2.2 World Bank logistics performance index: Pakistan 32
2.3 World Bank quality of trade- and transport-related
infrastructure 32
3.1 Saving in terms of distance (via Shanghai versus Gwadar) 50
5.1 T-shirt making costs 124
6.1 Tariff comparison of Pakistani exports to China, selected
products, 2013 158
7.1 TFP growth, 1970–2011 169
7.2 Labour productivity growth, 1990–2011 170
7.3 Pakistan: World Bank Doing Business ranking 181
7.4 Pakistan: measures of competitiveness 182
7.5 Corruption perceptions rank and score, 2018 183
7.6 Pakistan corruption perceptions over time 184
7.7 Pakistan: Country Policy and Institutional Assessment (CPIA) 185
7.8 World Justice rule of law index: Pakistan, order and security 186
7.9 World Justice rule of law index: Pakistan, regulatory
enforcement 187
7.10 World Justice rule of law index: Pakistan, constraints on
government power 187
7.11 Central government gross debt (% of GDP) 212
xi
Preface
There is excitement in Pakistan, the academic and political equivalent of winning

the cricket world cup. After years of failing to meet short-term International
Monetary Fund (IMF) programmes, Pakistan has been promised a well-financed,
long-term developmental partnership. This is the China–Pakistan Economic
Corridor, or the CPEC. Between 1970 and 2001, a desultory $7 billion of FDI
dribbled into Pakistan. China has promised to invest more than $60 billion in
roads, railways, energy, industrial parks and other projects between 2015 and 2030.
China promises this will not be driven by IMF-style conditionalities but that the
CPEC will be tailored to Pakistan’s domestic political and economic agenda.
The Government of Pakistan has proclaimed in a succession of government
plans that upgrading infrastructure is a priority to promote rapid and sustainable
economic growth. Practical efforts to follow these goals through in practice
continually failed in response to economic crisis, IMF-motivated budget cuts
or the lack of sustained political will. Here is the political will. The Dragon
from the Mountains. China has more than 40 years’ experience of fuelling its
own rapid industrial and export-led growth, supported by massive investment in
infrastructure. China is committed to the long-term. The US was ever fickle and
committed less to Pakistan than to wider geopolitics in Afghanistan, Soviet Russia
or Iran, in which Pakistan occasionally and accidentally proved useful. The CPEC
can be told as part of a wider story, that of the end of the US-led world order and
the creation of a new Eurasian supercontinent headed by China (Macaes 2018a,
2018b). What part will Pakistan play in this story?
Some CPEC supporters have daydreamed that the CPEC can help spur
Pakistan into emulating the rapid economic growth of the Tiger or Dragon
economies of the 1960s and 1970s and becoming, perhaps, a ‘Falcon Economy’.
The detractors are equally adamant. They variously claim that the CPEC is
an economic fig-leaf to cover the real geopolitical intentions of China, to
access oil directly from the Middle East and to gain control of the deep-sea
port at Gwadar in southwest Pakistan and near to the Gulf. Some have
labelled the CPEC as ‘predatory lending’, intending to push Pakistan into a
xiii
PREFACE
debt trap to increase Chinese leverage over Pakistan’s domestic and foreign
policy. Remember, say the detractors, the Suez Canal that tipped Egypt
into a nineteenth-century debt crisis, and eventual colonisation by Britain.
The problem with the economic debate is that so much of it is driven by
aspiration, hope and excited proclamation rather than rigorous analysis.
Commentators have tended to present the economy of Pakistan as either a
disaster that needs to be rescued by the Chinese and the CPEC or else as an
economy of wonderful potential that will be liberated by the opportunities of
the CPEC. While the excitement is perhaps understandable, there is need for
economics, theory and evidence to feed into the policy discussion in Pakistan.
This is especially timely as a new government was elected in Pakistan in 2018
which some commentators think has been interested in renegotiating aspects
of the CPEC.
One aspect of the careful thinking in this book comes from looking more
closely at China, not just as a source of financing and FDI, but at the economic
changes that are happening in Western China, that part of China to which
Pakistan will be tethered by the CPEC infrastructure. New roads do not just
make it easier for firms in Pakistan to export, they also make it easier for
Chinese firms to export to Pakistan. It makes a big difference, for example,
if industrial enterprises are opening in Western China and are readying
themselves to export high-quality and low-cost production to Pakistan. What
hope then for Pakistani industry? This book looks in some detail at economic
development in the western Chinese province of Xinjiang and thinks carefully
about the implications for Pakistan.
A deeper sense of history may give us some perspective and also highlight
the real potential of big infrastructure to promote long-run sustainable
economic development. There is another crucial body of evidence we can use
to help our thinking about the CPEC. There are huge numbers of very detailed
and rigorous economic studies of big infrastructure projects, dating from the
nineteenth century to the contemporary world. Almost none of these has yet
been drawn upon in discussions of the CPEC. That is something this book
will undertake for the first time. This will help take thinking about the CPEC
away from rhetorical political flourish and into more rigorous discussion
rooted in economic theory and economic evidence. The Government of China
promises that the CPEC will be a win–win partnership. The Government of
Pakistan promises that the CPEC will help integrate even the most backward
areas of Pakistan into economic growth and development. The historical
xiv
PREFACE
and contemporary literature looking at big infrastructure demonstrates that

such optimism is naive. Big infrastructure always generates both winners and
losers. Whether the losers can be compensated or whether they can migrate to
areas of new CPEC-induced opportunities will have a crucial bearing on the
long-term political viability of the CPEC. This book examines the important
lessons for the case of contemporary Pakistan.
One of the promises of the CPEC are the nine industrial parks or special
economic zones (SEZs) planned as part of the project and spread evenly across
Pakistan, to ensure, argues the Government of Pakistan, that every corner of the
country enjoys the benefits of the CPEC. There is a tendency among CPEC
commentators to list the resource endowments in the locality of each planned
SEZ and to assume that this will be enough to attract industrial enterprises
to process and export those endowments. If there is fruit near a planned SEZ
then there must be fruit-processing FDI waiting to pounce. Such writings
usually forget to note that there are thousands of such SEZs globally and do
not make the case for why invest in Pakistan, which comes near the bottom
of many international league tables for investment security. Pakistan itself has
decades of trying and largely failing to stimulate industrialisation through
SEZs; why are they suddenly going to work now? This book draws on
an extensive literature that has analysed in detail the functioning of SEZs
across various countries in Africa; the China–Africa engagement is well into
its second decade and offers important lessons for Pakistan. The outcome of
these efforts have hitherto been ignored in the CPEC scholarship.
This book is a study of the CPEC but ranges far beyond the Pakistan-
centric literature that is commonly used in related studies. This book delves
into the history and contemporary experience of big infrastructure, thinks
carefully about the economic development of Western China, draws from the
long-established Chinese investment projects across Africa, and grounds this
in a careful political economy understanding of contemporary Pakistan.
Will the CPEC turn Pakistan into a Falcon Economy or tip Pakistan into
debt-induced dependency on China?
xv
Acknowledgements
To Ranjana, of course, who was ever present and ever feigning enthusiasm for
infrastructure and spillovers. Thanks for some wonderful hospitality in Pakistan
where much of the rumination for and writing of this book took place. Ayesha
and Ali and wonderful colleagues and friends from Faysal Bank. Imdad and the
23rd batch. Mueen, the Malangs and a much more senior batch. Mike and Libby,
Andy and Marcela, and Jon and Varuni for their friendship and a wonderful home
at Aitchison College. The Rector Shahid Amjad Chaudhary, Professor Rashid
Amjad and also my dear friends Azam, Theresa and Waqar, all at the Lahore
School of Economics. In Japan, the generous academic hospitality and friendship
from the Institute of Economic Research (IER) at the University of Hitotsubashi,
in particular Professor Takashi Kurosaki and Michie Kano. Also at Oxford
where my wonderful colleagues Maryam Aslany, George Kunnath, Barbara
Harriss-White, Stephen Minay, Ali Jan and Imre Bangha not only coped but
thrived happily in my absence. To Anwesha Roy and Qudsiya Ahmed for generous
support and encouragement from submission to publication and the copy-editor
for incisive copy-editing, all from Cambridge University Press in New Delhi.
xvii
1
Introduction
The introduction first introduces and explains the China–Pakistan Economic

Corridor (CPEC), what it is, how much it will cost and what its geographical
contours and timewise evolution are. The introduction then places the CPEC
into five important background contexts that will be referenced throughout
the rest of this book. These are (a) the economic optimism the CPEC has
generated in Pakistan; (b) the long friendship between China and Pakistan;
(c) the economic development of Pakistan since 1947; (d) the evolution of
China’s political economy since the death of Mao in 1976; and (e) the global
history of big infrastructure projects. Finally, the introduction summarises
some of the key questions and findings of the chapters of the book.
WHAT IS THE CPEC?

The CPEC refers to a massive package of investment that has been promised
to Pakistan. The investments are inspired by a bigger Eurasia-wide Chinese
vision, that of the New Silk Road. This vision has subsequently become
known as the Belt and Road Initiative (BRI). China has promised to provide
much of the funding to Pakistan upfront, though controversy remains about
whether, how and on what terms that funding will be repaid. The investments
are concentrated in energy, transport infrastructure and the construction of
special economic zones (SEZs) to promote industry. Map 1.1 shows some of
the main projects of the CPEC. The nearly 3,000 kilometres (1,800 miles)
of roads and rail from Kashgar in China to Gwadar in southern Pakistan
can be clearly visualised. The CPEC includes oil and gas pipelines, railways,
highways, SEZs and fibre-optic networks (Sial 2014).
1
THE DRAGON FROM THE MOUNTAINS
Map 1.1 The Major Projects of the CPEC

Source: Council on Foreign Relations and Government of Pakistan.
Note: Map not to scale and does not represent authentic international boundaries.
The CPEC was initially projected to cost $46 billion, of which 71 per cent
($32 billion) was to be invested in energy, 8 per cent in rail, 13 per cent in
road links and 4 per cent in the Gwadar port (Boyce 2017: 12). By 2017, this
total had been raised to $62 billion. These numbers give a false impression of
precision about the CPEC. As with the entire BRI project, it is difficult to pin
the CPEC down, as Shafqat and Shahid argue, ‘Identifying and explaining
the various components of the CPEC is a tedious and complex task because
the information is not readily available, is scattered across sources or changes
frequently’ (2018: 24). Even within the same sources there are overlapping
and ambiguous lists, project lists are unstable and frequently updated and
existing projects are often repackaged as CPEC projects (Shafqat and Shahid
2018: 25).
The CPEC has been characterised by the Government of Pakistan in
baffling spatial terms. The Government of Pakistan’s (2017: 4) Long Term
Plan on the CPEC describes the spatial geography of the CPEC as ‘one belt,
three axes and several passages with a core zone and adjoining radiation zones’.
The core zone represents the belt and includes from China, ‘Kashgar, Tunshuq
city Atushi city and also Akto county in the Kizilsu Korghiz autonomous
2
INTRODUCTION
region of Xinjiang and in Pakistan Islamabad, parts of Punjab, Sindh, KPK,

Azad Kashmir, and Gilgit Baltistan’. Various node cities are also included.
There are also three axes that are horizontal links between Peshawar and
Lahore, Quetta and Sukkur, Karachi and Gwadar. Space is closely linked to
resources. The northwest and west of Pakistan are scheduled for the extraction
of minerals including gold, diamonds and marble; the central zone across
Punjab and Sindh for textiles, cement and household electronics; the southern
zone, including the coastal area around Gwadar up to Karachi, is to include
petrochemicals, harbour industry and iron and steel (Shafqat and Shahid
2018). The CPEC has also and somewhat less confusingly been characterised
over time. The CPEC projects have been divided up into short-term plans
due to be completed by 2020, medium-term (by 2025) and long-term (by
2030). The CPEC included a list of ‘early harvest’ projects due for completion
within the first few years after the programme was launched in 2015. The
list of mainly energy projects already underway by 2018 is impressive. These
include the 2,660 megawatt (MW) coal-fired power stations at Port Qasim
in Karachi, finished 67 days ahead of schedule; the 2,660 MW coal-fired
power plants at Sahiwal, Punjab, completed in October 2017, less than two
years after financial closure; and the Karot hydropower station, which is
expected to be operational by the end of 2021, almost four years after financial
closure (Government of Pakistan 2018). The CPEC can also be considered
by sector. It is difficult to think of an economic sector in Pakistan that has
not been subsumed by the CPEC label. The CPEC includes demonstration
projects in agriculture whereby Chinese companies are supposed to set up
best practice farms using modern irrigation, technology and livestock and
thereby provide learning opportunities for Pakistani farmers. Beyond lessons,
Chinese investment in agriculture is planned to encompass the entire supply
chain of agriculture, from seeds, fertiliser and growing to post-harvest storage
and transportation. Chinese companies will be given incentives to invest
in the newly opening SEZs including in telecommunications, mining and
minerals as well as consumer durables including home electronics and cell
phones. The roads and railways are the most manifest forms of transport but
the promise of CPEC ‘spillovers’ can be interpreted much more widely. Also
planned are a fibre-optic network, internet connectivity and cultural links such
as the promotion of Chinese television to provide a better understanding of
Chinese social life. Other projects include Baltistan University, the China–
Pakistan Joint Research Centre in Earth Sciences and the establishment of
3
research collaboration between Chinese and Pakistani medical institutions.

A Safe City Project has been started in Peshawar, which is structured around
monitoring and surveillance of urban areas with the use of explosive detectors
and scanners. The project is scheduled to be extended to other cities (Shafqat
and Shahid 2018: 34–7).
CPEC CONTEXT I: ECONOMIC OPTIMISM AND

THE LAUNCH OF A FALCON ECONOMY?
The first story about the CPEC is that of a sense of economic optimism
looking forward. The most enthusiastic supporters suggest that the CPEC
can help turn Pakistan into a new Asian miracle economy and so replicate
the earlier economic success of South Korea, Taiwan, Hong Kong, China
or Vietnam. We have had the Asian ‘Tiger’ economies, the Asian ‘Dragons’
and some have even started discussing African ‘Leopards’. Even the Chinese
government dangled the prospect of the CPEC helping Pakistan to become
an ‘Asian Tiger’ (Government of China 2015b: 5). Will we soon be discussing
the Pakistani ‘Falcon’ economy?
Shafqat and Shahid (2018: 16) have conducted a domestic accounting
exercise and divided domestic Pakistan politics up into ‘CPEC Enthusiasts’,
‘CPEC Opponents’ and ‘CPEC Reformers’. The ‘CPEC Enthusiasts’
seem the most numerous and vocal and are catalogued to include political
leaders from both the current (Pakistan Tehreek-e-Insaf ) and previous two
governments (Pakistan Muslim League [Nawaz] and Pakistan Peoples Party),
all of whom have been in power to negotiate crucial components of the CPEC,
government officials and many journalists. The Enthusiasts are often heard to
repeat the mantra that the ‘CPEC is a game-changer’ and that the CPEC
will bring about economic transformation and that this in turn will bring
peace to Pakistan. The Government of Pakistan has described the CPEC in
terms of its anticipated positive outcomes. The CPEC is ‘a growth axis and
development belt featuring complementary advantage, collaboration, mutual
benefits and common prosperity’ (Government of Pakistan 2017: 4). Such
optimism is supported by many scholars. The CPEC ‘will be a harbinger of
economic prosperity and well-being for Pakistan, China and the neighbouring
states’ (Hali, Shukui and Iqbal 2015: 160). The politics of these claims have
sometimes been steeped in nationalism and Enthusiasts have been known to
equate criticism or concerns about the CPEC with conspiracy against Pakistan
4
INTRODUCTION
and its friendship with China. There is clear and widespread support for the
CPEC in Pakistan which was sustained through the change of government in
Pakistan in the 2014 and 2018 national elections. The military have also gone
public on their strong backing (Sial 2014).
There are good and objective reasons for this optimism. The total value
of the CPEC (first estimated at $46 billion and now projected at more than
$60 billion) is a massive multiple of the cumulative $7 billion Pakistan received
in foreign direct investment (FDI) between 1970 and 2001 (Atique, Khan
and Azhar 2004: 709). The promised long-term commitment from China to
Pakistan dwarves any other foreign relationship for Pakistan since independence.
At the same time, there is no clear case to be made that the CPEC is nothing
but an externally imposed agenda. The CPEC builds on the long-standing
domestic government policy in Pakistan towards infrastructure investment,
especially in energy. There is good evidence that the quality of infrastructure
in Pakistan is poor relative to other large developing countries and has become
a significant constraint on economic growth (Loayza and Wada 2012). The
energy projects plan to add over 10,000 MW to capacity. This has already made
an immediate impact on reducing the chronic energy shortages Pakistan faced
over the preceding decade. This shortage necessitated that firms install high-
cost diesel generators which undermined firm competitiveness, led to power
outages of 10–12 hours a day by 2012–13 and added significantly to Pakistan’s
oil import bill. One estimate has that these shortages cost Pakistan 2 per cent
of gross domestic product (GDP) annually (Husain 2017). A report from
the International Monetary Fund (IMF) estimated that energy shortages
had been reduced between 2012–13 and 2015–16, from around 10–12 to
2 hours per day in industrial areas and around 6 hours per day in residential
areas (Shafqat and Shahid 2018). The CPEC has focused on increasing the
supply of energy and while there is evidence of reduced shortages there are
significant institutional-organisational problems with electricity supply that
the CPEC is not scheduled to tackle. These include inaccurate forecasts of
demand, water shortages, volatile fuel prices, persistently high transmission
and distribution losses, and a lack of political commitment by the government
to deal with these issues (Siddiqui et al. 2011). The electricity supply system
is strangled by an enduring and pervasive problem of circular debt. There are
massive payment arrears at every stage of the generation, transmission and
distribution system. This has ‘jammed the flow of funds through the power
supply chain and deprived fuel suppliers and independent power producers
5
of revenue’ (Alahdad 2011: 231). Consumer tariffs are insufficient to pay for
the operation and functioning of the electricity system. This in turn increases
arrears, and suppliers of fuels and independent power producers go unpaid.
This has undermined the viability of firms at every stage and their ability to
continue functioning in the market. Payment arrears were estimated at almost
$5 billion by 2011 (Alahdad 2011).
There is grumbling about the CPEC but this appears to be very much
a minority exercise. The ‘CPEC Opponents’ noted by Shafqat and Shahid
(2018) include those who argue that the CPEC will lead to an exclusive
focus on exploiting Pakistan’s natural resources and to an inevitable cultural,
social and political subjugation of Pakistan to China. A common theme
here is that Pakistan will struggle to repay CPEC loans and become locked
into a new cycle of debt dependency. In the extreme, the CPEC has been
caricatured as a ‘new East India Company’. Scholars and government officials
from India and the United States (US) have been consistently more sceptical
about the CPEC than their counterparts in Pakistan and China (Livemint
2019). ‘CPEC Opponents’ have been drawn from among leftist activists
and regional politicians, particularly those with a base in Balochistan or
Gilgit-Baltistan. One emotive summary of the CPEC is that it represents
the ‘imperialist aims of China and Pakistan to usurp Gilgit Baltistan under
the garb of the China Pakistan Economic Corridor’ (Choudhury 2017: 1).
Shafqat and Shahid (2018: 19) label themselves as ‘CPEC Reformers’ and
argue that the CPEC is credible and has been designed by China in a way that
can contribute to economic, human and social development in Pakistan. The
‘Reform’ label comes from their view that reforms are necessary, particularly to
ensure the greater involvement of local communities in the process of project
development, which they believe will help the CPEC maximise its potential
benefits. Reformers are noted by their frequent calls for more transparency in
the data regarding CPEC investments and related project funding.
This book for the first time has utilised the large and rigorous economic
literature on the impact of big infrastructure projects on economic outcomes to
think about the likely economic impact of the CPEC. It also pays much more
attention to China. Economic studies of the CPEC do devote much space
to discussing Chinese policy. How much money, what terms and conditions,
what sectors are China planning to invest in and other such questions. What
these studies forget about is what is happening across the border in Western
China. It makes a big difference, for example, if industrial enterprises are
6
INTRODUCTION
opening in Western China and are readying themselves to export high-quality

and low-cost production to Pakistan. What hope then for Pakistani industry?
This book looks in some detail at economic development in the western
Chinese province of Xinjiang and thinks carefully about the implications for
Pakistan.
CPEC CONTEXT II: THE CULMINATION OF

A LONG FRIENDSHIP?
The second story about the CPEC is that of a long-established history of
economic links between China and Pakistan. We can then be optimistic
because the CPEC is the culmination of a long history of successful and
peaceful cooperation. Pakistan was among the very first non-communist
countries to recognise communist China in 1950, this when China and the
US were on the verge of direct conflict in the Korean War. The 1950s saw
close military and political links between Pakistan and the US which delayed
the deepening of Pakistan’s relations with China. In January 1963 a long-term
trade agreement was signed which established bilateral trade and commercial
ties. In November 2006 China and Pakistan signed a comprehensive free-
trade agreement (FTA) (Sial 2014). These trade policy agreements have
consistently been supported by an emphasis on improving infrastructure.
Again, other than the scale, the CPEC is not particularly new for Pakistan.
The Chinese constructed the 1,300-kilometre Karakoram Highway in the
1960s to connect Hasan Abdal in the Pakistan Punjab to the Khunjerab Pass
in Gilgit-Baltistan where it crosses into China. The CPEC followed the same
pattern. The FTA in 2006 was accompanied by discussions to construct trade-
supporting infrastructure. The construction of the CPEC infrastructure links
from Kashgar in Western China to the deep-sea port of Gwadar in southern
Pakistan was discussed during the visit to Pakistan of Chinese Premier Li
Keqiang in 2013. The numbers were added together in April 2015 when
President Xi visited Pakistan and signed 51 agreements worth $46 billion.
This was the public launch of the CPEC vision.
The origins of the CPEC run from Chinese-led efforts in the planning of
the CPEC and then rapid and widespread acceptance of the plans in Pakistan.
The detailed initial planning for the CPEC came entirely from China.
The long-term plan began in November 2013 when the National Development
and Reform Commission (NDRC), the central planning organisation of
7
China, asked the China Development Bank (CDB) to compile a detailed

plan to guide China’s economic engagement with Pakistan up to the year
2030. The CDB also called on assistance from the ministries of Transport,
the National Energy Administration and China Tourism Planning Institute.
Details of this process came from the Pakistani media. The report was first
‘transmitted to the Government of Pakistan in 2015’. Pakistan’s immediate
response was inactivity and the report ‘gathered dust for a few months’.
Only ‘[u]nder prodding from the Chinese government, a team from Pakistan
met their Chinese counterparts in Beijing on November 12th 2015’, and the
plan was finalised by December of that same year (Dawn 2017a). Throughout
these initial years there was no wide input or discussion from academia, think
tanks or civil society in Pakistan. The long-term CPEC plan had to be leaked
to the media to come to public attention. The newspaper Dawn declared in
June 2017 that it had ‘acquired exclusive access to the original document, and
for the first time its details are being publicly disclosed here’ and further that
‘[t]he plan lays out in detail what Chinese intentions and priorities are in
Pakistan for the next decade and a half, details that have not been discussed in
public thus far’ (Dawn 2017a).
China and Pakistan both maintain a perception of the other as being a
reliable and enduring friend. China’s partnership with other countries—both
the geopolitically dominant US and the former Union of Soviet Socialist
Republics (USSR), and the smaller Albania, Vietnam, Algeria and North
Korea—have been close but have since cooled; with Pakistan the friendship
and practical cooperation has been remarkably resilient ( Jan and Granger
2016). Despite the absence of cultural similarities and common values between
the Islamic and communist nations, this alliance has remained close (Small
2015). Some have seen the explicit long-term commitment of China to the
CPEC as a refreshing alternative to the recurring short-term stabilisation and
adjustment packages offered by the IMF and to the alternating tendency of
the US to embrace and reject Pakistan in accordance with fluctuating US
foreign policy interests (McCartney 2011a).
Some scholars have argued that Chinese friendship and the CPEC
is a smile that hides the real motivations of China, which they suggest are
about international relations and geopolitics and not economic development.
In international relations much of this debate is conducted between ‘Liberal’
and ‘Realist’ theorists. The Liberals argue that Beijing is seeking to create a
positive-sum game in which China and its South Asian neighbours mutually
8
INTRODUCTION
benefit through better transport links, leading to more trade, investment

and production (Chen 2014). The Realists argue that Beijing is utilising the
CPEC and the BRI to strengthen regional dominance, challenge the current
US-led global–regional order and create opportunities to construct military–
naval bases throughout the Indian Ocean (D. P. Nicolas 2015). In the most
extreme version of this argument, some scholars argue that the ultimate aim
of Chinese foreign policy is to ‘push the United States out of the Asia-Pacific
region’ (Mearsheimer 2010: 389). It is true that Sino-Pakistan relations
have closely involved the military. China supplied much of the expertise for
Pakistan’s nuclear and ballistic missile programme in the 1980s and 1990s
(Chaziza 2016). The dependence runs both ways. Pakistan is now China’s
biggest overseas arms buyer, accounting for more than 40 per cent of Chinese
arms exports. These exports included in 2015 the $5 billion purchase of
eight submarines which was then the biggest defence deal in China’s history.
The CPEC may also help provide China with a reliable long-term naval base
at Gwadar, strategically located in the Indian Ocean close to the Persian Gulf
and so transforming China into a two-ocean naval power (Chaziza 2016).
A Chinese presence in the Indian Ocean would provide a hedge against
Indian ambitions to do the same (Garlick 2018). For some scholars Gwadar
represents one among a ‘String of Pearls’ or strategic naval bases across the
Indian Ocean, including Chittagong (Bangladesh), Hambantota (Sri Lanka)
and Woody Islands (Paracel Islands), among others. Other research has noted
that almost none of these ports has even the minimal facilities necessary to
support combat (D. P. Nicolas 2015: 35). Boni (2016) examines the motivations
of the civilian and military elite in Pakistan, arguing that they are seeking to
create a secure environment for Chinese investment in order to benefit from
Chinese military and diplomatic support for Pakistan’s wider geo-strategic
interests.
China clearly does hope to leverage its close relationship with Pakistan
as a gateway to influence other Islamic countries in Central Asia and the
Middle East. This has helped China supress separatist groups such as the East
Turkestan Islamic Movement formed among Uighur militants in Xinjiang.
This group had links outside China and so combating them requires inter-
regional cooperation. While Uighur separatists have received training and
support inside Pakistan, the Government of Pakistan has also suppressed
these militants under pressure from Beijing ( Jan and Granger 2016).
The crackdown on militants holed up in the Islamabad Red Mosque in 2007
9
is one such example. Pakistan also supported China and used its influence in
the Islamic world to head off any criticism of China’s handling of the 2009
Uighur ethnic riots in Xinjiang that left 200 people dead and 1,600 injured
( Jan and Granger 2016: 292). Pakistan is also supporting China in helping to
ensure there is no spillover of radical Islam from Afghanistan into Xinjiang,
particularly since the US decision to pull troops out of the region (Chaziza
2016). Trade and transport links with Pakistan are also seen through the prism
of China’s own problems with terrorism and unrest in Western China (Small
2015: ch. 4). Wolf (2019) explores the (potentially negative) impact of the
CPEC on religious extremism in both Pakistan and China.
CPEC CONTEXT III: ECONOMIC DEVELOPMENT

IN PAKISTAN SINCE 1947
A third story of the CPEC is looking back at Pakistan’s own economic history.
Here too often in the CPEC literature, the Pakistan economy is presented
as either a disaster that needs to be rescued by the Chinese and the CPEC
or else as an economy of wonderful potential that will be liberated by the
opportunities presented by the CPEC. We need to better understand the actual
numbers for Pakistan and think about them in relation to other developing
countries. Contrary to both these imagined Pakistan perspectives emanating
from the CPEC literature, the characteristic feature of Pakistan’s economy
since independence has been that of ‘resilience’, not ‘failure’.
Despite recurrent weather shocks (droughts and floods), intermittent surges
in the global oil price, policy shocks, civil wars and international wars, Pakistan
has experienced an annual average of 5 per cent GDP growth from the 1950s
to 2015 (McCartney 2011b). While not quite up to the rates experienced
by South Korea, Japan or Taiwan in the 1960s, Malaysia or Thailand in the
1980s or China or Vietnam in the 1990s, this is almost exactly the same long-
term growth story as that experienced by the now-lauded India. Pakistan has
avoided long decades of negative economic growth as experienced by much of
sub-Saharan Africa, or a lost decade of economic stagnation as much of South
America did in the 1980s. In the period after 1990 (as shown by Table 1.1)
Pakistan’s economic growth, even though slowing down slightly from its own
long-term average, is comparable to other successful Asian growth stories
such as Indonesia, Thailand, Hong Kong and South Korea.
10
INTRODUCTION
Table 1.1 GDP growth, 1990–2011
Country GDP growth, 1990–2011 (%)

Pakistan 4.3
China 9.9
India 7.2
Indonesia 5.2
Thailand 4.1
Hong Kong 4.0
South Korea 4.0
Source: Asian Productivity Organisation (2013: 18).
Pakistan’s performance becomes even more impressive when we compare

to a large cross-section of other countries as opposed to this small sample.
The most impressive aspect of Pakistan’s growth performance was its shift to
positive and sustained economic growth after independence. In the 50 years
leading up to independence in 1947 (what became) Pakistan experienced an
average growth of 0–0.5 per cent per annum. Between 1950 and 1960 growth
accelerated to 3.2 per cent per annum and then further between 1960 and 1992
to 6 per cent per annum (McCartney 2011a). Hausmann, Pritchett and Rodrik
(2004: 1) defined a ‘growth episode’ as a sustained 2 per cent improvement in
the growth rate. They found that short episodes of booms and busts were
common but that very few developing countries were able to sustain such an
acceleration over several decades. The jump in the growth rate and sustaining
that acceleration was an impressive and unusual economic performance (Kite
and McCartney 2017).
Not only was this a commendable average but it was also a remarkably
stable economic growth. GDP growth in Pakistan has fluctuated much less
than the other large countries of South Asia. Figure 1.1 shows that Pakistan
has not had a recession (a year of negative GDP growth) since at least 1960.
This has been true of Bangladesh (1972 and 1975), India (1965 and 1979) and
Sri Lanka (1972 and 2001).
To convert this into a more rigorous measure of stability, we can use the
coefficient of variation and standard deviation. These measures show how
much this growth has varied around the 5 per cent average. A higher value
of either measure represents greater variation from the average data value.
11
Figure 1.1 GDP growth trends, 1960–2016

Source: Afzal, McCartney and Asif (forthcoming).
Table 1.2 confirms the visual impression from Figure 1.1 and demonstrates
that growth stability is greater in Pakistan than in India and Bangladesh and
similar to that in Sri Lanka.
Pakistan has also managed to diversify its economy. At the time of
independence 75 per cent of Pakistan’s GDP and 99 per cent of her exports
were dependent on agriculture and raw materials. By the mid-1990s the
share of industry (at 25 per cent) had overtaken the share of agriculture
in GDP and 80 per cent of Pakistan’s exports were by then manufactured
goods (Burki 1999). In 1947 the vast bulk of Pakistan’s exports went to India.
By 2019 Pakistan exported to almost 200 countries, with the largest shares
going to China, the US, Afghanistan, Germany and the United Kingdom
(UK) (World Bank 2019b). There are, of course, problems and many scholars
question whether this long-term growth performance can be sustained into the
future. It is evident from Figure 1.1 that Pakistan’s growth has slowed down
considerably since 1990. Since 1990 and especially since 2006 Pakistan has
been the worst performing major South Asian economy, though, as noted in
Table 1.1, growth was still positive and comparable to other Asian economies.
Chapter 6 discusses in more detail some of these threats to sustainable growth,
including the failure to diversify and upgrade exports away from simple textile
products into more technologically demanding sectors, and Chapter 7 the
continuing challenges for Pakistan in relation to education, governance and
access to credit.
12
INTRODUCTION
Table 1.2 GDP growth stability analysis
GDP growth
India Sri Lanka Pakistan Bangladesh
Coefficient of variation 0.581642 0.474825 0.462566 0.899518
Standard deviation 3.10924 2.01546 2.394393 3.753301
Source: Afzal, McCartney and Asif (forthcoming).
CPEC CONTEXT IV: THE CHINA STORY SINCE 1976

The fourth story about the CPEC begins in China after the death of Mao in
1976. By 1979 Deng Xiaoping had risen to the summit of Chinese politics.
Once located there, Deng sought to open up China to the outside world,
motivated by a desire to acquire advanced foreign technology. As part of that
effort, he established a number of SEZs where foreign firms could invest.
Foreign enterprise was seen as a risk. Would the taste of capitalism and foreign
culture generate any clamour for wider political freedoms that undermined
the rule of the Chinese Communist Party? Very purposefully the first SEZs
were set up at great distance from the major metropolitan areas of Beijing and
Shanghai. Any emerging problems could be safely isolated. As is discussed
later in this book, the SEZs were perceived to be an enormous success in
terms of not just acquiring technology but also in promoting exports and rapid
industrialisation. Despite their initial isolation, the SEZs proved to be an
incubator for reform that spread from these enclaves to encompass the whole
of China. A second and related part of the story is that throughout the 1980s
Deng also targeted Chinese state-owned enterprises to be a central focus
of these efforts to open up. State-owned enterprises were pushed to bid on
overseas construction projects to both earn foreign exchange and to form joint
ventures abroad to increase their exposure to foreign technology and overseas
management skills. These reforms launched a transformation of China’s
engagement with the rest of the world, particularly the developing countries,
and marked a radical shift from the export of revolution and ideology that had
characterised China’s foreign policy in the 1960s and 1970s (Lovell 2018).
Twenty years later, Chinese President Jiang Zemin attended the 2000
United Nations (UN) Millennial Summit in New York. The developed world
focused on debates about promising aid targeted at poverty reduction and
13
human development. These promises and aspirations were encapsulated in

the Millennium Development Goals (MDGs), a broad set of targets related to
poverty and human development. China politely participated but looked back at
its own reform success in the 1980s and 1990s for inspiration in looking forward.
China drew on lessons from its own history rather than the then fashionable
trends in contemporary development thinking. One month later in October
2000, China launched and hosted the Forum on China–Africa Cooperation
(FOCAC). The success of this effort was startling and was one of the first
clear markers of the global diplomatic and economic rise of China. More than
40 African countries sent their foreign and economic ministers to participate.
China promised the delegates debt relief, training and investment (Brautigam
2009). This emphasis on production marked a sharp difference with the
UN Summit emphasis on human development. The launch of the FOCAC
marked the culmination both of China’s effort to re-engage with the
global economy and also of a diplomatic effort to explain its new strategy.
Throughout the 1990s, China invested much diplomatic effort in Africa.
In 1995 alone, Chinese vice-premiers visited 18 African countries. In 1996 the
Chinese president visited a further 6 countries (Brautigam 2009). This was
patient diplomacy. For the IMF, negotiating over the conditions that had to
be fulfilled was an integral part of any loan. For China, the discussion and the
vision came first and the lending after. There was no echo here, as some have
argued, of a new nineteenth-century-style ‘Scramble for Africa’. This was an
engagement that was carefully prepared and fully explained long in advance.
The 2000 Beijing Summit was warmly received across much of Africa. China
was no longer just opening up to the world motivated by an effort to acquire
technology. China was propagating its own vision of how the global economy
should work.
China’s motives for its tilt towards Africa were various. China needed raw
materials, particularly oil, from Africa to fuel its own rapid industrialisation.
Since 1993, China had become a net importer of oil. Partly this was about
international politics. A few countries in Africa still recognised Taiwan as
the true China and China wanted to change diplomatic minds (Brautigam
2009). This was also about the nature of industrialisation in China. The 2000s
saw a shift from promoting industrialisation in China to concentrating those
efforts into the creation of a number of national champions. The national
champions were firms that China hoped would become globally competitive
multinationals (MNCs) producing and controlling proprietary technology
14
INTRODUCTION
and brand names. China was heavily influenced by the rise of firms such
as Samsung, Toshiba and Panasonic in Japan and the other miracle Asian
economies from the 1960s onwards. China bought into the narrative about
MNCs controlling the evolution of international capitalism and wanted to
make sure it would participate in that control. In the early 1990s, the Chinese
government launched the policy ‘grasp the large and let go of the small’.
Small firms were privatised or closed and state support was focused on key
large enterprises. A national team of 120 enterprise groups was selected in
two batches in 1991 and 1997 by the State Council. These were concentrated
in strategic sectors such as electricity generation, coal mining, aerospace,
transport and machinery. These champions were given special help in the
form of access to technology, profit retention, rights to engage in international
trade, support from state-run research and development (R&D) laboratories
and access to credit (Nolan 2001). The national champions were intended to
achieve a global presence. In 1995 the Export–Import Bank of China (China
Eximbank) launched a policy of concessional loans to promote joint ventures
overseas in manufacturing and agriculture. In 1998 China Eximbank further
nudged these champions overseas by offering preferential loans to construction
companies to help them win contracts in the Middle East and Africa.
The scale of support in some cases was massive. The NDRC and Eximbank
provided $10 billion in credit to Huawei to expand global operations. China
National Oil Corporation (CNOC) received a loan of $1.6 billion for the
same purpose. As well as finance, the firms received diplomatic support, help
with foreign risk assessments, and easier emigration approvals. The economic
help was also directed to foreign countries. In 2000 China Eximbank began
supplying export seller credits to countries that wished to purchase imports
from China and by 2007 had become the world’s largest export credit
agency. The engagement with Africa was in part an effort to create a suitable
environment to host these national champions and help them restructure.
As part of their efforts to move into more technologically intensive sectors,
the champions were encouraged to relocate low-technology, labour-intensive
industries such as leather and textiles to other countries. Recalling its own
experience of successful reform in the 1980s, China proposed the setting up of
SEZs to host mature industries from China and supported Chinese firms going
global in groups to re-establish industrial networks overseas. The first two
African SEZs were announced for Mauritius and Zambia (Brautigam 2009).
15
This process was infused with Chinese characteristics and the lessons
of practical experience. China did not copy the IMF and lend or give aid
in exchange for policy conditionalities. This process was also heavily
influenced by emerging global norms; China was changing and learning
from its new global engagement. China’s preference for the state-to-state
economic engagement of the 1960s and 1970s was replaced by greater
reliance on markets and the private sector. During the 1990s China tended to
support the African private sector through cooperation with Chinese companies.
The Tanzania–Zambia railway, for example, that had been formally handed
over to African public-sector management in 1976 was revived by Chinese
loans for rehabilitation investment and in 1983 Chinese managers returned
to Africa (Brautigam 2009).
The comparison (and competition) with the US is striking. In 2000 the
US launched the Africa Growth and Opportunity Act (AGOA) to boost
exports from Africa. The effort proved to be an ‘abysmal failure’ (Brautigam
2009) and generated little US investment in African industry and only a minor
boost to African exports. The AGOA contained complicated rules of origin,
whereby the bulk of inputs used by the domestic industry, often textiles (such
as dyes and buttons) as well as raw materials (such as shoe leather), had to be
produced locally in Africa. African countries lacked the necessary broad base
of supplier firms to make this rule work. The AGOA rules kept changing and
were supplemented by various political conditions (Kaplinsky, McCormick and
Morris 2007). There was no government support to assist the relocation of
mature or labour-intensive US firms overseas. In practice these years saw a
political reaction against the idea of US firms relocating. Ross Perot enlivened the
1992 presidential election with the claim that the free trade deal with Mexico
had created ‘a giant sucking sound’ of factories, jobs and wealth leaving the US.
In contrast to China, the US government faced intense domestic pressure from
trade unions who sought extra protection to ensure the continued survival of
domestic production in textiles, shoes and electronics (Brautigam 2009).
The key themes of SEZs, no conditionalities, using overseas engagement
to promote industrial restructuring and creating new opportunities for export
culminated in the announcement in 2013 of the Chinese Belt and Road
Initiative (BRI). The BRI took those earlier themes onto a much larger global
stage. The BRI was planned to be a vast network of road and rail connections,
seaports, energy and manufacturing investment across Eurasia (Asia, Europe,
Central Asia and the Middle East) and Africa (Map 1.2). The BRI was first
16
INTRODUCTION
Map 1.2 The Chinese Belt and Road Initiative (BRI)

Source: Lommes, CC BY-SA 4.0 https://creativecommons.org/licenses/by-sa/4.0, via Wikimedia
Commons.
announced in 2013 during a visit by President Xi Jinping to Kazakhstan.

Since 2013 the BRI has become an integral part of Chinese foreign policy
under President Xi and working towards its success was made a constitutional
obligation in China (Boyce 2017). The BRI comprises six economic corridors
spanning an estimated 80 countries, containing more than two-thirds of the
world’s population and costing between $1 and $8 trillion, depending on
the source of ever more excitable estimates (Dawn 2018). China looks back
and feels it also ‘won’ in the 1980s and 1990s when it was a host to MNC
investment. That investment generated good profits for foreign investors
but left many benefits for China. It is reasonable to posit that the Chinese
government thinks such a mutually beneficial scenario is still feasible now
that Chinese MNCs are taking the lead.
To some, there is nothing particularly new in the BRI. Chohan (2017), for
example, argues that China is doing what all successful global exporters do.
The BRI, he argues, is just the latest manifestation of what he calls a global
‘surplus recycling mechanism’. By the 1970s and 1980s the post-war export-led
growth success stories, Germany and Japan, were running large surpluses on
17
their balance of payments. This pattern was similar to China after 2000, the
US in the 1950s or even the UK in the mid-nineteenth century. There are only
limited outlets to utilise the surplus in the domestic economy. In these earlier
cycles, the UK, US, Germany and Japan recycled their surplus in the form of
foreign investment and lending to deficit countries. In its own turn, China
has now utilised its own surplus to fund massive expansions of, first, domestic
infrastructure (discussed in detail in this book) and, second, the purchase
of US government debt (treasury bills), which, though safe, generate little
economic return. China is now seeking more productive overseas locations to
absorb that surplus. So, while the BRI may be novel in terms of its scale and
geographic reach, it performs a very similar role to these earlier manifestations
of surplus recycling. There is much truth in this story. The launch of the BRI
did coincide with China emerging as a net exporter of capital. In 2015 inward
FDI reached $135.6 billion and outward FDI grew by 18 per cent to $145.7
billion; China became a net exporter for the first time (Zhai 2018: 84).
For others, this excitement stems more from the fevered imagination of the
BRI watchers and their ability to construct for the BRI an ever more dramatic
scope and cost. One reason that the BRI is perceived to be ‘breathtakingly
ambiguous’ is that there is ‘no official definition for what qualifies as a BRI
project’. The ‘BRI brand has been extended to fashion shows, art exhibits,
marathons, domestic flights, dentistry and other unrelated activities’. This
allows lots of scope for creative accounting when estimating the total cost
or size of the BRI (Hillman 2018). It is also easy to over-exaggerate the
coordination and coherence of the BRI by relying on official publicity briefings
and planning documents. Analysing 173 Chinese-funded infrastructure
projects initiated between 2013 and 2017 across 45 countries, Hillman finds
that ‘there is no significant relationship between corridor participation and
project activity’ (2018: 5). This indicates that various interest groups in China
and elsewhere are repackaging their existing work and giving it a BRI label to
qualify for the incentives and the political benefits of being associated with the
initiative. This is less a problem for the particular case study of this book: the
BRI as it runs through Pakistan. This is the only economic corridor of the BRI
that Hillman (2018) found is significantly correlated with levels of project
activity. This could be explained by the fact that it is the only corridor where
China deals with a single negotiating partner rather than a host of different
governments.
18
INTRODUCTION
CONTEXT PART V: A LARGER GLOBAL HISTORY

OF INFRASTRUCTURE
One thing entirely lacking from current writing on the CPEC is any historical
perspective beyond wistful recollections about successful Sino-Pakistan
cooperation in the construction of the Karakoram Highway in the 1960s.
History can tell us a lot about the long-run impact of big infrastructure projects
on growth and development. There are two historical stories of big infrastructure,
the Suez Canal and the Panama Canal, that had at best only ambiguous impacts
on long-run economic development in Egypt and Panama respectively.
The 195-kilometre Suez Canal was opened in 1869 to link Port Said
in the north of Egypt to Port Tawfiq/Suez in the south. The canal quickly
became the most important waterway linking the Red Sea and Indian Ocean
to the Atlantic and the Mediterranean. It offered the global trading economy
a huge saving in terms of distance. The route from Jeddah (Saudi Arabia) to
Constanza (Black Sea) was reduced by 86 per cent compared to the long route
around the Cape of Good Hope at the southern tip of Africa (Mostafa 2004).
The journey of 11,560 nautical miles from Bombay to Liverpool was likewise
reduced to 5,777 nautical miles (Fletcher 1958). The canal had a dramatic
impact on patterns of trade. By 1871 96 per cent of all plain cotton goods
imported into Bombay were routed through the canal (Fletcher 1958). As this
book will later show, reductions in the cost of trade between Britain and India
led to the dramatic influx of cheap cotton textiles from Britain to India that
many scholars have argued led to the de-industrialisation and impoverishment
of India. The canal remained crucial for global trade, even a century after its
construction. The canal was closed for almost a decade from 5 June 1967 (the
beginning of the Six-Day War in the Middle East) and it was only reopened
on 5 June 1975. The closure had a significant and robust impact on global
trading patterns, proving the continued global importance of the canal (Feyrer
2009). The canal had a profound influence on the pattern of nineteenth-
century technological change by making it possible for steamships to trade
between Europe and Asia; steamships previously had been unable to hold
enough coal to travel around the Cape. The shift to steam accelerated a process
of technological innovations, including the screw propeller that replaced the
paddle wheel, the shift from wood to iron plating, and the development of
the compound engine which replaced the single-cylinder engine. The relevant
question for contemporary Pakistan is the extent to which the Suez Canal
stimulated economic growth and development in Egypt. The answer is
19
that, despite these profound international influences, it had little enduring

impact on Egypt. The global technological shift from sail to steam shipping
that occurred at the same time as the construction of the canal generated
few spillovers for Egyptian industry. The shift benefited the technological
leader and so gave a huge boost to the UK iron and shipbuilding industry.
The new technologies associated with steam shipping were developed by British
firms. The new steam ships were built in British yards. By 1880 80 per cent
of shipping passing through the canal was British, and by 1910 62 per cent of
shipping was still British (Fletcher 1958). If we flash forward to contemporary
Egypt, we find that in 2016 Egypt was stuck in negotiations with the IMF
(like contemporary Pakistan). Egypt was then exporting around $25 billion
per annum, comprised of chemicals, fertiliser, food industries and readymade
garments (much like Pakistan). Its imports at $59 billion ran way ahead and
left a deficit of around $35 billion (echoes of Pakistan). This deficit was covered
by more than $25 billion of remittance income (similar to Pakistan). Tourism
receipts declined from $12 billion in 2012 to $3.5 billion in 2016. In 2016
Egypt received $6 billion of receipts from shipping charged to use the canal
(Kenaway 2016; Daily News 2019). In 2018 the GDP per capita in Egypt was
around $2,400 and the country was classified as lower middle income (as was
Pakistan). The canal had not stimulated any particularly favourable dynamics
in terms of democratisation and the growth of civil society. The massive debt
accumulated in the construction of the canal led to Egyptian bankruptcy in
the later nineteenth century and its eventual colonisation by Britain (recall
those claims by ‘CPEC Opponents’). Much like Pakistan, Egypt has endured
three long military-backed dictatorships from the 1950s to the 2000s, Nasser
(1956 to 1970), Sadat (1970 to 1981) and Mubarak (1981 to 2011). Just as
now seems to be the case with the CPEC in Pakistan, the Suez Canal has
long provided the military with a justification to insert itself into discussions
of long-term economic planning (S. Marshall 2015: 12). The greatest piece
of engineering in the nineteenth century did not stimulate long-term growth
and the industrialisation of the Egyptian economy. In 2019, on the 150th
birthday of the canal, the Egyptian economy was remarkably similar in its
nature and its basic problems to that of Pakistan. Egypt was still talking about
transforming the canal-zone into a ‘major logistics hub and centre of heavy
manufacturing’. In 2019 Egypt was drawing on canal transit revenues as only
its third largest source of capital inflows to help fund a massive balance of
payments deficit. The biggest difference with Pakistan was (very unstable)
20
INTRODUCTION
tourism revenues. It was not big infrastructure that kept Egypt afloat; it was
the pyramids. Forget the Suez Canal; Egypt was just Pakistan with pyramids.
The construction of the Panama Canal between 1903 and 1914 was funded
by the US and then represented the most expensive infrastructure project in
US history. Its opening was delayed until 1920 by which time the direct cost
had soared to $326 million ($6 billion in 2006 dollars) which was 2.3 times the
initial cost estimate. By 1925 an extra $158 million had been spent on facilities
to ‘host’ the US military who were providing ‘security’ for the canal (bringing
the total to $9.1 billion in 2006 dollars). This investment was equivalent to
1.8 per cent of US GDP spread over nearly 20 years. As with the Suez Canal,
there was a dramatic impact on global distances. Bulk cargoes such as petroleum
from southern California and lumber from the Pacific Northwest no longer had
to be routed around the southern tip of South America. The distance from the
US west coast to Europe was reduced from 13,841 to 7,825 miles and proved an
evident boon for US shipping. By the 1920s and 1930s, 75 per cent of the cargo
transiting the canal was US in origin (Maurer and Yu 2008: 691). It seems clear
that the US managed to retain the bulk of the benefits for itself. Maurer and
Yu (2008: 688) estimate social returns of between 6 and 12 per cent to the US
from the canal. These were much better returns than were offered by alternative
investment opportunities in early-twentieth-century America. These benefits
were inflated by US leverage used in negotiating the original treaty and then
subsequently operating the canal. Panama signed the canal-zone (including
6 miles around the canal) over to the US on a 99-year lease. A treaty agreement
prevented Panama from imposing any tax on the canal and limited the US to an
annual rental payment of $250,000 to the government and an upfront payment
of $10 million. The US dispatched 2,000 marines to help ‘encourage’ Panama to
sign the agreement (Maurer and Yu 2008: 710). Looking back after more than
a century, Panama does seem in a better place than Egypt. GDP per capita in
Panama has reached almost $12,000, marking it as almost a high-income country.
The Colon Free Zone adjacent to the canal was established in 1948 and has
since become one of the most important distribution centres for Central, South
and North America and the Caribbean for wholesale food, textiles, household
domestic goods, construction materials, machinery and motor vehicles.
The zone also provides docking facilities for cruise ships, an international
airport and a railroad terminal. Eighty-seven per cent of Panamanian GDP
now originates from the zone. Expansion plans for the canal to allow its use by
larger ships are predicted to generate large increases in GDP for Panama up to
21
2025 (Pagano et al. 2012; Pagano et al. 2016). On the surface the Panama Canal
experience—excepting the rather forceful imposition of the original treaty—
does seem like a good model for Pakistan. The Panama Canal did generate huge
gains for the US, but Panama did not have to incur large debt in its construction
and over time clearly enjoyed its own benefits. The principal problem with
the canal though is that it has only benefited a small fraction of Panama’s
3.4 million population. The canal created less than 10,000 jobs in 2010 and
Panama was by then the second most unequal country by income distribution in
Latin America. Sustainable economic growth maybe but a model for sustainable
social development in Pakistan as a country of 200 million people it certainly
is not.
A sense of history may give us some perspective. History should warn
Pakistan of the dangers in allowing military-relevant infrastructure to be
funded by a neighbouring Great Power. History also highlights the ambiguous
potential of big infrastructure in stimulating long-run sustainable economic
development. The stories of the Suez Canal and the Panama Canal at most
only offer suggestive narratives. There is rigour elsewhere that CPEC scholars
have hitherto failed to learn from.
There is another crucial body of evidence we can use to help our thinking
about the CPEC. There is a large number of very detailed and rigorous recent
economic studies of big infrastructure projects,projects dating from the nineteenth
century to the contemporary world. Almost none of these have yet been drawn
upon to enlighten discussions of the CPEC. That is something this book will
undertake for the first time. This will help take thinking about the CPEC away
from rhetorical flourish and into a more rigorous debate rooted in economic
theory and economic evidence. Most of the existing CPEC studies lack ‘critical
appraisal of the projects or any sources that challenge their robustness’ (Shafqat
and Shahid 2018). The Government of Pakistan, for example, has estimated
that 800,000 jobs will be created in the SEZs even though in most cases the
land for these projects has not yet even been acquired (Daily Times 2019). This is
a politics of hope rather than the outcome of rigorous analysis. We need to think
carefully about how we draw on these studies of big infrastructure. Most of these
studies have been looking backwards to analyse the economic impact of finished
projects. The CPEC is not even due for completion until 2030. We need to
remember that Pakistan has a dire record of implementing and benefiting from
large-scale donor-funded projects, the catastrophic Social Action Programme
(SAP) in the 1990s being just one such example (Birdsall and Kinder 2010).
22
INTRODUCTION
THE MAIN FINDINGS OF THIS BOOK
Big Infrastructure: Big Problems or Big Benefits?

Chapter 2 notes that some good news for Pakistan, emanating out of the existing
literature studying the impact of big infrastructure projects, is the consistent
and positive story of the impact such projects have made on employment,
investment, growth and productivity. Any positive story though is only
conditional, dependent on whether new transport infrastructure represents a
transformational change to the existing transport system, the opportunity cost
of infrastructure investment, and the motivations for building infrastructure.
Will the CPEC create isolated pockets of China inside Pakistan or generate
spillovers that benefit the wider Pakistani economy?
CPEC Spillovers Rippling Outwards

Chapter 3 introduces the theoretical concept of the ‘leading sector’ and asks
whether the expansion of infrastructure can generate sufficient spillovers to
other sectors and thereby boost wider economic growth. Scholars of the CPEC
have made much of the dramatic reduction in distance for firms in China
to export goods to Europe; they herald the fact that the CPEC will make
markets more efficient. But efficient to do what? Will the CPEC evolve into
a transit corridor that allows Pakistanis to watch Chinese exports whizzing
through Pakistan to access markets in the rest of the world? Will the CPEC
lead to a surge in exports from China that undermines Pakistani industry?
Will the CPEC stimulate economic growth within Pakistan? Historical case
studies tell us that big infrastructure has induced all three outcomes.
Through the Eyes of Who? Evaluating the Success of the CPEC

Chapter 4 notes that in the existing CPEC literature there is little sign of
careful thinking about how to measure the success or otherwise of the CPEC
using any rigorous method. More common is to use a list of descriptive
statistics about CPEC plans and combine them with a mix of laudatory
claims about the transformative potential of the CPEC. The official rhetoric
about the CPEC from the Government of Pakistan is that it will transform
Pakistan’s economy. How do we evaluate the success of the CPEC, a project
23
that is confusing in its coverage across space, time and sector? This chapter
has chosen to do so against the Government of Pakistan’s own claims, that
the CPEC will have a complementary relationship with China and generate
gains for both countries whilst also transforming Pakistan’s economy. These
claims by the government have ignored both the likely impact of the economic
rise of Western China and also the fact that vast historical and contemporary
case study evidence shows that big infrastructure always creates both winners
and losers.
The Dragon Uncoils: Special Economic Zones (SEZs) from

Shenzhen to Africa
Chapter 5 introduces the major headline of the CPEC, the nine SEZs that are
planned to be established across Pakistan. The SEZs have become a real symbol
of the national nature of the CPEC, with one being promised for every province
of Pakistan. This is real concrete evidence, claim CPEC supporters, that the
benefits from the CPEC will spread out to encompass all of Pakistan.
The problem with these claims is that they are based on a corner of the CPEC
that as yet only exists in the dreams of economic planners and politicians.
Much of what passes for analysis of the likely impact of the SEZs is simply
the economics of aspirational hope, not rigorous thinking rooted in evidence
and theory. This chapter outlines the theoretical and empirical benefits of
SEZs as experienced in other countries and in non-CPEC SEZs in Pakistan.
Then it looks at the CPEC promise for Pakistan and the lessons we can draw
from the experience of SEZs in China and from Chinese-invested SEZs in
Africa. Why will SEZ-FDI flow to Pakistani SEZs rather than the thousands
that exist in countries with better governance? Why will the CPEC-SEZs be
a success when so many similar efforts have failed in Pakistan’s own past?
The Dragon’s Embrace: Pakistan–China Trade Policy

While the CPEC is about infrastructure and FDI, these promises are
closely influenced by overall trade relations between China and Pakistan.
Infrastructure links that open up trade between China and Pakistan will make
little difference if the two countries charge high tariffs on trade. Chapter 6
outlines the emerging patterns of trade between Pakistan and China,
24
INTRODUCTION
particularly in relation to the signing of the 2006 FTA. The FTA has helped
China boost exports of manufactured goods to Pakistan, but Pakistani exports
to China, which comprise mainly cotton yarn, have stagnated.
The Will of the Dragon: The Importance of an Industrial Policy

A review of historical and contemporary case studies in Chapter 7 reveals
some important lessons for Pakistan. Infrastructure is more likely to generate
positive spillovers into the domestic economy and SEZs to be more successful
in generating local industrialisation when the government combines them with
an industrial policy. Pakistan has all the characteristic features of an economy
suffering market failures that could justify the use of an industrial policy,
including low productivity, slow structural change and slow technological
upgrading. But does Pakistan have the necessary state capacity to successfully
utilise an industrial policy?
Conclusion: The Way of the Dragon or the Way of the Falcon?

The book concludes by arguing that there is little reason to think that the CPEC
will turn Pakistan into a ‘Falcon Economy’. At the same time, there is not
enough evidence to justify a conclusion that the CPEC is a Chinese-dictated
agenda that aims to turn Pakistan into a debt-dependent and subservient ally.
Although the CPEC infrastructure is not likely to transform Pakistan, there
is a huge amount of global evidence to demonstrate that big infrastructure
does have positive impacts on investment, productivity and growth over time.
The long-term relationship with China, a country with a long history of
successful developmental interventions, is likely to reorient Pakistan towards
a more consistent and long-term thinking in its economic policymaking.
Dannenburg, Kim and Schiller (2013) argue that Chinese FDI, with strong
Chinese-government backing in effect involves a ‘re-territorialisation of the
Chinese state abroad’. Governance is unlikely to be transformed but is likely
to be improved by this close exposure to a teacher with demonstrated patience.
With some irony it is often the supporters of the CPEC—those who label
the CPEC a game changer—that are doing the greatest disservice to Pakistan.
One excitable conclusion was that the CPEC is ‘a collection of infrastructure
projects in Pakistan to develop Pakistan’s shattered economy’ (Chen, Joseph
25
and Tariq 2018: 62). These views are a disservice because of the vigorous
enthusiasm with which they run down 70 years’ worth of moderate economic
success in Pakistan. In 2021 Pakistan is too big and too industrial for $60 billion
of investment over 15 years to make a transformative difference. Pakistan has a
GDP of $320 billion and is a middle-income country. Telling a more positive
story about Pakistan should not distract policymakers. There are real worries.
Pakistan faces challenges relating to sustaining growth over the next decades,
including the failure to diversify and upgrade exports away from simple textile
products into more technologically demanding sectors and problems relating
to education, governance and access to credit. The game-changer optimists
could unwittingly prove a distraction in making policymakers think that the
CPEC will provide a solution to all of Pakistan’s woes.
26
2
Big Infrastructure
Big Problems or Big Benefits?
The existing work on the CPEC tends to be structured around best guesses or
aspirational hopes about the likely impact of the big transport infrastructure
projects that are being planned and constructed. So far there is a complete
absence of academic engagement with the voluminous existing academic
literature of big infrastructure and the potential relevance of these results
for the CPEC. Existing studies of big infrastructure offer useful results and
research methodologies with which to think about analysing the impact of
the CPEC. These include modest studies looking at the impact of a single
piece of infrastructure, studies that aggregate numerous infrastructure projects
to measure their total impact (or ‘social saving’), studies that focus on the
macroeconomic impact of infrastructure projects and more recent studies that
have used hugely impressive big data sets to study infrastructure.
MODEST STUDIES OF SINGLE

INFRASTRUCTURE PROJECTS
The most widely used form of appraisal for a single road, rail or energy
infrastructure project is that of a cost–benefit analysis (CBA). A CBA attempts
to forecast the economic benefits to households and firms of infrastructure
projects by calculating changes in travel costs, including time savings, reduced
operating costs and improved safety. These direct transport impacts are applied
both to existing users and to potential new users. New users are those travellers
that shift to the now cheaper, faster or more reliable infrastructure by taking
new routes or longer journeys. The flows of benefits over time are quantified
in monetary terms, discounted to place a current value on the stream of future
benefits and then compared to the capital and operating cost of the project.
If the discounted benefits exceed the discounted costs, the project is profitable.
27
There are well-known problems with CBA. These include how to place a
monetary value on people’s time and how to quantify safety improvements or
reductions in the wear and tear on vehicles (Holl 2006). Conducting a CBA
is standard practice when planning and financing new infrastructure. The
academic literature on big infrastructure has used backward-looking CBA, not
forecasting before, but evaluating after, the construction of big infrastructure.
There are many such examples.
Gunasekara, Anderson and Lakshmanan (2008) examine the impact of
the 1987 rehabilitation of two roads connecting Colombo in central Sri Lanka
and the central city of Kandy with the north of the island. The improved roads
contributed to an increase in daily traffic from 2,000 vehicles per day in 1985
to 8,000 in 2001. The study uses (before and after) firm-level surveys from
1990 and 2000. The results show an increase in average firm size, especially
among those firms closest to the highway, driven by higher investment and a
resulting shift to more capital-intensive methods of production. By contrast,
more labour-intensive firms, such as textiles, tended to relocate away from
the highway to access cheaper land. There was an increase of 20 per cent
in incomes of those households closest to the highway relative to those
households located at a further distance.
Roads are the most important form of transport in contemporary India.
Road transport accounts for 65 per cent of freight movement and 80 per cent
of passenger traffic. India’s national highway system constitutes less than
2 per cent of this road network but carries more than 40 per cent of the total
traffic. At end of the 1990s, India’s highway network was in poor condition.
The major economic centres were not all linked by good roads and only 4 per
cent of roads had four lanes. Twenty-five per cent of roads were categorised as
congested, which was due to a combination of poor road conditions, increased
demand from growing traffic and crowded urban crossings. Frequent stops
were required to pay taxes when crossing state borders or to permit government
cargo inspection, which also increased costs, congestion and delays. The
surface condition of more than 25 per cent of national highways were labelled
as being in ‘poor condition’, meaning ‘riddled with potholes’. In 2000 it would
take four—five days to drive the 1,500 kilometres between Delhi and Kolkata,
five times longer than a journey of equivalent distance in the US. The National
Highways Development Project (NHDP) was launched in 2001 to expand
the highway network to four lanes and build 2,500 kilometres of new six-lane
expressways. The quality of the highway system was to be improved through
28
BIG INFRASTRUCTURE
the addition of bridges, bypasses and underpasses. The so-called Golden

Quadrilateral (GQ) portion of this was to connect the four major urban areas
of India: Delhi, Mumbai, Chennai and Kolkata. The GQ was 98 per cent
complete by the end of 2010 (Datta 2012). Ghani, Goswani and Kerr (2014)
found that the economic output in those districts close (0–10 kilometres) to the
GQ grew by almost 50 per cent in the decade after construction began.
These growth effects were absent in those districts that were 10–50 kilometres
from the GQ network. Michaels (2008) found that the construction of
highways in the US between 1956 and 1975 (40,000 miles) assisted the trucking
industry to grow rapidly and become the primary mode of transporting goods.
This road construction also increased retail sales, by 7–10 per cent per capita,
in those rural counties that were crossed by the roads relative to other rural
counties. The impact of the roads interacted with the characteristics of the
labour force. Those rural counties crossed by new roads that had an abundant
supply of skilled labour (as measured by the fraction of high school graduates
in the 25+ age group) saw a rise in the relative wages of skilled labour.
When the roads crossed counties with a less educated labour force, the relative
wages of unskilled workers declined. The roads were helping firms relocate to
areas offering a good supply of skilled labour. Using Russian firm-level data
between 1985 and 2003, Brown et al. (2008) found that transport infrastructure
reduced the costs imposed by geographic distance and boosted firm-level
productivity. Holl (2016) studied the impact of highway construction in Spain
between 1997 and 2007 using a sample of 74,500 manufacturing firms. She
found that highway construction gave a boost to firm productivity, the impact
of which varied between industries. Traditional manufacturing industries,
which tend to have a higher weight-to-value ratio, and so were more likely to
need roads for transport, experienced a more significant positive productivity
impact.
By the mid-1990s China’s long-distance train, truck and shipping networks
were so crowded that traders were facing ever-rising costs to book cargo space
in railways or trucks. One estimate suggests that this congestion more than
doubled the freight and handling charges faced by firms. Li and Chen (2013)
studied the new railway track that was constructed in 1994 alongside the
1,200-kilometre railroad (built in the 1960s) to connect Urumqi and Lanzhou
in northwest China. This doubled the capacity of the railroad. The results
show that the most conservative estimate of this return was 10 per cent and
could be as high as 50 per cent, depending on estimates of the transport and
maintenance costs.
29
‘Logistics’ refers to the movement, handling and storage of goods, from

the supply of raw materials, through production, to the final distribution to
customers. Logistics includes the cost of ordering, moving inputs and outputs,
the cost of storing inventories and the production costs if the delivery of inputs
is delayed (Holl 2006). While faster and more reliable transport affects all these
elements of cost, a CBA at most only generates rough estimates of the impact
of better infrastructure on firm production logistics. This is not surprising
since the impact of new roads and railways on logistics is difficult to estimate.
This is a problem when measuring the impact of transport infrastructure
on modern manufacturing. In modern (just-in-time) manufacturing, when
goods are supplied just as they are needed and the manufacturer only holds
a minimum supply of stocks, the reliability of transport and predictability of
delivery schedules are crucial (Holl 2006). From the wider literature there is
evidence that even crudely measured transport infrastructure does impact on
firm-level logistics.
Datta (2012) examine the impact of the GQ on firm-level outcomes.
They use data from two World Bank Enterprise Surveys for India that
provides a sample of more than 1,000 firms across 37 Indian cities. The data
covers the periods before the highway construction (in 2002) and when the
project was two-thirds complete (in 2005). The results show that the highway
construction allowed firms near the highway to cut costs by reducing their
average input inventory holdings. This effect was equivalent to seven days of
production. In China, the total length of roads doubled after 1990, reaching
2.6 million kilometres by 2008. The total length of expressways increased
from 147 kilometres in 1988 to over 50,000 kilometres in 2007. Li and Li
(2013) examine the impact of this road infrastructure investment on inventory
costs in China. They find that from 1998 to 2007 the inventory–sales ratio of
median and large manufacturing firms in China declined steadily, from 22 to
13 per cent, and so reached a level comparable to developed countries. Road
investments in China, they argue, made a crucial contribution to this decline
as faster and more predictable delivery schedules reduced the need for firms to
hold inventory as a safety buffer.
The studies reviewed in this section are good news for Pakistan in the
sense that they show a consistently positive relationship between infrastructure
(roads and railways) and output growth, improvements in firm-level logistics
performance, and higher productivity. The detailed methodology has yet to be
30
BIG INFRASTRUCTURE
Table 2.1 World Bank overall logistics performance index
Country 2007 2010 2012 2014 2016 2018

Pakistan 2.6 2.5 2.8 2.8 2.9 2.4
Malaysia 3.5 3.4 3.5 3.6 3.4 3.2
Sri Lanka 2.4 2.3 2.8 2.7 – 2.6
India 3.1 3.1 3.1 3.1 3.4 3.2
Bangladesh 2.5 2.7 – 2.6 2.7 2.6
Source: World Bank (2019b).
incorporated into studies of CPEC-related infrastructure. Some of this CPEC

literature has calculated how the new transport infrastructure connecting
China and Pakistan will reduce international travel distance (we discuss this
later in the book) which is at best only a very crude measure of how the CPEC
will impact firm logistics.
The need for such careful research is evident as there is clear evidence
that Pakistan performs poorly in terms of logistics. The World Bank logistics
performance index (LPI) provides a measure of the functioning of logistics
for more than 160 countries. There are six components of the LPI—customs,
infrastructure, ease of arranging shipments, quality of logistics services,
timeliness, and tracking and tracing. The LPI can clearly be used to create
a rationale for the CPEC. Table 2.1 shows that Pakistan generally performs
worse on the LPI than comparator countries. Stagnation, or even decline, in
the LPI between 2007 and 2018 is quite common, with much of this decline
being (as with Pakistan) concentrated in the years between 2016 and 2018.
This does seem surprising when logistics has been revolutionised by the use of
IT-technology during these years. As of 2018 there was still no evidence of a
CPEC impact on logistics in Pakistan.
Table 2.2 shows the LPI of Pakistan between 2007 and 2018. The index
shows some unexplained fluctuations but there is no indication of progress
between these dates. The signing of the CPEC and the completion of the
‘early harvest’ projects has not led to any improvement. By every measure
of the LPI, Pakistan performs worse in 2018 than it did in 2007, though
much of this has been caused by a sharp and unexplained drop between 2016
and 2018.
31
Table 2.2 World Bank logistics performance index: Pakistan
2007 2010 2012 2014 2016 2018

Ability to track and trace 2.6 2.6 2.6 2.7 2.9 2.3
assignments
Competence and quality of 2.7 2.3 2.8 2.8 2.8 2.6
logistics service
Ease of arranging competitively 2.7 2.9 2.9 3.1 2.9 2.6
priced international shipments
Efficiency of the clearance 2.4 2.1 2.8 2.8 2.7 2.1
process
Frequency shipments reach 2.9 3.1 3.1 2.8 3.5 2.7
destination in scheduled
time
Quality of trade- and 2.4 2.1 2.7 2.7 2.7 2.2
transport-related infrastructure
Overall logistics performance 2.6 2.5 2.8 2.8 2.9 2.4
index
Note: 1 = low, 5 = high.
Table 2.3 shows a measure of the LPI of particular relevance to the

CPEC, that of trade- and transport-related infrastructure. The measure is
consistently lower than for comparator countries, with the exception of Sri
Lanka, which provides some evidence that the emphasis of the CPEC is well
placed. However, since the launch of the CPEC, there has been no indication
of improvement. Again, there is the indication of a sharp and unexplained fall
in the measure between 2016 and 2018.
Table 2.3 World Bank quality of trade- and transport-related infrastructure
Country 2007 2010 2012 2014 2016 2018

Pakistan 2.4 2.1 2.7 2.7 2.7 2.2
Malaysia 3.3 3.5 3.4 3.6 3.4 3.1
Sri Lanka 2.1 1.9 2.5 2.2 - 2.5
India 2.9 2.9 2.9 2.9 3.3 2.9
Bangladesh 2.3 2.5 - 2.1 2.5 2.4
32
BIG INFRASTRUCTURE
There are weaknesses with the CBA studies that were reviewed here.
None of the studies considered costs or especially the opportunity costs of
construction. Did the benefits, for example, of being able to transport skilled
people across China on the high-speed railways (HSR) create more benefits
than a counter-factual alternative of increasing the local supply of skilled
people by spending the money instead on improving education? The in-depth
focus of the CBA tends to miss out on these bigger questions. One exception
in the literature is that of Ansar et al. (2016) who examine a database of
95 road and rail infrastructure projects in China, built between 1984 and 2008
and costing roughly $65 billion. They ask whether the projects were profitable,
measuring to see if they generated a benefit-to-cost ratio (BCR) of greater
than one. They also compare the macroeconomic outcomes to a dataset of 806
transport projects built in developed countries. The results show that China’s
cost performance was no better than that of developed countries. Seventy-
five per cent of the transport projects in China suffered cost overruns and
actual costs were on average 30 per cent higher than estimated costs. China
did, however, complete projects significantly quicker, by 2.6 years on average.
This was due to the rapid speed with which China was able to remove and
resettle populations affected by the infrastructure construction and also by
paying less attention to quality, safety and environmental considerations.
The CBA conducted prior to construction in China was hopelessly unreliable.
Two-thirds of the 156 projects generated new traffic flows that were below
forecasts. The average traffic usage in the new roads and railways was 41 per
cent below forecast and for some projects it was 80 per cent below forecast.
The remaining one-third of projects experienced traffic usage more than
60 per cent above forecasts. Chinese planners expected the BCR to exceed
costs by 40–50 per cent. From the sample of 95 projects, 55 had a BCR of less
than 1, meaning that only 28 per cent were genuinely profitable. The YuanMo
expressway, for example, had an expected BCR of 1.5. A combination of
benefit shortfalls and cost overruns pushed the BCR to below 1. This was
caused by a near 50 per cent shortfall in traffic usage on the road. Lost revenue
due to lower traffic volumes were exacerbated by a 53 per cent shortfall in
forecasted toll rates, and a 24 per cent cost overrun. Even after eight years, the
first-year traffic forecast had not been met. These losses did not even include
the costs of population resettlement and the initial land acquisition costs.
The final BCR was between 0.2 and 0.3, depending on estimates of wider
economic benefits (Ansar et al. 2016: 376). Overall, the lower-than-expected
33
benefits and huge spending and debt incurred to construct the road and rail
infrastructure in China generated significant macroeconomic risk. Between
2000 and 2014, according to McKinsey (2015), China invested $29.1 trillion.
This nearly matched the estimated rise in China’s total debt, from $2.1 to
$28.2 trillion. Much of this went into low-productivity road and rail and other
infrastructure projects. This infrastructure development has left a legacy of
macroeconomic risks including debt and non-performing loans as well as the
crowding out of alternative investment opportunities. While there is relatively
low official government debt (55 per cent of the GDP), overall debt is raised
by high corporate debt and financial institution borrowing (by state-owned
enterprises and state banks). China’s total debt-to-GDP ratio is currently
around 282 per cent, compared to 269 per cent in the US and 160 per cent in
Brazil (Ansar et al. 2016).
The absence of easily accessible planning documents for Pakistan makes
it hard to judge the extent to which a detailed CBA has been carried out for
the various CPEC projects. If this work has been undertaken, there is every
reason to believe that estimates for Pakistan will be similarly over-optimistic.
The Medium-Term Development Framework (MTDF) was a government
effort between 2005 and 2010 to invest $36 billion to build and upgrade
infrastructure. Individual infrastructure projects built under the MTDF took
much longer to complete than planned and budgeted for. Irrigation works
took on average 18 years to complete, with the construction time of projects
running between 3.4 and 30.4 years, and 8 years on average for roads, ranging
between 4.6 and 13.6 years. In a survey of 57 contractors, almost 40 per cent
reported delays in completing construction projects ranging from six months
to two years (World Bank 2007b: 5). We should be extremely sceptical about
the promised benefits of CPEC projects in relation to planned costs.
In the extreme, infrastructure projects may not even be constructed with
any realistic expectations of generating enough economic benefits to offset
costs. Construction may be entirely driven by political considerations such
as bringing prestige to a particular politician for being associated with major
construction work and the resulting employment in his/her constituency or
generating an opportunity to hand out construction contracts to political
supporters or business contacts. White elephants are spending projects that
have no economic value. There are many such examples. The International
Olympics Committee (IOC) promised sustainable development as a
crucial legacy for investment in hosting the winter and summer games.
34
BIG INFRASTRUCTURE
In reality the sports facilities created to host these events are frequently left as large
white elephants. Such infrastructure has boosted patriotism and international
prestige but often contributed little to tourism or long-term alternative use for
the facilities. Host populations have been left holding long-term debt burdens.
The long-term plans for the 2008 Greek Olympic legacy seemed impressive.
The Olympic Press Centre was to become the new Ministry of Health, and the
Weightlifting Centre a new university. In practice, within a few years, 21 of the
22 venues were in a state of disrepair (Mangan 2008: 1874). The creation of
white elephants even happens with projects that seem to have a clear economic
rationale. Green (2004), for example, examined a 15-year programme to set
up 52,000 stations to charge solar batteries in northern Thailand at a cost of
$11 million. A survey showed that 60 per cent of solar stations had some
technical problems and 90 per cent failed to work properly during the five-
month rainy season. Reported problems were not acted upon and the centralised
decision-making failed to set up mechanisms to get feedback from users on the
poor functioning. The continued central subsidies were not sustainable.
White elephants can be frequently observed in government programmes.
If firms doing work under a government-financed technology programme
believe that the net benefit to the public is negative, it is not in their interest to
reveal that information to the government or the public. The white elephant is
kept hidden from view for as long as possible. Such information might prompt
the government to abandon the programme and then firms would not make a
profit from it (Keck 1988). The Concorde supersonic airplane in France and
the UK is an example of a technology investment that continued long after
scientists were aware that they would never turn an economic profit. Robinson
and Torvik (2005) show how white elephant infrastructure projects may
occur even without the information complications of advanced technology.
A politician may promise an unproductive investment project for a particular
region and voters will keep the politician in power to continue enjoying the
benefits from the project, knowing that any alternative politician will likely
close it down, not having a support base in that area. It may be thought that
as the CPEC plans were initially drawn up in China, China will be immune
to planning, designing and paying for such prestige white elephant projects.
But this is arguably exactly what did happen in Hambantota, Sri Lanka.
Here China constructed a port, an international airport and an international
cricket stadium in the relatively isolated home village of the then president,
Rajapaksa. Not surprisingly, none of the projects turned out to be viable and
35
the cost, combined with a lack of economic benefits, left Sri Lanka unable to
service its debts to China. Media reports suggest that all the feasibility studies
projected the non-viability of the projects (New York Times 2018). The lack
of easily available information about the financing and expected benefits and
costs of CPEC projects could all contribute to the emergence of CPEC white
elephants.
AGGREGATING MODEST STUDIES TO

MEASURE SOCIAL SAVING
Moving beyond these modest studies of the local impact of single infrastructure
projects, there have been various attempts to measure the aggregate impact
of infrastructure improvements over an extended period of time. Fogel
(1966) pioneered a method called ‘social saving’ to measure the incremental
contribution of US railways to the US economy in the nineteenth century.
He defined the social saving of railways as ‘the difference between the actual
cost of shipping goods in that year and the alternative cost of shipping exactly
the same goods between exactly the same points without railroads’ (Fogel
1966: 34). This was an early attempt to think rigorously about some important
aspects of firm logistics. Fogel argued that even by the late nineteenth century,
the extensive system of existing and potentially expandable waterways in the
US offered a viable alternative to railway transport. Despite the enormous
expenditure on the railways, he argues that by 1890, the social saving of
railways only amounted to less than 3 per cent of US GDP. The reason for the
limited benefits were that the main wheat and cotton growing regions and the
bulk of iron-ore deposits were all located close to natural waterways. Railways
offered few extra economic gains (Fogel 1966). This pioneering work has
inspired a large number of studies attempting to estimate or re-estimate social
savings in different countries and across different time periods.
Other estimates of social savings show wide variations even among
countries with close geographical proximity. Social savings of the Argentinian
railways in the nineteenth century are estimated at 19.9 per cent of GDP in
the case of freight and 1.7 per cent in the case of passengers. These estimates
are four times higher than the estimates of social savings in neighbouring
Uruguay. The key reason for this difference was that a much greater proportion
of the national population lived in the capital city in Uruguay than they did
in Argentina, which reduced the need for rail transport. Also, 67 per cent of
36
BIG INFRASTRUCTURE
Uruguayans lived in regions with a port or with easy access to river navigation,
which only happened with 28 per cent of the Argentinean population.
By 1870 the rivers carried a very large proportion of the total freight transport
of Uruguay and their prominence did not disappear with the construction
of the railway network (Herranz-Loncan 2011). In Brazil unit savings on
freight services were large because waterways were not a viable alternative for
most overland shipments. Navigable rivers were poorly situated and coastal
shipping could only rarely substitute for overland transport. Merchandise
typically went to market on the back of mules carried across scarce and crudely
constructed turnpike roads. The cost of this transport often exceeded the
average charge for railway travel in 1913 by a large margin. Estimates of the
social savings generated by the railways for freight services are not surprisingly
very large, between 18 and 38 per cent of Brazil’s GDP in 1913. There was
a different story for passenger travel. Brazilians in c. 1900 tended to travel
only short distances and with low wages; the value of the time they spent in
travelling was minimal. Using data on stagecoach prices, speed of travel and
travel distances shows that the social savings from passenger travel was under
4 per cent of GDP (Summerhill 2005). Similar to Brazil, social savings were
large in Mexico. Before the railroad, Mexico depended almost exclusively on
overland transportation. Except for some local freight services across three
lakes near the central highlands, there was little scope for using internal water
transport. Most of the Mexican population and economic activity was located
far from the two coasts in plateaus and mountain valleys. Coastal shipping
consequently never played the role it did in Europe and the US. Together, the
geography of Mexico meant that unit savings from railroad freight transport
were high (Coatsworth 1979).
There are predictably many criticisms of Fogel and his pioneering method.
Fogel makes strong assumptions about the feasibility and likely impact in the
nineteenth-century US on the costs of transport caused by a shift from the
actual railways to the counter-factual waterways. Fogel based his estimates
on the actual market prices of waterway transport in the 1890s without
accounting for the possibilities of costs escalating if waterways were burdened
with the extra traffic then carried by the railways. Fogel makes no mention of
the comfort and convenience of passenger travel. This when passenger activity
in 1890 generated 25 per cent of operating revenues for the US railroad
companies. As with the more limited CBA studies, these larger social saving
approaches tend to ignore the wider and indirect benefits of infrastructure
37
improvement. Fogel, for example, does not account for the way in which
railways permitted the massive increase in US grain exports in the nineteenth
century, which provided the foreign exchange necessary to import the capital
goods and technology needed for industrialisation. Chapter 3 shows that
China will benefit from the CPEC by getting greater access to markets and
raw materials from Afghanistan and also from the possibility of importing
oil directly from the Middle East, via Gwadar in Pakistan. The impact on
railways also had a significant impact on various measures of well-being that
are difficult to capture in estimates of social savings. In the US and elsewhere,
the railways allowed grain and other foodstuffs to be transported throughout
the year. In Russia freezing winter temperatures and in India heavy monsoon
rains made transport a seasonal activity. Railways allowed consumers to move
away from seasonal consumption patterns and to maintain consumption levels
more evenly throughout the year (David 1969; White 1976). The root of
these various methodological problems is that there are so many potential
changes induced by an investment in big infrastructure that it is impossible to
convincingly account for all of them. More generally, there are ‘deficiencies of
partial analyses, which accept the existing structure of prices and production’
(David 1969: 513). This is a key claim of many supporters, that the CPEC
will be a ‘game changer’, something which social saving, for all its apparent
benefits, cannot engage with.
Despite these criticisms that are some clear lessons for Pakistan, we can
draw from this literature. The positive impact of large transport infrastructure
depends on some key questions. These questions include how the new
infrastructure will impact on the costs of transportation and whether the
new infrastructure will be more than a marginal improvement over existing
transport options (Coatsworth 1979; Summerhill 2005). We return to these
questions later in the book. Chapter 4 shows that there will likely be large
distance savings in direct transport from Pakistan into Western China.
This is, however, unlikely to have much aggregate impact on trade or passenger
travel. The alternative to the proposed CPEC roads and railways is the
existing transport system of Pakistan. Chapter 4 also shows that this existing
transport system is already relatively efficient as measured by price differences
between cities. The proposed railway connection to Western China, for
example, would likely run alongside the existing Karakoram Highway.
We should not underestimate the Fogel problem of ignoring passenger
comfort and convenience. Whoever has travelled the few hours on the
38
BIG INFRASTRUCTURE
relatively new highway from Lahore to Islamabad will readily appreciate

the unusual sensation of ‘passenger comfort’ in Pakistan. The road is smooth
and fast, and the pleasant drowse for non-drivers is broken only by a visit to
the service station for snacks (including what one Pakistani patriot of my
acquaintance called the best burger in the world). There are none of the traffic
jams or jerky driving amidst chaotic roads that characterise much of the rest
of the road system in Pakistan. Unfortunately, the highway is really a luxury
for the privileged few. The cost of the route means it is relatively empty and
vehicles using it tend to be confined to the luxury end of the market, people
for whom the road is a marginal improvement over flying.
LONG-TERM MACROECONOMIC
STUDIES OF INFRASTRUCTURE
An alternative method is to move away from the micro to macroeconomic
studies that examine the impact of new infrastructure systems on GDP growth
or entire economic sectors such as agriculture, manufacturing or urbanisation.
These studies tend to use time-series data to look at the relation between
infrastructure and economic variables. These studies allow us to think about
big, CPEC-sized macroeconomic questions.
This research has covered the impact of infrastructure on growth in
contemporary developing countries. By 1965, for example, 10 per cent of
villages were electrified across India but this included almost 50 per cent in
Tamil Nadu and less than 3 per cent in West Bengal. Rud (2012) uses the
variation in the timing of electrification to measure the impact of electricity
infrastructure provision on state-level industrialisation between 1965 and
1984. He finds that electrification boosted manufacturing output through
an increase both in the number of factories and output growth among
existing small firms. Xu and Nakajima (2013) examine the impact of highway
construction in China between 1988 and 2011 by comparing the economic
performance of cities with and without highway access. They find that newly
constructed highways boosted industrial growth in connected regions via
more investment and higher output. Remote county-level cities more than
300 kilometres from large cities did not benefit from newly constructed
highways. Highway construction promoted the growth of heavy industry but
not that of light industry. Xu (2016) examines the efforts undertaken by the
Chinese Ministry of Railways between 1997 and 2007 to increase the speed
39
and capacity of and reduce the delays of train travel in China. He finds that
exports from China to Central Asia via railways increased by approximately
30 per cent after the construction project compared with exports via other
means of transport. One interesting result was the impact on the geography
of exporting activities. The improved market access benefited efficient inland
producers mainly in the east of China. The market share of Xinjiang exporters
(located in the far west of China) in exports to neighbouring Central Asia
actually shrank. Y. Lin (2017) examines the passenger (not freight) oriented
HSR project in China. She finds that being connected to the HSR network
in China leads to a significant increase in the GDP and employment of urban
areas. After being connected to the HSR network, a city experienced an
18 per cent increase in the number of passengers travelling by train. There
was a large, 12 per-cent drop in the number of passengers who travelled by air.
This demonstrated that travelling by the HSR is a close substitute for air travel
but not road transportation. There is no significant impact of the HSR on the
volume of goods transported by the railway, which is not surprising as the
HSR is predominantly a passenger service. The study found that an increase
in HSR-induced market access led to higher growth in those industries that
require ‘non-routine cognitive skills’ such as intuition, creativity and human
interaction. In 2000s China, these skills were particularly in demand in sectors
such as finance, insurance, real estate, IT and business services. All these
sectors showed growth as a consequence of HSR-induced market access.
The HSR generated competitive effects through its ability to move highly
skilled employees around China at high speed.
Such studies have also been done for the now developed countries back
when they too could be classified as developing. Haines and Margo (2006)
examine US local (county) level impact of gaining access to a railroad in the
1850s. They use detailed evidence from old maps to test whether a rail line
passed through county boundaries in 1850 or 1860 (or both). Rail access data
is then linked to county-level information on economic outcomes in 1850
and 1860. They use a difference-in-difference approach, which compares
outcomes in a treated group (counties that gain rail access in the 1850s) with
a control group (counties that did not gain rail access in the 1850s) before and
after treatment (rail access). They find that rail access led to only a marginal
increase in land prices (the measure of economic growth as the US was in
the 1850s still a predominantly agrarian-based economy). This small impact
is likely to be because a large fraction of the population lived in counties
40
BIG INFRASTRUCTURE
in 1850 that were already served by rail or water transport (recall Fogel).
Donaldson and Hornbeck (2016) show that as the US national railroad
network expanded from 1870 to 1890, county-level increases in market access
were capitalised into substantially higher agricultural land values. There was
a significant economic impact but it took longer than allowed for by Haines
and Margo (2006). Removing all railways in 1890, Donaldson and Hornbeck
(2016) estimate, would have lowered the total value of US agricultural land
by 60.2 per cent, equivalent to an economic loss of 3.22 per cent of GDP.
Atack et al. (2010) use a different proxy measure for economic growth
and find that railway construction had a causal impact on both population
density and urbanisation in nineteenth-century US. Their results show that
more than half of Midwestern urbanisation in the 1850s can be attributed
to the causal impact of railroad construction. Yamasaki (2017) examines
the impact of railway construction in Japan, which started in the 1870s and
had reached about 60 per cent of all Japanese counties by 1900. He finds
that railway construction stimulated structural change from agriculture into
industry and associated urbanisation. Tang (2014) finds that the construction
of railways after the 1870s in Japan had a positive impact on firm investment
and production.
And finally for the case of contemporary developed countries. Leah and
Munnell (1990) found a positive impact on output, employment, growth and
private investment from public capital investment across 48 US states and at
the national level between 1970 and 1988. By contrast, Garcia-Mila, McGuire
and Porte (1996) found that investment in highways, water and sewers in
the US between 1970 and 1983 across 48 states had no impact on private
output. Cantos, Gumbau-Albert and Maudos (2005) found a positive impact
from the growth of transport infrastructure on private sector production
in Spain between 1965 and 1995, the most significant impact coming
from road investment. The impact was particularly strong for agriculture,
a sector in which Spain emerged as a successful exporter during these years.
The impact of airports was most strongly felt on both agriculture and industry,
and the impact of railways on construction, agriculture and services. Crescenzi
and Rodriguez-Pose (2008) found that the local endowment of transport
infrastructure was a useful predictor of economic growth across 15 European
Union (EU) countries between 1990 and 2003, though the impact lasted
only three or four years and education had a stronger impact on economic
growth.
41
The provision of infrastructure can help explain differing regional patterns

of growth. Demurger (2001) found that investment in transport infrastructure
and telecommunication accounted for a significant part of the observed
variation in the growth performance of 24 Chinese provinces and autonomous
regions between 1985 and 1998. Shi, Guo and Sun (2017) found that
infrastructure had a positive impact on economic growth across 30 Chinese
municipalities and provinces between 1990 and 2013.They find that the growth
impact varied across different types of infrastructure. For example, roads were
negatively related to growth before 2006, though between 2006 and 2013
they had a significant and positive impact. The impact of railway construction
was positive across the full sample period and the impact increased over time,
but the link was not statistically significant. The expansion of telephone
landline usage made a strong contribution to the GDP per worker in the
early 1990s but zero impact after 2006. The results show that over-investment
may have occurred for some forms of infrastructure. This may be the result of
diminishing returns and a crowding out effect on non-infrastructure. Lewis
(1998) found that road and especially water infrastructure had a significant
impact on incomes across 32 Kenyan municipalities between 1993 and 1996.
The generally positive finding has been found to also extend to other
economic variables such as productivity and exports. Aschauer (1989a,
1989b) used data from 1949 to 1985 and found that infrastructure spending
on streets, highways, airports, electrical and gas facilities, mass transit, water
systems and sewers has a significant impact on productivity growth in the US.
The slowdown in public capital expenditure was an important contributory
factor behind the slowdown in US productivity after the early 1970s. Herranz-
Loncan (2006) found that the construction of railways in Spain between 1890
and 1914 boosted national productivity by allowing a (limited) shift from
alternative, less efficient means of transport such as waterways or animal
powered transport. Duranton, Morrow and Turner (2014) found that a
10 per cent increase in highways within a city in the US between 1956 and
2007 caused a 5 per cent increase in its exports measured by weight rather
than value. This suggests that highways pushed cities to specialise in sectors
that have high weight-to-value ratios.
The good news for Pakistan is the very consistent (if not quite
100 per cent) finding that infrastructure has a positive impact on economic
growth and closely related variables such as productivity and exports over time.
42
BIG INFRASTRUCTURE
The relationship holds for developed countries now and in their own past,
for contemporary developing countries, across regions in the same country
and across different types of infrastructure, including all those associated
with the CPEC, roads, railways and energy. Highways consistently have a
greater impact on sectors that require the transport of heavy goods, such as
manufacturing and agriculture. This is good news for Pakistan which hopes
that the CPEC will boost both domestic sectors.
THE MACROECONOMIC STUDIES

BECOME IMPRESSIVE
The empirical scale of some of these macroeconomic studies is truly impressive.
Donaldson (2010) created a new dataset with almost 7 million observations
on district-level prices, output, rainfall and inter-regional and international
trade in colonial India. The rainfall data came from the all-India network of
3,614 meteorological stations that recorded the amount of rainfall at each
station on every day of the year to link local rainfall to local crop output.
This data was supplemented with a digital map of India’s railroad network
in which each 20-kilometre segment was coded with its year of opening.
The dataset allowed him to track the evolution of India’s district economies
before, during and after the expansion of the railroad network. Donaldson
found that when a district was connected to the railroad network, its real
income increased by 18 per cent. Income in those districts that were bypassed
by the railways declined by 4 per cent (a negative spillover effect). He also
found that when a district was connected to the railroad network its real income
became less responsive to local rainfall. The railroad made it easier to import
agricultural produce from other districts if local output was reduced by bad
weather conditions. Researchers have not just collected enormous amounts of
data, they have now started using high-technology to do so. G. Khanna (2016)
measures GDP by using night-time lights as measured by weather satellites.
These satellites circle the Earth 14 times each day and take pictures between
2030 and 2200 hours at night. They use algorithms to filter out other sources
of natural light using information about the lunar cycles, sunset times, the
northern lights and other occurrences such as forest fires and cloud cover.
Given the lack of reliable sub-regional level GDP data in India, this measure
is a useful proxy measure of overall economic activity. Lights data is calculated
43
at approximately every 1 square kilometre, which are then aggregated to

correspond to the boundaries of sub-district levels in India. The results
show that in 1992 a 100-kilometre increase in distance from a highway was
correlated with a 12 per cent fall in income. By 2012 this had halved to about
a 6 per cent difference in income. Khanna interprets these results as showing
that the costs of distance were declining with the construction of highways in
India (and other factors such as market-oriented economic reforms).
There were distinct differences in the impact of highway construction in India
and China. In India highway construction efforts focused on connecting its
largest economic centres. In China the highway programme had the goal of
connecting all intermediate-sized cities with a population above 500,000 and
all provincial capitals regardless of population. Overall China invested about
ten times more in its highway network than India did and so the density of
connections was much greater. Alder (2014) asks what would have happened
to India had it invested enough to follow a similar road-building strategy as
China did. Alder (2014) applies the same policy objective in China to Indian
conditions and chooses routes such that the costs of building the roads on
Indian terrain are minimised. He collected detailed data on the costs of road
construction in India as a function of terrain and calculated transport costs on
the imagined road system between all 590 mainland districts in India using an
algorithm to generate the shortest path journey. In the empirical analysis, the
transport costs are related to income which is measured using satellite data on
night-time lights as a proxy for income. The results show a positive economic
impact, but taking into account the construction costs of the counterfactual
roads, only a modest gain relative to existing infrastructure. This is evidence
again of diminishing returns to road investment in China—only a marginal
boost to growth from ten times the investment!
None of these statistical complexities has been absorbed by the Pakistan-
centric CPEC literature. Compare the richness of the research agenda in these
reviewed studies with Tehsin, Khan and Sargana (2017) who ask whether
the CPEC will contribute to sustainable economic growth in Pakistan. This
study does not define ‘sustainable economic growth’ and does not review any
of the wider theoretical or empirical work focused on sustainable economic
growth. The paper does suggest sustainable growth to be related to security,
government institutions and investment but ultimately the only evidence
that is presented on the ‘economic significance of CPEC’ are two newspaper
44
BIG INFRASTRUCTURE
reports related to some comments made by the IMF representative in Pakistan

and data regarding some seafood exports from Pakistan to Western China.
THE ENDOGENEITY PROBLEM WITH

MACROECONOMIC STUDIES AND
THE SOLUTIONS
The finding from many of these studies is that of a positive correlation
between infrastructure and economic variables such as GDP growth, earnings
and urbanisation. As infrastructure is built, the economy experiences GDP
growth. This relationship does not answer any questions about causation.
This is the well-known problem of the endogeneity of infrastructure or the
problem of working out what causes what. Infrastructure can boost the economy.
Infrastructure can reduce costs of production and so boost the profits and
efficiency of firms leading to more output and investment. Better ports, roads
and airports will increase the ability of a country to export. The greater ease of
face-to-face meetings (recall those HSRs in China) will improve knowledge
exchange. Quicker travel will improve the efficient allocation of resources such
as raw materials and human labour. The direction of causation may work in
the opposite direction. An economy experiencing economic growth, perhaps
caused by economic reforms, more foreign aid or the discovery of oil, will
be generating traffic congestion and capacity shortages in ports and airports.
Disgruntled households and firms may then be motivated to engage in political
mobilisation and protest to seek a policy redress to problems of congestion.
The economic growth will be providing the government with more tax revenue
and a political motivation to look for a solution. Infrastructure provision may
then be expanded in response to the demands and shortages generated by
economic growth. Studies of infrastructure and economic growth have used
various ways to test for causality.
Using Granger’s causality tests to account for causality, Looney (1997)
found that private investment in manufacturing stimulates a subsequent
increase in infrastructure in Pakistan, including energy, non-rail transport
and communications. Infrastructure provision was increased in response to
needs created by the private sector. Choosing what infrastructure to construct
was not a random process but was here done in response to prospects
for economic growth or economic growth that had already occurred.
45
Chandra and Thompson (2000) used data on interstate highway construction

and economic activity in the US from 1969 to 1993 to engage with this problem.
They paid close attention to smaller cities on the assumption that these
cities had gained access to an interstate highway because they coincidentally
happened to fall on the transport route between big cities and had not been
purposely selected to receive a new highway. The cities were connected by
accident so there was no problem with causality. They found that the total
earnings increased (between 6 and 8 per cent) in these counties, mainly in
the services and retail industries. This was evidence that infrastructure was
causing economic growth. This method was replicated by Hornung (2014)
who studied the link between railroad construction and economic growth
using data for 978 Prussian cities between 1838 and the mid-1860s. To deal
with the problem of endogeneity, they assumed that the purpose of railways
was to connect important cities and hence smaller cities located on a direct
line between these important cities gained access to the railroad by chance.
They found a causal and significant impact from railways to urban population
growth. The mechanisms worked through inducing migration and allowing
factories in cities with railroad connections to expand in size. Another method
to find the causal pattern of highway construction on trade in the US was used
by Duranton, Morrow and Turner (2014). They used the exogenous variation
in exploration routes between 1528 and 1850, in railroad routes circa 1898, and
in a 1947 plan of the interstate highway network to study the effect of highway
construction on bilateral trade between large US cities. As noted earlier, they
also found a causal trail from infrastructure investment to economic growth.
Li and Chen (2013) found that in Western China in the 1990s, the railroad
showed a problem of one-way congestion. Only goods shipped eastbound
through the railroad were constrained by the rail-shipping capacity while those
shipped westbound were not. Hence, the westbound trade was not expected to
have been affected by the expansion of railway capacity and so could be used
as a control group to check for the direction of causation. They found that
within three years of railway construction, there was a 40 per cent increase in
goods being shipped eastwards and the origin–destination price differences
(a measure of market efficiency) had decreased by about 30 per cent. In contrast,
both the volume and price of transport changed little for the westbound goods.
The evidence was consistent with the implication of the one-way capacity
constraint feature. This is evidence that infrastructure was driving economic
change and having a sizeable impact. They estimated an annual rate of social
46
BIG INFRASTRUCTURE
return to the project of between 10 and 50 per cent. Banerjee, Duflo and Qian
(2012) delved into China’s history to account for causality in contemporary
China. In the late nineteenth and early twentieth centuries, the Chinese
government and a set of western colonial powers built railways connecting
the historical cities of China to each other and to the newly constructed treaty
ports. Banerjee, Duflo and Qian (2012) compared areas closer to these lines to
areas further away and interpreted the result of this comparison as the overall
effect of any transportation infrastructure, including both the original railways
and any other infrastructure added later. This method provided the research
with an exogenous source of variation in access to the transportation networks
that dealt with the endogeneity problem and allowed the authors to conclude
that transport investment caused economic change in China between 1986
and 2006.
There is no discussion or engagement with the issue of endogeneity or
causality in the Pakistan CPEC literature. Is this a potential problem for
thinking about the likely impact of the CPEC? Thinking about China–
Pakistan relations and of China’s growing involvement in the global economy
after 1979, it is clear that the ‘economic success’ or otherwise of Pakistan was
never a motivating factor in China’s growing engagement with Pakistan.
The milestones in the relationship were never economic change in Pakistan.
Pakistan was able to facilitate better relations between China and the US in
1971; Pakistan provides China with access to a deep-water port at Gwadar;
Pakistan can be a conduit of oil from the Middle East to China; and Pakistan
has a useful role to play in limiting Islamic extremism in Western China. China
is not initiating the CPEC in response to prior economic success or failure in
Pakistan. Once we look more closely at the exact plans for the CPEC, there
are, however, some doubts. The eastern route of the CPEC that runs through
the more populated and economically successful Punjab does seem to have
been cited to take advantage of prior economic growth and industrialisation.
The western route is frequently justified in the CPEC literature on the claims
that it will bring development to the more backward regions. Again, prior
economic growth, or the lack of it, has influenced the location of the CPEC.
So, the location and routes of the CPEC at a more disaggregated level are
influenced by prior economic performance and any future analysis should
consider seriously the endogeneity problem. The studies reviewed here have
provided a rich array of methods to do this; it just needs the academic will to
incorporate them into Pakistan studies.
47
IN SUMMARY
There is a massive and rigorous existing literature that has looked at the impact
of big infrastructure projects on economic outcomes, ranging from firm-level
to the macroeconomic. None of this is referred to in the existing CPEC
literature. One positive for Pakistan from this review is the consistent and
positive story of the economic impact of big infrastructure on employment,
investment, growth and productivity. We must remember though that the
positive impact is highly conditional, dependent on whether new transport
links represent a transformational change to the existing transport system;
the opportunity cost of investment is important and very marginal gains to
infrastructure financed by massive borrowing are possible; the motivations for
building infrastructure are important and the threat of white elephants, or
projects with no economic rationale or infrastructure enclaves that exist only
to serve another country, are real possibilities. We also need to think about
distributional issues which we will turn to in a later chapter. The following
chapter explores this last point in more detail and asks whether the CPEC
will create isolated pockets of China inside Pakistan or generate spillovers that
benefit the wider Pakistani economy.
48
3
CPEC Spillovers Rippling Outwards
Will the infrastructure investments of the CPEC have a transformative impact

on Pakistan’s economy? This chapter introduces the theoretical concept of
the ‘leading sector’ and asks whether the expansion of a single economic
sector (such as infrastructure) can generate sufficient spillovers to other
sectors so as to boost wider economic growth. We should ask whether public
investment in infrastructure will crowd in private investment and whether
CPEC infrastructure spending will generate employment for local workers
and demand for locally produced inputs in construction. The use of economic
theory has hitherto been conspicuous only by its absence from existing studies
of the CPEC.
EFFICIENT MARKETS AND THE CPEC

The CPEC literature often focuses on the reductions in distance that will be
created for China by the CPEC investment. Table 3.1 shows the dramatic
savings in travel distance the CPEC will make to trade for China. Instead
of goods from central China being transported first to Shanghai, then from
Shanghai by sea to the rest of the world, they can travel directly across Western
China, down through Pakistan and via Gwadar to the rest of the world.
The CPEC infrastructure will reduce the distance from Central China to
the Middle East by 7,580 miles and more than 10,000 miles from Western
China. This will minimise the fuel and travel time from 45 to 10 days
and avoiding security and robbery risks in the sea-bound Malacca route
(Ahmad 2017).
49
Table 3.1 Saving in terms of distance (via Shanghai versus Gwadar)
From To Via Shanghai Via Saved Saved (%)

(miles) Pakistan (miles)
(miles)
Central China Middle East 11,206 3,626 7,580 68
Central China Europe 17,801 10,928 6,873 39
Central Asia Pakistan 10,601 3,081 7,520 71
(Gwadar)
Western Middle East 12,537 2,295 10.242 82
China
Western Europe 19,132 9,597 9,535 50
China
Western Pakistan 11,932 1,750 10,182 85
China (Gwadar)
Source: Ahmad (2017: 87).
There is an assumption in much of the CPEC literature that if infra-

structure reduces distance and thereby creates more ‘efficient’ markets, this
must be a good thing for Pakistan. This idea does have a solid academic lineage
in the writings of neoclassical economists. In such writing, efficient markets
contribute to the optimal allocation and reallocation of resources to their most
productive uses in response to changing prices and profit opportunities. This
creates the necessary conditions for economic growth. The CPEC literature
makes this assumption uncritically and ignores the varied impacts that history
and the contemporary world demonstrate that efficient markets may have.
Efficient markets can have a devastating impact. More ‘efficient’ markets
contributed to the massive famines in railway-era colonial India. The failure
of crops in Orissa in 1865–6 occurred in a region without roads and ports
which made it harder to import food supplies from outside. Thereafter, it was
expected that as the railway network spread, any food crop failure would be
alleviated by supplies being imported from cheaper/surplus areas. The famine
in 1868–9 in the North-West Frontier occurred in an area well supplied by
the railways. Even after massive railway construction, the estimated mortality
from starvation and disease crossed 1 million in the Deccan 1876–8, and
North-West Provinces in 1877–8, the country-wide famine in 1896–7 (an
estimated 4.5 million dead) and also in 1899–1900. The railways had instead
50
CPEC SPILLOVERS RIPPLING OUTWARDS
changed the structure of production in Indian agriculture and facilitated a

general shift to producing cash crops for trade rather than food crops for local
consumption. The output of food crops per head stagnated in British India
from 282.41 kilograms in 1885 to 287.95 kilograms in 1895 (Habib 2006:
84). The second effect of the railways was to connect inland areas to ports
which facilitated the export of foodgrains, especially rice and wheat. In 1875
British Indian ports exported 1.22 million tonnes (2.3 per cent of foodgrain
production) of foodgrains and in 1895 about 2.49 million tonnes (3.9 per
cent). During the famines of both 1896–8 and 1899–1900, averting these
exports would have been enough to avert the famines (Habib 2006).
Efficient markets may have a minimal impact on the domestic economy.
One of the most striking examples of absent spillovers is that of transnational
oil company investment in Equatorial Guinea. US oil and gas companies have
invested $50 billion and supply 17 per cent of US net crude oil and oil products
from Guinea. The country has a per capita income equal to Denmark. The lived
reality for Guineans is sporadic electricity, rampant malaria and typhoid, lack
of running water in homes, open sewage systems, and poor provision of public
services including food, health and education. The expatriate workers live in
gated compounds with manicured lawns, landscaped gardens, paved roads,
fire hydrants and food shipped directly from Europe or the US. The largest
gated compound generated enough electricity to power the entire country for
24 hours every day. Even the telephones inside the gated compounds have
Houston area dialling codes. Locals come and go as maids and gardeners,
must wear badges inside and public transport is prohibited from entering the
compound. Entry and exit are strictly monitored; locals need an invitation
to enter and are compelled by curfews to leave (Appel 2012). A news story
appeared in 2018 which seemed to hint at such a trend emerging in Pakistan.
It has often been suggested that Chinese residents in Gwadar and elsewhere
live in gated communities that are isolated from the rest of Pakistan.
In February 2018, 1.3 tonnes of vegetables were carried by China Southern
Airlines from Urumqi to Islamabad. This was the first time a shipment of
vegetables was air freighted from Urumqi. The vegetables consisted mainly
of Chinese yams, mushrooms and chilli peppers. The delivery was to supply
the Chinese in Pakistan who were unable to return to China for the spring
festival. There was a suggestion that such exports would continue after the
new year (Hortidaily 2018).
51
THE CPEC: A TRANSIT CORRIDOR OR AN

ECONOMIC CORRIDOR?
An important question for Pakistan is whether the CPEC will evolve as a
‘transit corridor’ (minimal domestic spillovers) or an ‘economic corridor’
(Esteban 2016). A transit corridor would be created by linking Xinjiang in
Western China with the Indian Ocean port of Gwadar in south Balochistan.
Pakistan may simply become a transit route that increases China’s trading
access to the rest of the world. This access would include the Chinese need
for imported natural resources. Since 2009 China has overtaken the US as
the world’s largest importer of crude oil and also imports natural gas and coal.
More than 50 per cent of China’s oil comes from the Middle East but 80 per
cent of China’s oil needs come through the Straits of Malacca across nearly
10,000 miles of mainly ocean (D. P. Nicolas 2015). China has also recently
increased its energy imports from Russia but remains wary about national
restrictions on Chinese investments or joint ventures in Russia’s energy sector.
China has been alerted by past demonstrations of Russian willingness to
use the energy dependency of other states as a leverage to exercise political
pressure as was the case with Ukraine (Brugier 2014). Pakistan’s geo-strategic
location provides Beijing alternative routes for oil and gas supplies from the
Middle East. The Gwadar Port complex was inaugurated in December 2008
as a deep-sea port. In April 2015 Pakistan handed over 40-year operational
control rights of the Gwadar Port to Chinese Overseas Port Holdings Ltd.
A crucial role of the CPEC is likely to be the direct import of oil. Gwadar
is seen by many scholars as a central part of China’s rise and it ‘is another
important piece in Beijing’s economic expansion and transition to being a
global power’ ( Jan and Granger 2016: 293). The Iran–Pakistan gas pipeline
was conceived in the early 1990s and while the Iranians have completed their
560-mile portion of the pipeline, construction of the pipeline on the Pakistani
side was long delayed due to US-led international sanctions against Iran.
China has agreed to assist Pakistan in constructing its 485-mile section.
The China Petroleum Pipelines Bureau (CPP) began construction of the
liquefied natural gas terminal in Pakistan and the Iran–Pakistan pipeline under
the CPEC. The two projects were scheduled to cost around $2.5 billion with
the Chinese company pledging 85 per cent of the investment (Chaziza 2016).
In November 2017 a spokesman for the prime minister said that Pakistan is
ready to complete the gas pipeline with Iran if international sanctions on Iran
were lifted (Samaa TV 2017). China could save around $2 billion per annum
52
if it were to use the CPEC to import 50 per cent of its current volume of
oil supplies (Esteban 2016). In January 2020 the media were reporting little
progress on the pipeline, which they blamed on US pressure and that Pakistan
faced sanctions for violating the terms and conditions of the original agreement
(PressTV 2020). The completion of the CPEC might also enable Beijing to
better access Afghanistan’s mineral wealth. This would require that China rely
on the influence of Pakistan to ensure there is no spillover of radical Islam
from the wider region into Xinjiang. This has become particularly important
in light of the US decision to pull troops out of the region (Chaziza 2016).
For China the cost to Western and Central China’s economic trade with
Central Asia, Middle East, Europe and Africa will be reduced. As well as the
import of raw materials, the creation of a transit corridor would help Chinese
firms export goods through Pakistan and, via Gwadar, to the rest of the world.
Recall the case of the Suez Canal discussed in the introduction to this book.
The canal has been one of the key transit points of the world economy for
150 years but has had little impact on the Egyptian domestic economy, save
for transit feed being the third largest source of remittance income. This is the
fate for Pakistan seen by some observers. If the CPEC is finally only a transit
corridor, its potential to foster socio-economic development in Pakistan will
be limited (Esteban 2016). If the CPEC serves only as a transit corridor, the
spillovers for Pakistan will be minimal and ‘I would not like the people of
Gilgit-Baltistan only selling eggs and fixing tyres of those who would travel
from China to Gawadar’ (Choudhury 2017: 226).
Recall from the introduction how the US used its economic and military
muscle to maximise its own benefits from the construction of the Panama
Canal or how the construction of the Suez Canal tipped Egypt into
bankruptcy and colonial conquest. History gives us further cause for concern.
Historical case studies show that the emergence of a ‘corridor’ is more likely
the greater the role and influence of outside investors. Historical evidence
(1850 to 1900) on railway building in the Balkans shows that imperial
governments (the Austro-Hungarian and Ottoman Empires) and outside
investors (such as Germany and the UK) tended to build railways that
better integrated the entire region and opened it up to international trade
by connecting up big cities across international frontiers. In the period from
the end of the nineteenth century to the 1950s, the national interests of the
by now independent Balkan states prevailed. The political fragmentation of
the region induced by the independence of many smaller countries changed
53
the nature of railway-building. Newly independent national governments in

Bosnia, Serbia, Hungary and others switched to building railway systems that
integrated the national economy by linking up towns and cities. The density of
these networks increased but connectivity between the countries in the region
declined (Stanev, Alvarez-Palau and Marti-Henneberg 2017).
The opposite case is that of an economic corridor. This is the case whereby
infrastructure investment stimulates local economic links. An economic corridor
would occur if the CPEC stimulated economic development within Pakistan.
The Government of Pakistan certainly hopes that this will be the outcome.
A key part of the CPEC for Pakistan has been the proposal of nine SEZs
across Pakistan. The most advanced of these is the 9 square-kilometre Gwadar
SEZ which will accommodate industrial units for mines and minerals, food
processing, agriculture, livestock and energy. It is hoped that the SEZs
will attract Chinese investment, technology and know-how that will boost
industrial growth and help diversify the structure of Pakistani economy
(Esteban 2016). Chapter 5 discusses the proposed SEZs in more detail.
THE CPEC AND CROWDING IN PRIVATE INVESTMENT

There is a long-standing body of theoretical work, from the early years of
development economics, that emphasised the importance of a ‘big push’
(something like the CPEC) to launch a poor developing county into ‘self-
sustained economic growth’. Rosenstein-Rodan (1943) argued for the
‘simultaneous planning of several complementary industries’ on the basis
that employment and income growth in each would create a corresponding
linkage through the demand for the output of the other industries and lead to
broad-based sustained economic growth. Rostow (1956) writes of a ‘take-off
into self-sustained growth’ when over two or three decades the economy and
society transform themselves in such way that subsequent economic growth is
more or less automatic. There is more to this than just policy change as these
‘[i]nitial changes in method require that some group in society have the will
and the authority to install new production techniques’ (Rostow 1956: 154).
Relevant for our study of the CPEC is his argument that the
beginning of take-off can usually be traced to a particular sharp stimulus.

The stimulus may take the form of a political revolution which affects the
balance of social power and effective values, the character of economic
54
institutions, the distribution of income, the pattern of investment outlays and

the proportion of potential innovations actually applied. (Rostow 1960: 36)
Could the unusually wide support for the CPEC in Pakistan encompassing
both political and military elites be likened to such a Rostowian ‘political
revolution’?
Rostow provides us with a useful framework in which we can analyse the
success (or otherwise) of the CPEC. A take-off, he argues, requires three
related conditions:
Firstly, a rise in the rate of productive investment from 5% or less to 10%

of national income. Secondly, the development of one or more substantial
manufacturing sectors, with a high rate of growth. Thirdly, the existence
or quick emergence of a political, social and institutional framework which
exploits the impulse to expansion in the modern sector and the potential
external economy effects of the take-off and gives growth an on-going
character. (Rostow 1960: 39)
The relevance of this to contemporary Pakistan at first glance may appear a

little tenuous. Pakistan boosted its investment from 5 to 10 per cent of GDP
and beyond in the early 1950s and it has remained well above 10 per cent ever
since. Rapid manufacturing growth that created a modern industrial sector
(textiles) can be dated back to the 1960s. As was shown in Figure 1.1, Pakistan
has experienced an average of 5 per cent GDP growth per annum and,
according to World Bank data, has not had a recession since at least 1960.
If that is not ‘growth with an on-going’ character (McCartney 2011b),
then what is? So perhaps here, we should modify Rostow. We should not
be analysing the CPEC as potentially initiating a take-off but instead as
potentially restarting a stalled take-off.
There are important market failures in public investment. These include the
sheer length of many public investment projects; big irrigation networks may,
for example, take 10 or 20 years to complete, which makes the private sector
reluctant to participate. Big infrastructure may be too large and long-term
for domestic capital markets to finance (Pakistan has an enduring problem
with supplying long-term lending for industry, see Chapter 4). Much of the
economic benefit from big infrastructure is external to the original investment.
The benefits of power supply, for example, are not just apparent in the higher
profits of power stations but in the opportunities for profitable investment
55
created for industrial firms and farmers (Hirschman 1958). There are also
coordination problems. For a landlocked region (in Pakistan) or country, there
are crucial external benefits from investment in transport infrastructure by
regions or countries lying between them and the coast. Improving the railway
line in Punjab, for example, would have little effect unless there was similar
improvement to the line as it passed through Sindh or Balochistan on the
way to the ports in Gwadar or Karachi. Why should Sindh or Balochistan
take into consideration those external benefits to investment for Punjab?
These market failures create a rationale for the government to provide public
investment. This leads us to question whether the increased Pakistani public
and Chinese investment in infrastructure will crowd in domestic private
investment. Crowding in occurs when private sector investment is conditional
or contingent on public investment. This is an example of a spillover.
There has been some statistical work to question whether public investment
(in transport and infrastructure in particular) crowds in private investment
in the Pakistan context. The general finding is that public investment has a
positive impact on private investment (A. H. Khan 1988; Hyder 2001; Naqvi
2002; Ahmed and Qayyam 2007) though some argue the opposite (Ghani
and Ud Din 2006).
We can ask whether the CPEC is likely to crowd in enough private
investment to promote rapid economic growth in Pakistan. The investments
in energy and infrastructure are projected to amount to $55 billion by 2030.
There are valid reasons to believe that this volume of public infrastructure
investment is simply not enough to crowd in any substantial amount of private
investment. The CPEC investment represents about 19 per cent of Pakistan’s
current GDP (about $280 billion), or roughly 1.5 per cent of GDP per annum
over the next 10 years. The CPEC is projected to boost Pakistan’s investment
ratio from 15 to 16.5 per cent of GDP over the next decade (Iqbal 2017).
CPEC investment would represent about 6 per cent of the annual investment
budget for Pakistan over a 15-year period (Husain 2017: 4). This increase
is not large and would remain well below the 30–35 per cent investment
shares that have characterised rapid (8 per cent or more) and sustained growth
rates elsewhere in the world, such as China after 1978 or India after 2003.
Figure 3.1 shows that during the 1990s economic growth in Pakistan
fluctuated around a declining trend, falling to as low as 1 per cent in 1997.
From 2001 and particularly after 2003, growth accelerated rapidly, reaching
nearly 8 per cent in 2004–5, which made Pakistan then among the fastest
56
Figure 3.1 Pakistan GDP growth (annual %)

growing countries in the world. This growth was broad-based; agriculture

grew by more than 6 per cent in both 2004–5 and 2005–6, manufacturing by
around 15 per cent per annum in 2003–4 and 2004–5 and the service sector
by 8–9 per cent in 2004–5 and 2005–6. This is the pattern of growth that the
CPEC optimists are hoping for.
The investment consistent with the slowing economic growth of the
1990s and rapid acceleration after circa 2003 is clear. Figure 3.2 shows that as
growth ebbed in the 1990s, investment as a share of GDP was also declining,
from about 21 per cent of GDP in 1992 to 16–17 per cent of GDP in 2003.
The slowest growth in Pakistan’s recent history is consistent with the level of
investment that can be funded by the CPEC. The acceleration in economic
growth after circa 2003 saw a surge in investment to about 23 per cent of GDP
between 2006 and 2008. This was a good performance relative to Pakistan’s
own history though lagged behind levels in other rapidly growing Asian
economies during the 2000s, including India (30 per cent or more) and China
(40 per cent or more). There is no surprise to this close relation. Investment
has been found to be the variable most closely associated with economic
growth (Levine and Renelt 1992) and specifically with a literature focusing
on explaining episodes of boom and bust (Hausmann, Pritchett and Rodrik
2004; Jones and Olken 2004).
57
Figure 3.2 Pakistan gross capital formation (% of GDP)

Figure 3.3 shows that India also experienced an economic boom around
2003. The differences were that growth went to higher and more sustained
levels in India and was accompanied by a bigger and more sustained rise in
investment, to over 35 per cent of GDP. The slowdown did come, but not until
2011 rather than 2005 as in Pakistan.
Aside from these concerns about the small quantity of CPEC investment,
there are also reasons to doubt the likely quality of CPEC investment.
Figure 3.3 A comparison with India

58
The Government of Pakistan responded to a perceived infrastructure

shortage in the mid-2000s by planning extensive investment. The Medium-
Term Development Framework (MTDF) allocated $36 billion to upgrade
roads, railways, air, power, water, irrigation and other infrastructure. A review
of the complete implementation process (planning and approvals, financial
allocations, detailed engineering and physical construction, and starting
operations) showed weaknesses in all these stages. Individual infrastructure
projects took a long time to complete. It took on average 18 years to complete
projects in irrigation and power, with some projects taking more than 30 years.
Roads took an average of 8 years to complete. In a survey of 57 contractors,
almost 40 per cent reported delays in completing construction projects ranging
from 6 months to 2 years. Almost half of the contractors reported lodging
claims for additional compensation to cover delays and unanticipated costs.
The World Bank concluded that any ‘attempt to meet the expanded public
infrastructure needs through existing processes, resources and skills, will lead
to colossal wastage of scarce resources and frustrate all efforts to meet delivery
targets’ (World Bank 2007b: 5). Chapter 1 noted the impressive speed with
which some ‘early harvest’ CPEC projects have been completed. So, has the
China connection helped mitigate these long-standing quality concerns?
PRODUCTION SPILLOVERS
Spillovers extend much beyond the market failures that may tie public and
private investment together. We also need to consider production spillovers
that may tie together different economic sectors.
Walt Rostow is best known for his pioneering work on the importance of
‘the leading sector’, a sector whose growth can pull up the rest of the economy.
From a study of history he plucked various examples of such ‘leading sectors’.
These included the Swedish pulp industry (1890–1920), cotton textiles in
Britain (1819–48) and silk thread exports from Japan (1900–20). Rostow even
argued that the ‘growth and modernisation of the armed forces’ played a role
as the leading sector in the take-off of Germany, Japan and Russia, something
of evident potential relevance to Pakistan. What makes his work particularly
relevant for thinking about the CPEC is his argument that the introduction of
the railroad has been ‘the most powerful single initiator of take-offs’ (Rostow
1960: 56). Railways were, he argues, ‘decisive in the US, France, Germany,
Canada, Russia and played a very important part in Sweden, Japan’ (1960: 56).
59
The railways, he concludes, made three significant contributions to the take-

off. First, they reduced internal transport costs, widened the market size
for producers and brought new products to market. Second, the railways
permitted countries to boost exports by transporting goods from interior
regions to ports or across borders. Third, the railways generated backward
spillovers that stimulated the growth of other modern sectors such as coal, iron
and engineering. There was, noted Rostow, nothing automatic about these
railway-induced benefits. The expansion of railways contributed much less to
economic growth in nineteenth-century India and China, Canada pre-1895
and Argentina.
History and economic theory also offer us a counterexample to the
leading sector of Rostow and show why there is nothing automatic about
the developmental benefits of big infrastructure projects. The best example
is that of the plantation or mechanised mining project. These have been
termed ‘enclave sectors’. In enclave sectors the original investment is often
financed from overseas, the entire output is exported, the capital equipment is
imported, the skilled personnel and senior management are from overseas, and
the bulk of any profits or royalties earned on the investment are remitted back
to shareholders in the home country. These projects generate so few spillovers
in the host economy that they might as well be labelled ‘domestic investment
on the part of industrialised countries’ (Weisskoff and Wolff 1977: 608).
In the extreme as was discussed in Chapter 2, an investment project may not
generate economic benefits for anyone; it may be a white elephant.
The empirical measure of the impact of a leading or enclave sector is
‘spillovers’. The empirical measure of spillovers from any particular industry is
the proportion of its total output that does not go to final demand but rather
goes to other industries and also the proportion of its input that are sourced
from other industries (Hirschman 1958). Albert Hirschman pioneered much
of this work in the 1950s. Using data from the US, Japan and Italy, Hirschman
found that iron and steel created more spillovers than any other industry.
He also found significant spillovers in chemicals, coal and petroleum products,
paper and paper products, non-ferrous metals, and textiles. Agriculture, he
found, has few spillovers; much of its output is destined for immediate
final consumption or exported in unprocessed form. There is some scope to
boost spillovers from agriculture, by, for example, using locally constructed
machinery in production or processing raw materials, such as coffee beans
into coffee powder, to add value (Hirschman 1958: 106–7). More recent work
60
has extended the emphasis on spillovers through demand. Other sources

of spillovers include firms learning from the production, managerial or
organisation methods in other firms (imitation); when employees learn skills
in a leading firm and transfer them to other firms (skill transfer); when a
leading firm forces other firms to adopt new technology and become more
efficient in order to compete in the market (competition effect) (Gorg and
Greenaway 2004).
Crucial for our thinking about the CPEC is the extent to which FDI
may induce spillovers elsewhere in the economy. The potential for spillovers
is likely to be greater than for domestic investment as MNCs are likely to
have access to more advanced technology. The source of spillovers is similar
to that already discussed. Local firms may respond to competition from FDI
firms by adopting new technologies to defend their market share. Local
firms may emerge to supply the FDI firm with inputs. Local spillovers
may actually decline if the FDI displaces the local production of inputs in
favour of imported alternatives. How local firms respond to FDI firms
armed with more sophisticated technology is likely to depend on a host of
local characteristics. Firms that are larger, more market-oriented, use more
skilled labour and have easier access to investment-financing bank credit
are those more likely to respond positively and stimulate the creation of
more local spillovers (Seyoum, Wu and Yang 2015). One concern about the
long run, mentioned but not evaluated in detail in the CPEC literature, is
that massive infrastructure investment from China may lock Pakistan in to
importing China-complementary inputs and technology in the future.
For example, Chinese-built train tracks may give China a monopoly on
supplying compatible trains. Other examples may include fibre optics, global
positioning system (GPS) and telecommunications where Chinese technology
standards may become locked in (Choudhry 2018: 12). This could have the
effect of reducing potential domestic spillovers.
Concern about the apparent absence of domestic spillovers was noticeable
even in the apparently highly successful Indian software–IT sector during
the 1990s. Throughout the 1990s output and export growth in the sector
consistently exceeded 30 per cent per annum. By 2004–5 total revenues
of the IT services and software sector reached $16.5 billion and exports
$12.2 billion. India had never recorded such rapid growth rates in any
manufacturing sector in its history (Balakrishnan 2006). Throughout
the 1990s more than 90 per cent of software–IT output was exported.
61
Software firms were producing for markets overseas and their learning was
geared to the needs of the high-technology economy of Silicon Valley in the
US. How was such learning of relevance to the developing-country needs
of India? The skills acquired by software engineers structured around this
foreign engagement not surprisingly more often led them to physically migrate
overseas than to find other well-paid opportunities for their skills within India.
The software sector was highly dependent on imports of finished software
and hardware, meaning that despite the impressive headline export figures,
the sector’s net exports were much lower (Chakraborty and Jayachandran
2001; Balakrishnan 2006; Kapur 2007). Something seemed to change in the
early 2000s and observers became much more optimistic about the spread of
domestic spillovers (Kite 2013). Domestic entrepreneurs started developing
software apps that were well suited to Indian conditions. A notable example was
Paytm, which allows people to make tiny payments online for small purchases
from informal roadside vendors. Many of these start-ups were launched by
professionals who had gained experience with domestic software–IT firms and
then left to launch their own firms. Skills started being transferred to the rest
of the Indian economy rather than disappearing into international migration.
HISTORICAL AND CONTEMPORARY EVIDENCE

OF SPILLOVERS FROM INFRASTRUCTURE
There is a dearth of any work on spillovers from the CPEC; this must become
a priority for research. The lesson from the study of spillovers is that we need
to go beyond measures of the success of the CPEC itself—freight volume at
Gwadar or transit volumes on the new highways, for example. The expansion
of one sector can promote economic growth through wider spillovers and
these need to be carefully considered in any discussion of the CPEC. There
are at most some vague hints in the CPEC literature that the CPEC may
lead to a few such local spillovers. These hints tend to be very anecdotal—
for example, that much of the equipment for coal-fired power stations comes
from dismantled (polluting) power stations in China. This sort of statement
is often dropped into discussion without any data or effort to think about its
aggregate importance. There is a huge range of historical and contemporary
studies of how big infrastructure has generated domestic spillovers, none of
which is referenced in the CPEC work, but that do provide important lessons
for Pakistan.
62
Connell (1885) argues that railways in colonial India failed to promote the
spillovers that would lead to broad-based economic growth. India, he argued,
lacked the surplus land that would permit a turn away from subsistence
agriculture towards export-oriented cash crops. Other authors have instead
emphasised the positive impact of the railways on agriculture but noted that
this was at the expense of domestic industrialisation. The more common view
is that the railways did permit agriculture to diversify and produce new crops
for the export market, including indigo, opium and cotton. The flip side of
this was that (remember the concern about transit corridors) the railways
permitted the easier import of manufactured goods. India’s traditional cotton
textile industry declined between 1820 and 1860; first Indian exports of cloth
declined and then later handwoven cloth and hand-spun cotton yarn lost
market share to imported cloth and yarn manufactured in English mills (Roy
2002). By 1880–1 British manufacturers were supplying more than half of the
total domestic cotton cloth consumption (Habib 2006: 94–5). Other sectors
to decline included the jute handloom weaving and silk of Bengal, Kashmir
shawl manufacture in Srinagar, hand paper, glass and iron (Habib 2006).
The British Indian government supported a ‘Buy British’ policy whenever
possible (Habib 2006). The Suez Canal opened in 1867 and cut the costs of
exports from the UK further and allowed heavy engineering goods such as
engines and rails to be exported from Britain to India at a low cost. These
spillovers were not just the consequence of market forces building on existing
competitive strengths. The pattern of spillovers was also influenced by
government policy. Except for ballast for railway tracks and coal, everything
needed for railway construction and usage, right down to railway sleepers (the
wooden planks upon which the metal track rests), was legally obligated to be
imported from Britain. Railway materials and stores counted for 7.3 per cent
of total Indian merchandise imports in 1897–8. The stimulating spillover
effects were felt by British industry alone (Rothermund 1993; Habib 2006).
Others have emphasised the more positive, if harder to measure, spillover
impacts of the railways. The Indian National Congress met for the first time in
Bombay 1885 and could not have done so without the railways which brought
in delegates from distant provinces. The growing Indian press depended on
the railways for their circulation (Rothermund 1993). The postal system was
started in the 1850s and carried by the railways. The number of letters and
packets carried by post increased from 85 million in 1869 to 1,043 million in
1914 (Habib 2006). Without politics, the press and the post, the growth of
nationalism in India would have been much harder.
63
There is a similar story regarding railway investment in nineteenth-century

Brazil and Mexico. In Brazil railway infrastructure investment led to greater
specialisation in agricultural exports (especially coffee from Sao Paulo) and
parallel dependence on foreign manufactured imports. Backward spillovers
from the railways were limited and did not lead to the growth of modern
sectors in Brazil, such as modern coal mining and the iron and steel industry.
A few backward spillovers were established and promoted the growth of
logging for fuel, the establishment of maintenance shops and eventually
the assembly of some rolling stock. This lack of backward spillovers can be
explained by the absence of established industrial skills and also of easily
accessed iron ore deposits and coking coal. There was some positive impact
on skill development as the railways created a cadre of professional staff,
managers, regulators and planners in the public sector (Summerhill 2005).
On the Mexican Central Railway (the country’s longest), minerals and
fibres accounted for 1.3 per cent of total freight tonnage transported in 1885
and almost 60 per cent in 1908. There was also a big increase in the carrying
of coffee, chickpeas, chicle, India rubber and livestock to ports. The export
sector cornered about three-quarters of the benefits enjoyed by the railways
in Mexico by 1910. Backward spillovers had little impact on the development
of Mexican industry. Imported inputs as a percentage of operating costs were
42.67 per cent in 1891, fell to 29.33 per cent in 1896 and rose to 47.99 per cent
in 1900 (Coatsworth 1979: 955). The railways were constructed and
operated with rails, locomotives, rolling stock, spare parts, iron bridges and
supervisory/engineering personnel imported from abroad. On occasion
even fuel (coal and wood), ties for laying tracks and unskilled labour were
imported. In Mexico ‘railroads contributed far more to economic growth
[in the US]—did so precisely by mortgaging the country’s future to an increasing
dependence on the North Atlantic economies’ (Coatsworth 1979: 956).
There was no domestic political effort to create domestic spillovers and, as
in colonial India, government policy was skewed towards supporting export-
oriented agriculture. Freight rates on the Mexican railways were lower for
agricultural products than other exports throughout the late nineteenth
century. This was a policy supported by both the railroad companies and the
Mexican government, designed to stimulate the growth of an export-oriented
agricultural economy in the uninhabited mountains through which railway
lines passed. The government hoped to further benefit by mobilising tax
revenue from taxes on exports (Coatsworth 1979).
64
The story in nineteenth-century Germany was very different. Due to

technological backwardness and lack of industrial capacity, when German
railroad construction began around 1835, the German engineering and iron
industries were not capable of producing the key inputs for railroad, rails and
locomotives. German iron production was not developed much beyond the
technological level of the sixteenth century. In 1835 more than 90 per cent of pig
iron was produced in small furnaces using charcoal and more than 80 per cent
of bar iron was still produced by medieval methods. During the early years of
railroad construction up to the 1840s, foreign imports of rails and other inputs
dominated the market. The government responded by initiating a process of
import substitution. The policies used by the German government are discussed
in more detail in Chapter 7. Many iron processing plants using modern British
technology were established and existing ones enlarged their capacities. By the
1850s most rails were produced in Germany. In the 1840s the production of
rails depended on imported coke pig iron. Overall pig iron production increased
from 148,000 metric tonnes in 1851 to 804,000 in 1866. Backward spillovers
from railway construction were crucial in driving this growth. The demand for
rails and rail fastenings from 1860 to 1871 absorbed 28.1 per cent of the pig iron
production and 23.8 per cent of the pig iron consumption in Germany between
1857 and 1868. After 1854 all locomotives except a few from Austria were
supplied by German producers. By 1875 locomotive construction accounted for
74 per cent of the output of the entire engineering sector (Fremdling 1977: 588).
The studies discussed so far in this section have explored how infrastructure
generated spillovers in terms of economic sector, boosting either agriculture,
industry, exports or imports. Spillovers can also operate over space. A study of
transport infrastructure in contemporary India by G. Khanna (2016) explored
the impact of road and rail routes connecting the four largest Indian cities.
He finds that regions neighbouring those in which the infrastructure was
built were also affected by better transport connections because of spillovers
across regions. Ignoring these external impacts would lead to estimates of the
beneficial effects on income that were only one-quarter of the true overall
impact of these routes.
Two lessons emerge from these stories: First, that big infrastructure
may simply boost existing areas of production (comparative advantage).
In developing countries this may mean the expansion of agriculture rather
than modern industry. Second, the policy context is important: are the
domestic government (or influential foreign governments) actively promoting
the spread of spillovers to domestic industry?
65
LABOUR AND EMPLOYMENT

Labour is required for infrastructure construction work. It is the case that
Chinese loans often stipulate the use of a Chinese contractor which reinforces
the tendency to subcontract to members of familiar networks at home.
In Pakistan and also across infrastructure projects in Africa, China has also
often been accused of importing its own, cheaper, skilled and unskilled labour
from China. There is concern that the widely documented failures of the
education system in Pakistan mean that the country lacks the skilled labour
necessary to take up jobs in construction-related CPEC projects, regardless of
Chinese intentions. Even prior to the CPEC, the majority of stakeholders in
Pakistan’s construction sector agreed that local contractors did not have the
capacity to expand and that they already had too much work. The construction
sector has long been small by international comparison. In 2005 the share
of the construction sector in Pakistan’s GDP was only 2 per cent, compared
to 4–8 per cent in India, China, UAE, Indonesia and Bangladesh (World
Bank 2007b: 66). The planned infrastructure investment by the government
in the mid-2000s (in roads, water, irrigation and earthquake reconstruction)
required an estimated 7,700 engineers, 62,000 skilled workers and about 1,000
administrative support staff for project activities. There was a widespread
agreement among the majority of stakeholders surveyed that Pakistan lacked
such skilled human resources. Stakeholders argued that contractors lacked
professional management and personnel with the necessary execution and
implementation skills (World Bank 2007b: 52). These infrastructure projects
were being planned even as the number of engineers being trained in Pakistan
showed a steep downward trend. Pakistan was training approximately
900 civil engineers each year between 1999 and 2004, and the numbers
dropped to about 560 engineers during 2005. There was a similar trend in other
engineering disciplines. The limited pool of engineers was also being drained
away by demand from electronics, computer science and the service sector
more generally (World Bank 2007b: 53). A significant number of educated
engineers were also migrating out of Pakistan each year. Professional salaries
are lower in Pakistan than those offered by regional competitors. Salaries are
twice as high in Malaysia, three times in Indonesia and Thailand and about
seven times higher in the UAE. This makes it hard for Pakistan to retain
its engineers, construction workers and consultants (World Bank 2007b).
In 2005 around 950 engineers went abroad for employment, which was
almost 38 per cent of all graduating engineers. In the six months to June 2007,
66
975 engineers were reported to have migrated overseas. Even before the
CPEC, there was an estimated shortage of between 3,000 and 5,000 civil
engineers. Broadening the perspective of migration shows that in the early
2000s, each year on average 60,000 trained workers, foremen, surveyors,
quantity surveyors, technicians and others found jobs overseas and the trend
was increasing over time (World Bank 2007b).
Concerns about unskilled labour seem more misplaced. In survey evidence
from six large CPEC and related road infrastructure projects, Zia and Waqar
(2018) found that 52,000 direct jobs had been created with an average ratio
of Pakistani to Chinese employment of 18:1. These included the $2.94 billion
Multan to Sukkur portion of the Peshawar to Karachi highway (13,881
Pakistani and 1,293 Chinese people employed), the $865 million Lahore to
Multan portion (16,676 Pakistani and 570 Chinese) and the Faisalabad to
Multan highway (3,543 Pakistani and 97 Chinese). The ratios are optimistic
but the total of 50,000 temporary construction jobs unearthed by the survey
is small in a country of more than 200 million people. By comparison, in 2013
the textile sector in Bangladesh employed 4 million people, mainly young
women. The Pakistan government remains optimistic and has proclaimed
that eventually 800,000 jobs will be created in the SEZs. These figures lack
any real credibility and ‘the use of rather crude and unscientific methods for
projecting this number has created even more doubts’ (Daily Times 2019).
A blazing headline on 8 January 2019 proclaimed that the CPEC would create
700,000 new jobs. On closer inspection, the article noted that only 75,000 jobs
had so far been created by the CPEC and referenced this to ‘Chinese Embassy
Documents’ (Express Tribune 2019a). On 5 February 2020, CPEC Authority
Chairman Lt-Gen (retd) Asim Bajwa claimed that the CPEC would generate
‘massive employment opportunities’. He could only point to 5,000 jobs already
created in the energy generating projects and another 15,000 that would
be created by further projects under construction (Business Standard 2020).
The ‘potential’ of the CPEC to create jobs in Pakistan is asserted in numerous
academic studies without any effort to survey existing projects or undertake a
rigorous effort to forecast the likely impact of the CPEC on industry, exports
or agriculture.
Even if the pessimism and optimism both seem misplaced, it is clear
that the Government of Pakistan could be doing more to actively boost the
employment-generating potential of the CPEC. Turkmenistan, for example,
requires that 70 per cent of project workforce consist of local employees and
Uzbekistan requires China to contribute only managers and not labourers.
67
BUILDING MATERIALS
The media have on occasion made much of how the CPEC’s construction has
created demand for locally sourced building materials. The construction of the
Sukkur–Multan highway (M-5) (a CPEC project) involved
[a] total of 100 million cubic meters of earth, 30 million cubic meters of stone,
500,000 tons of bitumen, 60 million bricks, 6 million tiles, 1 million tons of
cement, 500,000 tons of steel, more than 9,200 sets of machines and tools, 500
million liters of oil, 16.11 million square meters of turf, 360,000 seedlings and
a large number of daily supplies have been purchased from the local market.
(The News 2020)
Beyond such claims we lack rigorous studies of whether the CPEC has
stimulated spillovers to local supplies of building materials.
Prior to the launch of the CPEC, Pakistan was producing about 23 million
tonnes (MT) of cement per annum, of which about 10 per cent was exported.
From the remaining, about 60 per cent was consumed by the housing sector,
20 per cent by the industrial sector and the rest was available for public sector
construction projects (World Bank 2007b). There is currently much scope
for increased consumption of cement in Pakistan. In 2018 the per capita
consumption of cement in Pakistan was 140 kilograms, lower than China and
India, and significantly less than the global average of 400 kilograms (Global
Village Space 2018). This will increase. According to the 2017 population
census of Pakistan, the population of Islamabad has grown by 149 per cent
since 1998, with an annual growth rate of 5 per cent. Urbanisation will only
continue. The Pakistan Planning Commission projects that by 2030 the
country will have over 100 million urban citizens, representing more than
half of the population. This will generate an enormous demand for cement to
construct housing. There is also a need to significantly upgrade the existing
housing stock. Of the 19 million housing units in contemporary Pakistan,
only 55 per cent are estimated to be ‘pucca’ houses that were designed to be
permanent dwellings (Global Village Space 2018). These demands were earlier
supplemented by government efforts to boost infrastructure investment in the
MTDF. This effort was expected to raise domestic demand for cement for
major infrastructure projects from 3.3 MT in 2005 to 6.7 MT in 2010. While
consumption headed ever upwards, the World Bank and others were sceptical
about the ability of Pakistan to expand domestic production. The World Bank
68
anticipated that domestic cement capacity would rise to 42 MT installed

capacity by 2009 but did expect significant shortages in Pakistan. This would
be relieved, they argued, by the increasing imports of cement, to around
2–3 MT per annum (World Bank 2007b). This perception only hardened
with the launch of the CPEC. According to some estimates, at least 3 MT
of demand will be generated through the CPEC to build roads, highways,
ports, dams and bridges. In the 12 months to February 2018, local cement
consumption increased by more than 9 per cent (Baig 2018).
The prospects for greater production in Pakistan looked bleak relative to
the excess capacity and presumably export readiness of much of the rest of
the world. In early 2018 the Bangladesh local cement industry had a capacity
utilisation of 54 per cent. Cement consumption was 27.1 MT in 2017 against an
annual production capacity of 50.2 MT from around 45 companies of various
sizes. In Algeria the cement market has been characterised with overcapacity
since 2017, with new capacity still being added and excess capacity expected
to reach 15 MT. In Uganda three new cement plants or upgrades to existing
plants were opening in 2018, which together were expected to dwarf local
demand. Overcapacity in cement production has been evident in Indonesia
since 2016, with capacity doubling between 2012 and 2015, leaving capacity
utilisation only at around 65 per cent and excess capacity of 27.7 MT (Global
Cement 2018b). In 2018 the capacity in India was only around 65 per cent
(Financial Express 2018).
Dwarfing all of these was the dragon in the world market, China. In 2015
China’s cement production accounted for 57 per cent of global output and
was about nine times greater than the second-largest producer, India. Rapid
growth of cities had generated an urbanisation rate of 55 per cent in China by
2015. The expansion had created an enormous demand for cement to construct
housing and urban infrastructure. Even so, this expansion of capacity was well
beyond domestic demand and left the Chinese cement industry suffering the
consequence of enormous excess capacity. China’s cement capacity in 2014
was 3.1 billion tonnes per year and total production was almost 2.25 billion
tonnes, resulting in a utilisation rate of 73 per cent. At the end of 2009, the
Chinese government issued guidelines aimed at curbing overcapacity by
ordering that any new capacity should be balanced by cuts in existing capacity.
Three hundred and sixty MT of obsolete capacity was scrapped between
2009 and 2014 but new projects continued to be built. These measures only
69
slowed down the rate at which the problem was getting worse. Between
2009 and 2014 the utilisation rate fell from 76 to 73 per cent and the scale of
overcapacity increased from 450 to 850 MT (European Union Chamber of
Commerce 2016).
Despite World Bank pessimism and threats of cheap imports flooding
in from the world market, the CPEC does seem to have started generating
spillovers that are leading to the expansion of the domestic cement industry.
This pessimism had largely ignored the long history of successful state sector
efforts to promote cement production in Pakistan. The cement industry was
severely underdeveloped at independence. While the first cement factory had
been established in what became West Pakistan in 1921, by 1947 the total
installed capacity was only 470,000 tonnes per annum and production was
300,000 tonnes per annum. By 1954 there were severe shortages, with demand
of 1 MT outstripping capacity of 600,000 tonnes per annum. The 1960s saw
successful state-developmental interventions in cement. The state-owned
Pakistan Industrial Development Corporation (PIDC) took the initiative of
expanding production in the country. Two cement factories were established at
Zealpak and Maple Leaf with installed capacity of 240,000 and 100,000 tonnes
respectively (Global Village Space 2018). The cement industry has continued
to respond in recent years in a way that has confounded pessimism. Output
tends to fluctuate dramatically but reached a 25-year peak in 2019. Instead of
domestic shortages pulling in imports, domestic production expanded—this
time led by the private sector as opposed to earlier state-led expansion. Here
were CPEC spillovers in the making. In the 12 months to February 2018,
cement exports had increased by more than 18 per cent (Baig 2018). Pakistan
even started exporting to India. India removed customs duty on cement imports
from Pakistan in 2007. Cement imports from Pakistan to India increased
24 per cent between December 2017 and March 2018 (Business Standard
2018). Data from the India Directorate General of Foreign Trade showed that
1.27 MT, or 76 per cent of total cement imports, were imported from Pakistan in
the 12 months to March 2018 (Global Cement 2018a). The angst had migrated
from Pakistan to India. In India the president of the Cement Manufacturers’
Association complained about imports of cement from Pakistan damaging
the local industry (Global Cement 2018a). He claimed that production costs
in Pakistan were 10–15 per cent cheaper than those in India (Tribune India
2018). In February 2019 exports had risen almost 70 per cent over the
previous 12 months to 0.51 MT (Global Cement 2019). Here the Chinese
70
had been more prescient than the pessimistic World Bank and had declared in
their own planning documents,
The Pakistani cement industry enjoys rapid growth. Driven by strong

domestic demand for infrastructure construction and from the export market,
the industry’s advantages can be increased gradually. The Plan suggests the
existing cement capacity be maintained, technical transformation be reinforced,
and outdated equipment be replaced by environment-friendly processing
equipment with high energy efficiency. (Government of China 2015b: 14)
It is not all good news for Pakistan as the cement industry remains concerned
that it has lost markets in Afghanistan to Iran and is vulnerable to direct import
competition from Iran, where energy costs are lower. High government taxes
in Pakistan mean domestic producers can be uncompetitive compared with
imports of under-invoiced cement imports from Iran (Global Cement 2018a).
The combination of domestic sales and exports and lack of imports led
to capacity utilisation surging to levels much higher than the other countries
listed earlier. Capacity utilisation in Pakistan reached more than 91 per cent
in 2018. These high rates of existing capacity utilisation encouraged domestic
cement manufacturers to spend an estimated $2.25 billion on new production
capacity. Maple Leaf Cement Factory told the Pakistan Stock Exchange
(PSE) they had placed an order with the Danish firm FLSmidth for the
establishment of a new cement production line with a daily capacity of 7,300
tonnes. Bestway Cement, a subsidiary of Bestway Group UK and the largest
cement manufacturer in Pakistan with around 17 per cent of the market,
informed the PSE in March 2017 that it would set up a brownfield cement
plant with a capacity of 6,000 clinker at its Farooqia site in northern Pakistan
(Global Village Space 2018). The Flying Cement Company had ordered a
vertical roller mill from Germany’s Loesche for a new 7,000-tonnes-a-day
production line in Lahore (Global Cement 2018a). Kohat Cement Company
Ltd ordered four vertical roller mills (VRM) from Germany’s Loesche in
order to expand its cement plant in Kohat, 160 kilometres west of Islamabad.
When the new line enters production, the plant’s capacity will nearly double
to 5 MT a year (Global Cement 2018a). In the 12 months to February 2018,
local production had increased from 26.3 MT to 30.1 MT, representing a
growth of more than 14 per cent (Baig 2018).
It was not merely about production. The cement industry in Pakistan
also saw sharply rising profit rates. DG Khan’s sales rose by 30 per cent to
71
$271 million in the year that ended in June 2018 and its profits from $63.6
to $72.4 million. Maple Leaf Cement saw profits falling due to higher costs
but sales rising from $194 to $208 million in the year to June 2018 (Global
Cement 2018a). The spokesman of the All Pakistan Cement Manufacturers
Association (APCMA) said that the cement industry was among the highest
contributors to the national exchequer; its contribution increased from
Rs 39 billion in 2012–13 to Rs 117 billion in 2016–17 (Baig 2018).
Concerns remain about the potential for CPEC-induced production
spillovers in other sectors. Many state enterprises in China are not subject to
market-based disciplines such as profit maximisation and cost minimisation
but rather are tasked to fulfil broader political goals. These goals include
increasing market share, expanding local employment and developing
production capabilities such as adopting new technology. Local governments
have long tolerated non-performing loans and low profitability, and have
pushed state-owned banks to continually extend new credit and provide cheap
inputs such as land and energy in pursuit of these goals. While the majority
of the locally generated taxes must be passed on to Beijing, local governments
can keep business tax revenues and so have an incentive to encourage local
investment and production to maximise output and thereby boost local tax
revenues. In response to the 2008 global financial crisis, the central Chinese
government instituted a large fiscal stimulus, including a large lending
programme, that generated a new round of massive investment in low-return
manufacturing projects. The career and promotion prospects of party and
government officials are closely influenced by their ability to promote local
industrial production (European Union Chamber of Commerce 2016).
By the mid-2000s, the World Bank was predicting similar problems
with steel (as it did with cement), another vital input into construction
activity. The total consumption in Pakistan was estimated at 4.7 MT during
2004–5. This comprised of Pakistan Steel Mills, then producing about
1 MT of quality steel, and other local re-rolling and smelting mills (which use
imported scrap) producing about 2.3 MT of steel. The deficit of 1.4 MT was
met through imports. The World Bank predicted that increasing infrastructure
investment (planned under the MTDP) would raise import requirements
from 2 MT during 2006–7 to 3 MT during 2009–10. The main concern for
the construction industry is with respect to the production of quality rolled
billets. These are used to produce reinforcement bars that are commonly used
in construction. In 2004–5 Pakistan Steel Mills produced about 0.3 MT of
72
steel billets while the demand for long products was about 2.5 MT (World
Bank 2007b). The steel sector in Pakistan has failed to adjust. In June 2020
it was confirmed that Pakistan Steel Mills had fired 9,350 workers as well
as run up wage and pension payment arrears since 2016. It was described by
some as a loss-making liability (Gulf News 2020). During 2020 there have
been media reports that Pakistan Steel Mills are seeking assistance from
the Bank of China to fund a rehabilitation package (Dawn 2020), though
it denies that it will be therefore considered part of the CPEC and open to
Chinese investors (Express Tribune 2020a). Elsewhere in the steel sector, the
Pakistan Association of Large Steel Producers (PALSP) was warning that
further openness to trade in steel billets, bars, wire rod and structures would
lead to mass closure of factories and job losses (Express Tribune 2020b).
Over the horizon was China, where the steel industry had evolved
according to very different drivers. By 2015 the capacity in China’s steel
sector had reached over 2,300 MT. From 2000 to 2015, nominal steelmaking
capacity grew by an average of 82 MT per year, an annual increase that
equalled total US production. China’s overcapacity grew from almost zero
in 2000 to 336 MT in 2015. Even then, China planned to add another
41 MT of steelmaking capacity by 2017 (Brun 2016). In part, this growth
was caused by strong domestic demand from infrastructure construction, real
estate, machinery and the automobile industry (European Union Chamber
of Commerce 2016). Much of this expansion also reflected excess capacity.
Capacity utilisation ratios declined from 95 per cent in 2002 to around 70 per
cent in 2015 (Brun 2016). The incentives to promote local industrialisation
already noted mean that it is often common practice for local officials to
provide implicit lending guarantees to steel firms to attract investment
without consideration for existing overcapacity. The Chinese steel industry in
the mid-2010s had $480 billion in outstanding loans (Brun 2016). China has
made efforts to export steel, often at very low prices. Steel exports from China
to the world quadrupled between 2005 and 2016 to 111.6 MT, doubling in the
four years alone after 2012 (Brun 2016). During this period India, Malaysia,
Indonesia, South Africa and Thailand responded with attempts to protect their
markets from Chinese exports (European Union Chamber of Commerce 2016).
The pressures towards increasing output and capacity thwarted continued
attempts by the central government to curb the expansion of the industry.
In 2009, when China’s planning agency ordered several blast furnaces to
be closed, Hebei Tianzhu Iron and Steel Group received a $750,000 bonus
73
for dismantling four blast furnaces. The compensation was then promptly
reinvested to build a larger blast furnace and expand production capacity
(Brun 2016: 39). In October 2013 the Chinese government issued a guideline
requiring that steelmaking capacity in China should be reduced by 80 MT
by 2018 or about 10 per cent of China’s 2013 production. Instead, capacity
continued to grow (Brun 2016; European Union Chamber of Commerce
2016). The litany of failed efforts to cut back capacity include the Steel
Industry Revitalisation Plan, 2009, the State Council Circular, 2010, the 12th
Five Year Plan, 2011, and the State Council Guidance on Excess Capacity,
October 2013 (United Steel Workers 2017). It is clear that a small but growing
fraction of this excess production is leaking into the Pakistani market. Imports
of iron and steel have been steadily increasing, from $324 million in 2013 to
$1.1 billion in 2017 (Pakistan Business Council 2019: 16).
As a result of Chinese urbanisation and the resulting need for construction
materials, there has been increasing demand for construction glass.
The capacity in the flat glass sector in 2007 was 650 million weight cases
per annum with production of 574 million weight cases per annum, giving a
capacity utilisation rate of 88 per cent. In 2009 this translated into half of the
world’s production of flat glass. The 2008 stimulus package produced a boom in
market demand with sharply rising prices and the addition of new production
lines. By 2014 the total annual production and capacity expanded to 831 and
1,046 million weight cases respectively, giving a capacity utilisation rate of
79 per cent. China’s flat glass industry is highly fragmented. The industry
began to experience significant problems in 2012, as overcapacity led to
falling prices and companies started experiencing financial difficulties. Again
the Chinese government attempted to reduce overcapacity. The industry
responded by consolidating, through mergers and acquisitions. Kibin merged
with Zhejiang Glass, and the Triumph group acquired a large number of small
producers. There were also some efforts to upgrade production. Triumph
increased its capacity in the higher value-add indium tin oxide (ITO) and
ultra-thin segments. There were also government efforts to shift production
to Western China. Overall, these efforts had little impact and provincial
governments continued to allow new capacity to be built (European Union
Chamber of Commerce 2016).
Another important building material (for roads) is bitumen; again the World
Bank was predicting imminent shortages in Pakistan (and hence imports). They
argued that the local production of asphalt/bitumen was expected to remain at
74
around 300,000 MT while the total amount required was estimated at nearly
400,000 MT during 2005 and was expected to increase to 438,000 MT by
2010. By the mid-2000s a shortage of bitumen meant that several ongoing road
projects were suffering delays (World Bank 2007b). Some firms have made
much of the CPEC boosting demand for their output of bitumen. Spectre
Energy has advertised the link between themselves supplying bitumen (18,000
MT) in 2019 to the China Railway First Group Co. Ltd and the China State
Construction Engineering Corporation (CSCEC) for use in CPEC projects
(Spectre Energy 2019). At an aggregate level, there have been no obvious
CPEC-induced spillovers to boost domestic production, which has actually
declined over the last decade from a peak of 460,000 MT in 2005 to a low of
155,000 MT in 2012 and witnessed some revival to 223,000 MT in 2016, but
further decline in 2017. Outside the construction sector but of relevance for
Pakistan more widely were other industrial sectors in which China experienced
massive capacity growth and resulting overcapacity. These included China’s
electrolytic industry, where capacity utilisation rates declined to 78 per cent in
2015, and paper and paperboard, where utilisation rates fell to 84 per cent in
2014. In the petroleum industry, for the chemical by-products (fertiliser, urea,
methanol, chlor alkali, soda ash, calcium carbide, tyres, hydrogen fluoride,
ammonium phosphate and silicone methyl monomer), capacity
utilisation rates were all below 80 per cent (European Union Chamber of
Commerce 2016).
The Chinese government is very clear that the BRI ‘is designed to uphold
the global free trade regime and the open world economy’ (Government of
China 2015a: 1). The BRI ‘follows market operation. It will abide by market
rules and international norms, give play to the decisive role of the market in
resource allocation and the primary role of enterprises, and let the government
perform their due functions’ (Government of China 2015a: 2a). In order to
boost investment and trade, ‘[w]e should strive to improve investment and
trade facilitation, and remove investment and trade barriers for the creation
of sound business environment within the region and in all related countries’
(Government of China 2015a: 3). The problem of excess capacity in Chinese
manufacturing, its existence and implications have not yet been discussed
in the predominantly Pakistan-centric writing on the CPEC. What will be
the cost to Pakistan of further deepening free trade with a country that has
massive overcapacity and is ready to export that output at extremely low cost
to Pakistan?
75
There is an alternative and more optimistic (for Pakistan) implication of

the BRI. It is clear from initial pronouncements that China sees the BRI
as a way to reduce domestic excess industrial capacity, including in the steel
sector, by transferring it overseas. The new China-led Asian Infrastructure
Investment Bank (AIIB) and $40 billion Silk Road Fund provide funds
to help implement this policy. He Yafei is the former vice minister of the
Overseas Chinese Affairs Office of the State Council and former vice minister
at the Chinese Ministry of Foreign Affairs. He argued in an article (published
in the South China Morning Post in 2014) that the agreed solution to the
problem of industrial overcapacity in China was to implement the ‘Going
Out Strategy’ for Chinese enterprises and export the overcapacity. He argued
that this would be a win–win situation, moving production from China with
excess capacity to those countries with inadequate capacity (Yafei 2014).
Chinese banks have been instructed to assist through credit, trade finance and
international insurance to support the Going Out Strategy and in particular
overseas mergers and acquisitions (M&As). Hebei Iron and Steel Group, one
of the largest steelmakers in China, signed an agreement to develop 5 MT of
steelmaking capacity in South Africa by 2019 (European Union Chamber of
Commerce 2016). It is not clear whether rising exports of steel from China
to Pakistan or whether rising domestic production of cement in response to
CPEC demands will be characteristic of CPEC spillovers in Pakistan. Will
the spillovers generate local industrialisation or leak out of Pakistan into
Chinese imports?
One lesson about infrastructure that is lost in these short-term predictions
from the World Bank about shortages and inadequate domestic production
capacity is that big infrastructure is very different from marginal changes.
Marginal infrastructure changes involve incremental gains such as alleviating
bottlenecks, cutting costs for producers, and reducing travel time for
commuters. The benefit from such changes should be quickly and clearly
evident. The impact of a big change in infrastructure cannot be so easily
measured or even anticipated. We are no longer considering just promoting
a more efficient economy but of ‘calling forth and enlisting for development
purposes resources and abilities that are hidden, scattered, or badly utilised’
(Hirschman 1958: 5). Economic growth then is not about the efficient
allocation of existing resources but if the ‘economy is to be kept moving ahead,
the task of development policy is to maintain tensions, disproportions, and
disequilibria’ (Hirschman 1958: 66). In considering the success (or otherwise)
76
of the CPEC, we should not expect to see only ‘improvements’ but also
considerable and often unexpected changes that may in turn be associated
with significant adjustment costs such as shortages of construction inputs or
of credit among firms desperate to expand, bankruptcy of firms no longer able
to compete, forced migration as workers move to growing areas, and so on.
IN SUMMARY
Scholars of the CPEC have made much of the dramatic reduction in distance
for firms in China to Europe to herald the fact that the CPEC will make
markets more efficient. But questions about efficiency have tended to stop there.
Scholars have rarely followed up this question with the question: efficient to
do what? Will the CPEC evolve into a transit corridor, meaning that Pakistan
will be watching Chinese exports to the rest of the world whizz past from
China to Gwadar? Will the CPEC lead to a surge in exports from China that
undermines Pakistani industry? Will the CPEC stimulate industrialisation
and economic growth within Pakistan? Historical case studies tell us that
big infrastructure has induced all three outcomes. This chapter started to
answer this question. There are well-documented impacts in Pakistan from
public investment in infrastructure crowding in private investment. The
volume of public investment resulting from the CPEC will be insufficient
to raise private investment to the 30 per cent or more of GDP that has been
consistent with sustaining 7–8 per cent or more economic growth in other
countries. There are also concerns about the quality of public investment in
Pakistan. This chapter also looked at production spillovers and found that
history offers us widely varied cases. In Mexico, India and Brazil, for example,
the construction of railway spillovers leaked out of the domestic economy
to promote industrialisation overseas; in Germany those spillovers stimulated
domestic industrialisation. There are some grounds for optimism in regard to
domestic cement production and possibly with employment where the CPEC
does appear to be generating domestic spillovers. Concerns remain though
regarding iron and steel, glass and bitumen. Chapter 4 now asks how we can
judge the success or otherwise of the CPEC.
77
4
Through the Eyes of Who?

Evaluating the Success of the CPEC
In the existing CPEC literature, there is little sign of careful thinking about
how to measure the success or otherwise of the CPEC using any rigorous
method. More common is to use a list of descriptive statistics about the
promise of the CPEC and combine it with a mix of laudatory claims about
the transformative potential of the CPEC.
The official rhetoric about the CPEC from the Government of Pakistan
is that it will transform Pakistan’s economy. The former Minister of Planning
and Reform, Professor Ahsan Iqbal, gave a speech to the Pakistan–China
Joint Cooperation Committee in which he claimed that by 2025 the CPEC
will help Pakistan achieve 8 per cent annual growth, increase exports from
$25 to $150 billion, raise the tax-to-GDP ratio to 16–18 per cent, investment
to 22–25 per cent and domestic savings to 18–21 per cent (Ali et al. 2017:
193). Since the CPEC was announced and the early projects were completed,
none of these outcomes has yet come to pass. Between 2012–13 and 2017–18,
growth of GDP has remained stuck in the 3–5 per cent per annum range,
exports have actually declined from 11 per cent to 7.9 per cent of GDP (to
around $22 billion), total investment has remained stagnant at 15–16 per
cent of GDP (down from 20 per cent in 2005–6) and domestic savings have
declined from 14 to 10 per cent of GDP (Government of Pakistan 2019).
Figure 4.1 later in the chapter shows that tax revenue has been rising in the
last few years in Pakistan, but, at around 12 per cent of GDP, remains a long
way below the government aspiration.
Although the CPEC aspirations have not yet been achieved, it is too early
to make any kind of definitive judgement. The CPEC is not even due for
completion until 2030. The method in this chapter is to evaluate the CPEC
against both this speech and also the more formal claims of the Government of
Pakistan. The official government planning document states that ‘The CPEC
78
THROUGH THE EYES OF WHO?
is a growth axis and development belt featuring complementary advantage,

collaboration, mutual benefits and common prosperity’ (Government of Pakistan
2017: 4, emphasis mine). This book will evaluate the CPEC according to
whether the relationship is one of ‘complementary advantage’ with China (the
importance of Western China), whether there will be ‘mutual benefits’ (the
distributional impact within Pakistan) and whether it will lead to ‘common
prosperity’ (if it will have a transformative impact on Pakistan’s economy).
The evidence will draw upon historical and contemporary infrastructure
projects and an analysis of both contemporary Pakistan and China.
‘COMPLEMENTARY ADVANTAGE’:
THE IMPORTANCE OF CHINA
After 1978 China made a break with the Maoist effort dating from 1949 to
promote regional equality across China. Deng Xiaoping put greater emphasis
on economic growth and sanctioned some provinces to get rich before others.
In 1978 economic reform and opening up to trade and FDI were permitted
in a few eastern coastal states (Lu and Deng 2011). The western provinces
retained state controls that in various ways subordinated them to the growth
efforts of the eastern provinces. The western provinces were compelled to
sell their raw materials at state-controlled, low prices and to purchase the
consumer goods being produced by eastern industry at rising market prices
(Christoffersen 1993). By the end of the 1990s, 86 per cent of China’s GDP
and 95 per cent of FDI was located outside the western provinces. Domestic
political pressures were growing to rebalance China’s economy.
The rebalancing was initiated by President Jiang Zemin who in 1999
launched the Western Development Programme (WDP) or Open Up the
West campaign to promote economic development in the west. The campaign
had a broad remit that covered economic, ecological and security concerns
(Lai 2002). The WDP was a state-led effort to pump state investment, skilled
labour, foreign loans and private capital into Western China The initiative
covered six provinces (Yunnan, Gansu, Sichuan, Guizhou, Qinghai and
Shaanxi), three autonomous regions (Ningxia, Xizang [Tibet] and Xinjiang)
and one provincial-level municipality (Chongqing). The Chinese government
have been clear in the 12th Five-Year Plan (2011–15) and other documents
that its China-wide efforts aim to
79
restructure and move to higher value-added areas of economic activity; for

the inland regions to attract domestic and international companies to move
operations inland, including the establishment of trade processing base (in
other words becoming part of transnational production networks in the way
that coastal regions did in earlier decades). (Summers 2016: 1632)
The Chinese government made conscious comparisons between the new

Chinese west and the development of the American west in the early twentieth
century (Moneyhon 2003). Specific policy initiatives included government
investment in infrastructure, budgetary support for provincial development
plans, reduced taxes on income, and tariff exemptions on imported industrial
equipment. The construction of the Qinghai–Tibet Railway was made an
almost tax-free project (Lu and Deng 2011). Between 2001 and 2010 the
share of loans from the China Development Bank going to Western China
surged to reach a quarter of all lending (Lu and Deng 2011). The WDP
included a long list of big infrastructure projects: the eight major cities of
Western China were connected in a 12,600-kilometre highway building
programme, and more than 200,000 kilometres of local highways and 150,000
kilometres of minor rural roads were newly constructed or upgraded. In 2000
the Chongqing–Huaihua Railway, in 2001 the Qinghai–Tibet Railway and
in 2002 the Xi’an–Hefei section of the Xi’an–Nanjing Railway were built to
more deeply integrate the region into the national railway network (Lai 2002).
Despite its large 1.6 million square-kilometre area, which is one-sixth of
China, Xinjiang has a population of only around 23 million people (Griffiths
2017). This can be explained by geography. Xinjiang ‘is largely a desert
wasteland cleaved in two by the Tianshan Mountains…. [T]he physical
environment of the region is simply too harsh to support large populations (by
Chinese standards), even with modern agricultural and industrial technologies’
(Strafor 2013: 3). Xinjiang is rich in natural resources that include 30 per cent
of all of China’s oil reserves, 34 per cent of gas reserves and 40 per cent of
coal reserves as well as chrome, rare metals, gold, nickel, salts and building
materials (Government of China 2007). Beijing has long worried about
potential instability in Xinjiang and its long distance from the centre of
China. In 1759 Xinjiang was incorporated into the Qing Empire. The ensuing
250 years saw frequent changes in the status of Xinjiang. The collapse of the
Qing dynasty in 1911 saw a progressive weakening of central authority. In the
1930s a separatist Muslim/Turkik rebellion in southwest China established
the First East Turkestan Republic, which was soon after defeated by central
80
Kuomintang troops. Another uprising in 1937 led to Soviet intervention and

rule for a decade. At the close of World War II, the USSR backed the creation
of an independent East Turkestan Republic that included Xinjiang. The idea
was quashed by the return of communist Chinese troops (Griffiths 2017).
The emergence of independent Central Asian states with the collapse of the
USSR in 1991 inspired separatist movements in Xinjiang and elsewhere.
Already in 1990 Chinese soldiers and police had clashed with the Islamic
Party of East Turkestan which advocated closer union with Central Asia
or even secession. Unofficial estimates suggest that dozens were killed
(Moneyhon 2003).
The WDP had a substantial presence in Xinjiang and included a
2,500-kilometre, $1.4 billion pipeline running from gas fields in Xinjiang to
Shanghai as well as more than $8 billion spent on highways, power plants,
dams, telecommunications and railways (Moneyhon 2003; Becquelin 2004).
Building infrastructure in areas rich in natural resources will lead to local
industrialisation, claim the Government of Pakistan in its plans for the
CPEC. The case of Xinjiang is instructive as this effort strengthened the
status of Xinjiang as both a transit corridor and a source of raw materials
for manufacturing on the eastern coast. Rather than fuelling local industries,
the West–East Pipeline transferred energy from the Xinjiang–Kazakhstan
border to cities around the Yangtze and Pearl River deltas (Strafor 2013).
In 2010 71 per cent of exports from China to Central Asia exited through
Xinjiang, but the bulk of these were being manufactured outside of Xinjiang.
The share of Xinjiang in these exports was actually declining over time. Better
infrastructure had boosted the relative competitiveness of coastal China
and they were pushing Xinjiang aside to capture more of the export market
through Central Asia (Xu 2016). This generated growing political concerns
in Xinjiang, that Beijing was extracting oil, coal, aluminium, wool and cotton
from Xinjiang to benefit the coastal region. There was investment but where
were the new local jobs? (Christoffersen 1993).
These province-wide economic concerns interacted with local structures
of minority politics. Many of China’s 55 minority groups live in the west,
especially in Xinjiang. There was a realistic perception that Beijing was
promoting migration to the region to strengthen ties with the centre.
In the 1950s the Han Chinese constituted about 6 per cent of the population
of Xinjiang and by 2000 more than 40 per cent (Becquelin 2004). By 2000
approximately a quarter of the population of Xinjiang remained in poverty
81
and a disproportionate number of these were Uyghurs. What jobs were

created tended to go to migrant Han workers rather than indigenous Uyghurs,
particularly those skilled positions for university graduates (Becquelin 2004:
375). Among the 20,000 oil workers in the Tarim Basin, relatively few jobs
went to minorities. In the Taklamakan Desert oil exploration project, only
253 of 4,000 technical workers came from minority groups (Moneyhon 2003).
The combination of resource extraction from Xinjiang, a lack of local
employment and a perception of discrimination was a classic cocktail for
resentment. The situation was similar to that in Balochistan which provides
much of Pakistan’s domestic gas supply but experiences much lower levels of
development than the rest of Pakistan. There have been at least four major
insurgencies in Balochistan since 1947. In 2009 conflict over perceptions
of inequality and relative deprivation in Xinjiang spilled over into street
demonstrations and riots in Urumqi (Griffiths 2017). There was also an
increase in terrorist attacks in Xinjiang. In March 2013 terrorists attacked
a train station in Kunming, leaving 29 dead and over 130 wounded (Brugier
2014). Kashgar, the starting point of the CPEC, has a population of nearly
4 million people, of which almost 90 per cent are Muslim Uyghur. Terrorist
attacks have been concentrated in the area around Kashgar and the Tarm
Basin in southern Xinjiang. This region is the home of the majority of the
province’s nearly 10 million Uyghur Muslims (Strafor 2013).
To reduce this local political opposition, Beijing has replaced the
infrastructure-oriented policy of the WDP since 2010 with a new strategy
to boost local economic growth by focusing on promoting local industrial
production (Brugier 2014). This new effort is much larger in scale and scope
than the preceding WDP. The region was granted ‘extraordinarily important
strategic status’; Kashgar is now host to a new SEZ along with Alatau
and Khorgas as the crossing points to Central Asia to special trade zones.
Further incentives for foreign investors, including access to cheaper imported
technology, were granted (Griffiths 2017). China clearly sees the CPEC
as bound up with its efforts to promote development in Western China.
The CPEC is intended to ‘further enhance the Western Region Development
Strategy, promote Xinjiang’s economic and social development, speed up the
“Belt and Road” Initiative, give greater play to China’s advantages of capital,
technology and project operating capacity, and form new open economic
system’ (Government of China 2015: 6).
82
A key question for the purpose of this book is to ask whether Pakistan and
Xinjiang are complementary or competitive economies. A ‘complementary’
economic relationship is one in which two countries or regions have a
very different comparative advantage. This could variously be explained by
differences in natural resource endowments, skill levels of the population,
availability of land or histories of learning in production. In the complementary
case, existing and potential patterns of production and exports between two
countries or regions are likely to be very different leading to potentially
large gains from trade. For example, China has been exporting capital
goods and cheap consumer goods to Africa, and in return Africa has been
exporting the raw materials and energy China needs for its manufacturing
sector. Both are likely to gain from this complementary exchange (Chapter
6 discusses this relationship in more detail). The opposite is a ‘competitive’
economic relationship in which two regions or countries have a very similar
pattern of comparative advantage. In this case, existing and potential patterns
of production and exports are similar enough that there exist few potential
gains from trade. For example, China’s export of consumer goods to Africa
may displace local producers leading to the loss of output and employment.
There are both direct and indirect competitive impacts. The direct impacts
include the trade impacts just discussed. Indirect impacts occur as a result
of China’s economic interactions with third countries. China’s demand
for commodities has contributed to their prices rising at a global level and
Ethiopia has benefited from the higher prices it receives for exports of animal
feed (Kaplinsky, McCormick and Morris 2007).
So are Pakistan and Xinjiang complementary or competitive economies?
The Government of Pakistan is in no doubt about this and argues that the
China–Pakistan relationship is a complementary economic relationship and
consequently there are significant potential gains for both countries. China, it
argues, has advantages in ‘infrastructure construction, high-quality production
capacity in equipment manufacturing, iron and steel and cement industries as
well as financing for investment. Pakistan they note is rich human and natural
resources, has huge potential for economic growth, broad market prospects and
a geo-strategic location’. The CPEC should then proceed with an ‘all weather
strategic partnership of cooperation, concepts of harmony, inclusiveness,
mutual benefits and sustainability’ (Government of Pakistan 2017: 11). The
Government of China is likewise very clear: the relationship is complementary,
‘Through major cooperative projects concerning infrastructure construction,
83
industry and people’s livelihood, it [CPEC] seeks to complement each other’s

advantages, promote an efficient division of labor and cooperation, and achieve
mutual benefits, win-win results and common development’ (Government of
China 2015b: 3).
The big picture of China does offer evidence to support the ‘complementary’
perspective. After 40 years of rapid economic growth, China has absorbed
most of its surplus labour, and with wages are rising rapidly, Chinese exporters
have lost competitiveness in labour-intensive industries such as clothing,
footwear and other light manufacturing. These sectors dominated the structure
of Chinese exports in the 1980s and early 1990s. China has responded by
upgrading. In the late 1990s new export sectors emerged, such as office
machinery, telecommunications and furniture. In the 2000s there was further
upgrading into consumer electronics and machinery. China is following a very
traditional pattern that has clear parallels with other economic stories in Asia.
As wages rose with economic growth, a successful Asian exporter has relocated
less competitive labour-intensive industries to other Asian countries.
This process started with industry relocating from Japan to South Korea,
Singapore and Hong Kong in the 1960s and 1970s, to Thailand, Malaysia and
Indonesia in the 1980s, and to China and Vietnam in the 1990s and 2000s.
This pattern is often referred to as the ‘Flying Geese’ model of economic
development, with Japan traditionally as the lead goose. China’s huge labour
force has taken much longer to absorb and it is still continuing in Western
China. The process of international relocation of labour-intensive industry
has begun in China. The first countries to receive Chinese industrial FDI were
low-wage Vietnam, Laos and Cambodia (Hamid and Hayat 2012). While
China as a whole may appear to be a complementary economy to that of
Pakistan, this labelling forgets the specific situation of Western China and the
WDP and Xinjiang in particular.
The petrochemical industry accounts for 60 per cent of industrial value
added in Xinjiang. This comprises four large oil fields, petroleum refining and
spin-off petrochemical-based industries. These spin-offs include chemical
fertiliser, plastics and chemicals such as ethane, sodium hydroxide, soda ash,
sulphuric acid and resin. The three coal bases at Urumqi, Hami and Aiver
Valley offer a similar potential for development (Government of China 2007).
Closer trade links have permitted the increase of such imports (from China
as a whole) to Pakistan. This is an example of a complementary economic
84
relationship, as equivalent production capacity does not exist in Pakistan.

Imports of fertiliser from China to Pakistan increased from $229 million in
2013 to $487 million in 2017 and of organic chemicals from $378 million to
$791 million over the same years (Pakistan Business Council 2019: 16).
The problem for Pakistan is that the economy of Xinjiang is not locked
into a structure determined by its existing patterns of resource-based
comparative advantage. Beijing is making an enormous effort to alter the
trajectory of economic growth in Xinjiang. Rather than rely on the capital-
intensive, low employment potential of the petrochemical sector, the Chinese
government is seeking to promote more labour-intensive sectors in Xinjiang
to boost employment. Particularly since 2010 the Chinese government has
focused efforts on promoting SEZs in Xinjiang. There is a focus on IT,
biotechnology, electronics, bio-medicine and scientific research in the Urumqi
Economic and Technological Development Zone, the Urumqi Hi-Tech
Industrial Development Zone and the Shihezi Economic and Technological
Development Zone. Traditional manufacturing such as cotton textiles,
processing of agricultural products, building materials and processing minerals
are to be found in the SEZs at the Xinjiang Kuytun Economic Development
Zone, the Xinjiang Hoxud Economic Zone, the Xinjiang Huocheng Economic
Development Zone and the Xinjiang Miquan Industrial Park (Government
of China 2007). Outside of these specific zones, there are numerous sectors
targeted for development. These include textiles and garments, processing of
silk, processing coal into chemical products, solar, wind and other renewable
energy, glass products and packing containers, among others (US–China
Business Council 2010; Government of China 2015b: 11).
Since 1994, the centrally mandated fiscal decentralisation has meant that
the local government keeps all locally generated business and income taxes
levied on local enterprises and 25 per cent of the value-added tax (VAT) levied
on enterprises located in its region. The new incentives to build industrial
capacity in Xinjiang have been added to a China-wide tax system that gives
state governments an incentive to promote local industrialisation and protect
industry against competition from other states (Lu and Tao 2009: 168).
Production for local consumption, import substitution and exports generates
tax revenue for the local government. Importing from other states or from
overseas is a revenue loser. The local focus of industrialisation is supported
by empirical evidence. A survey of Nordic firms located in Western China,
85
for example, found that their location decision was based primarily around
the lure of the local market, government incentives, a low-wage and well-
educated local labour force and good infrastructure. None of the companies
surveyed mentioned the possibility of sourcing inputs from or exporting to
Pakistan (Gyllestal and Ekstrom 2013). This is not surprising; in a wider
empirical study Ke (2010) finds that the growth of large cities in Western
China boosts the growth of nearby cities and reduces the growth of nearby
rural counties but the effect only operates to the distance of 100 kilometres.
The localisation incentives reduce the spillover effects of economic growth
in Western China.
Textiles constitute some 8.5 per cent of Pakistan’s GDP, 40 per cent of
its manufacturing labour force and almost 60 per cent of its total exports.
The Government of Pakistan is clear in its hope that the CPEC will boost the
sector further. It proclaims that the CPEC will
[p]romote the quality, value addition, competitiveness and efficiency

improvement of the textile and garment industry, expand the size of the
textile industry, and increase the supply of high value-added products to
adopt the means like export processing to establish a regional cooperation and
development model based on complementary advantages, and mutual benefits.
(Government of Pakistan 2017: 16)
The obvious problem for Pakistan is that the Chinese government is also
set on promoting textiles and garments in Xinjiang (Abbas and Ali 2017b).
The Government of Pakistan hopes that the CPEC will contribute to the
diversification of its domestic industrial structure. It lists chemical and
pharmaceutical, engineering goods, iron and steel, light manufacturing and
home appliances, and construction materials as the sectors due for a CPEC-
boost (Government of Pakistan 2017). The problem again is that these sectors
are remarkably similar to those the Chinese government is promoting in
Xinjiang.
Xinjiang has become one of the most rapidly growing and competitive
textile regions in China. China formulated a 10-year plan for textiles in
Xinjiang in 2014 to establish textile industrial parks and clothing factories.
This was backed by a dedicated $3.2 billion fund as well as subsidies to
electricity, cotton and bank credit (Yarns and Fibres 2017). Starting in 2014,
China’s leading garment and apparel makers including Ruyi Group, HoDo
Group and Huafu Fashion Co. invested in Xinjiang and built factories
86
(Xinhua 2019). In 2014 the Korla Economic and Technological Development

Zone was set to become the largest production base of rayon viscose in China
(Globaloue 2015). In 2015 Henan Xinye Textile Co. opened a 110,000-spindle
plant, the third such factory it had opened in 12 months. Employment in the
three factories was expected to reach 3,000 people within five years (China
Daily 2015). In 2018 the world’s largest textile mills for spinning coloured
yarn was set up in Xinjiang. The mill cost $735 million and was expected to
have 1 million spindles by the end of 2018 (Business Recorder 2018). In 2018 a
plant with an annual output of 100,000 tonnes opened in Xinjiang to produce
nylon from corn, rather than the usual petroleum (Xinhua 2019). An industry
source was quoted as saying that in 2018 a textile factory was set up every
two days in Xinjiang and by 2017 the province hosted more than 1,800 such
factories (China Daily 2017) or 2,700 factories (Xinhua 2018). In 2017 alone,
employment in textiles in Xinjiang rose by an estimated 112,000, and by 2019
400,000 people were working in the industry (Xinhua 2019). Media reports
heralded another near 100,000 textile jobs created in 2019 (China Daily 2020).
The 2014–23 10-year plan aimed to boost this to 1 million by 2023 (Sourcing
Journal 2014). The output in the industry was planned to rise from 30 billion
yuan in 2014 to 400 billion yuan in 2023, by which time Xinjiang will have
the largest cotton textile base in China and the most important clothing
export base in Western China (Macaes 2018a: 101). The textile industry in
Xinjiang is now globally efficient and price competitive after heavy investment
in blowing-carding machinery, combining machines, automatic winders and
shuttle-less looms (Macaes 2018a: 100). The 2019 trade war with the US hit
textile exports from Xinjiang. The US increased tariffs on 87 per cent of textile
and clothing exports from China on 1 September and increased them further
on 15 December. Media reports suggested that exports of textiles, cotton yarn,
clothes and hats from Xinjiang’s Aksu Textile Industrial Park dropped 86 per
cent to $35.87 million in the first 10 months of 2019. Big contracts with
firms such as Disney were abruptly cancelled. Aksu Park showed its dynamic
capabilities by rapidly shifting exports to Russia, Germany, Southeast Asia
and BRI countries (Global Times 2019).
Compared to this large investment in Xinjiang, there is no indication
that Chinese producers are relocating textile production to Pakistan. Total
FDI in Pakistan textiles (including China and everyone else) fell steadily
from only $60 million in 2007 to $15 million in 2017 (Pakistan Business
Council 2018: 13). The textile trade between Pakistan and China is currently
87
at very low levels but the trend is not in Pakistan’s favour. In 2017 Pakistan
had a deficit on garments trade with China; exports of garments from
Pakistan to China increased from zero in 2008 to only $40 million in 2017
while in that same year China exported $65 million of garments to Pakistan
(Pakistan Business Council 2018: 62). This is not surprising. Pakistan is seen
as a means to support the growth of the textiles industry in Xinjiang.
The Chinese government argues that ‘China can make the most of the
Pakistani market in cheap raw materials to develop the textiles and garments
industry and help soak up a surplus labor forces in Kashgar to develop the
city into an industry cluster area integrating textiles, printing and dyeing,
cloth weaving and garment processing’ (Government of China 2015: 100).
There is little reason to suppose this effort will promote industrialisation in
Pakistan. The Chinese government is clear about how Pakistan will contribute:
‘The garment and textile industry should be greatly developed in the Kashgar
(Xinjiang, China) Economic Development Zone through importing raw
materials from Pakistan’ (Government of China 2015b: 14). In 2017 the
Santex Group, a leading Italian textile and apparel producer, organised ‘Future
Textile Road’ to bring together important players to explore the future of the
textile industry. The effort stimulated a dialogue between ‘Xinjiang, China and
Europe’, aiming to provide a platform for international cooperation, linking
partners and promoting textile exchanges (Textile Focus 2017). There was no
indication of any Pakistani participation.
Chinese aspirations run beyond creating a link between Western China
and Pakistan. ‘By 2023, Chinese province Xinjiang is about to become not
only China’s largest cotton textile and apparel market but also that of Central
Asia and Europe’ (Business Recorder 2018) and Xinjiang was to become China’s
‘largest garment export processing base’ (Textile Focus 2017).
Some look instead to links in agriculture as a potential long-term benefit for
Pakistan. The agricultural sector accounts for around 20 per cent of Pakistan’s
GDP and employs over 40 per cent of the country’s labour force. Due to
shortage of arable land and freshwater resources, China increasingly needs to
import land-extensive crops (such as wheat and rice) to feed its population.
Agricultural goods now constitute 9 per cent of China’s total imports (Abbas
and Ali 2017b). Before the mid-2000s the main imported foodstuffs were
soya for animal feed from Latin America. A decade ago, the per capita average
calorie consumption in China had reached 3,040 or 90 per cent of the level in
high-income countries. It has risen even higher since. The future of China’s
88
food imports is also likely to reflect changing patterns of consumption and not
just greater volumes (Kaplinsky, McCormick and Morris 2007). Hence, it is
not surprising that some have argued that
the main thrust of the plan [CPEC] actually lies in agriculture, contrary to the
image of CPEC as a massive industrial and transport undertaking, involving
power plants and highways. The plan acquires its greatest specificity, and lays
out the largest number of projects and plans for their facilitation, in agriculture.
(Dawn 2017)
There is plenty of evidence for an agricultural emphasis in the CPEC, reflected

more in intensity of activities rather than impressive headline financing
numbers. The CPEC plans outline targets to strengthen the entire agricultural
supply chain in Pakistan, starting with the provision of seeds, fertiliser, credit
and pesticides, through production, harvesting, storage and finally marketing
of the final output. Chinese enterprises have been allotted a leading role in
setting up a new capacity in fertiliser, meat and milk production, improving
nationwide logistics in transport of farm products, as well as in processing
vegetables and cotton. Those Chinese businesses entering agriculture will
receive significant assistance from the Chinese government in the form of
grants and low-interest loans. The CPEC plans to harness the scientific and
practical expertise of the Xinjiang Production and Construction Corps in areas
related to irrigation, mechanisation, scientific breeding and the development
of hybrid varieties. The setting up of the Sino-Pakistan Hybrid Rice Research
Centre at Karachi University is hoped to pioneer collaborative research to
produce new high yield seeds in rice and other sectors (Kamal and Malik
2017). Will the CPEC lead to a new Green Revolution in Pakistan?
Various authors have highlighted the potential of agriculture in Pakistan
that they suggest can be liberated by CPEC investments. Ahmad (2017)
argues that the northern areas through which the CPEC transport passes
has advantages in regard to seasons and high altitude and so can produce
higher-value crops. Gilgit-Baltistan has ideal conditions for growing of
fruits such as grapes, apples, peach, cherries, almonds and apricots. In Punjab
existing agricultural products include mangoes, guavas, potatoes and onions,
in the Sindh dates and bananas, in Khyber Pakhtunkhwa (KPK) peaches and
tomatoes, and in the region of Peshawar peaches, citrus, strawberries, apples,
melons and apricots (Ahmad 2017: 98). Abbas and Ali (2017a) note that the
Rashakai Economic Zone in KPK is located in an area that produces wheat,
89
maize, tobacco, rice and sugar beets and has natural resources such as oil, gas,
mines of marble, gems stones, emeralds, uranium, electricity, salt, lithium and
steel. There is an assumption in this literature that agricultural endowments
once liberated by transport access will somehow automatically turn into
agricultural production or local agricultural processing industries. We discuss
this tenuous assumption in more detail in Chapter 7. The second assumption,
as discussed already in this chapter, is that this naïve optimism ignores the
very similar natural endowments next door in Xinjiang. Yet again, the list of
agricultural sectors with potential in Pakistan is uncomfortably close to the list
of those sectors being promoted in Xinjiang.
Despite its obvious problems with inhospitable terrain, agriculture in
Xinjiang does enjoy the benefits of long sunshine, high accumulated air
temperature and a frost-free period that is conducive to growth of crops and
fertile soils (Government of China 2007). Some products are local specialities
that are not directly in competition with producers in Pakistan. These include
safflower, medlar, Hami melon, Turpan grapes, Korla pears and Hotan
pomegranates. The wider competitive threat is evident. Similar to Pakistan, the
current leading agricultural products in Xinjiang are cotton, fruit, grain, beet,
horticulture and livestock husbandry. By the early 2000s Xinjiang accounted
for more than one-third of China’s total cotton crop and 8 per cent of the
world’s total. By 2018 Xinjiang accounted for 75 per cent of China’s cotton
crop. Xinjiang offers China’s second largest area of pastureland, which is ideal
for sheep farming and hence fine-wool production (HKTDC 2018a, 2020).
Xinjiang was also the biggest producer of tomatoes in China and the third in
the world (Government of China 2007). It is not surprising then that cotton
exports, representing 60 per cent of Pakistani exports to China, have been
falling steadily, from $2.65 billion in 2013 to $1.51 billion in 2017 (Pakistan
Business Council 2019: 13).
Recent trade data offers striking evidence of the real success of efforts to
promote the economic development of Xinjiang. The 23 million people of
Xinjiang are rapidly catching up with the exports of the 200 million people
of Pakistan. Total exports from Xinjiang in 2016 reached $15.61 billion and
imports only $2.05 billion. In 2019 exports and imports reached $18.04
billion and $5.67 billion respectively. In 2016 the major exports from Xinjiang
included garments, shoes and textiles and predominantly headed to markets
in Kazakhstan, Kyrgyzstan and Tajikistan. These three countries together
accounted for 68 per cent of Xinjiang’s total exports in 2016 and 65 per cent
90
in 2018 (HKTDC 2018a, 2020). Further improvement of CPEC trade and

transport links will enable exporters in Xinjiang to turn more attention to the
large regional market of Pakistan. This chapter has demonstrated that this is
likely to generate competitive dangers for Pakistani agriculture and industry.
The grand Chinese strategy of the BRI is explicitly outward looking and
has the ultimate goal of creating a new Eurasian supercontinent. The vision
of Eurasia comprises a web of ports, roads, railways and Chinese-dominated
industrial parks that economically unifies China, Central Asia, the Middle
East and Europe. China plans to export to this vast Eurasian market, to
import goods that can no longer be profitably produced within China, and for
Chinese firms who have relocated production outside of China to export across
Eurasia. One of the pioneers of these BRI dyanmics was the Chinese province
of Hebei which announced plans to relocate capacity in order to produce
20 million tonnes of steel, 5 million tonnes of cement and 3 million units of
glass overseas by 2023 (Macaes 2018a). The announcement of the plans and
associated policy incentives have boosted outward FDI from China (if not
yet to Pakistan). Both state- and non-state-owned companies have increased
FDI in infrastructure and industry across Eurasia (Du and Zhang 2018).
The Chinese government has been vocal in proclaiming the generous and
outward-looking nature of the BRI: ‘The connectivity projects of the initiative
will help align and coordinate the development strategies of the countries along
the Belt and Road, tap market potential in this region, promote investment
and consumption, create demands and job opportunities, enhance people-to-
people and cultural exchanges, and mutual learning’ (Government of China
2015a: 2). The mutuality is evident in guidelines issued to Chinese companies
by China’s State Council in July 2018. Chinese business was pushed to ‘look
to Belt and Road countries as new sources of imports, strengthen strategic
cooperation, and increase imports of high-quality products that meet the
needs of upgraded domestic consumption in order to expand the scale of trade’
(Macaes 2018a: 173). Pakistan has been promised a
household appliance industrial park should be built near Lahore through joint
venture. Household appliances, such as refrigerator, ice tank, washing machine,
air conditioner, TV, microwave oven and small appliances, can be produced
by absorbing foreign capital, adding investment and introducing technology.
Pakistani household appliance industrial development should move from
assembling imported parts to producing them locally. (Government of China
2015b: 14)
91
If the industrialisation of Xinjiang dominates China–Pakistan trade relations

then there are real competitive dangers for Pakistan. Alternatively, if the BRI
principle dominates the operation of the CPEC, then Pakistan may benefit
from FDI that seeks to relocate production facilities to Pakistan and greater
opportunities to export to China.
‘MUTUAL BENEFITS’: THE IMPORTANCE

OF DISTRIBUTION
The second official claim by the Government of Pakistan is that of ‘mutual
benefit’, that everyone will gain from the CPEC, or, as the Chinese government
frequently repeats, it is a ‘win–win relationship’. It is often simply assumed in
the literature that every place connected to the CPEC will benefit. Abbas
and Maaz list not only China and Pakistan, but also Iran, Afghanistan, India,
Central Asian Republics and the overall region as the beneficiaries of the
CPEC. They argue that the
CPEC is a flagship of growth and development as the enhancement of

geographical spillovers having improved road, rail and air transportation
system with frequent and free exchanges of services and through people to
people contact will enhance understanding through academic, cultural and
regional knowledge as well as the volume of trade and business activities, level
of production and energy movement will also be increased tremendously as a
result of co-operation among the countries along this belt. In this way, it is a
win-win model that will result in a well-connected, integrated region of shared
destiny, harmony and development, CPEC is a journey towards economic
regionalisation in a globalised world. It has the vision of generating peace,
development and win-win situation for all nations of this region. (Abbas and
Maaz 2017: 1)
After an all-party conference on 28 May 2015, an agreement was struck among

political interests in Pakistan to give priority to the western route of the CPEC
which runs through parts of Balochistan and KPK. Ahmad (2017) argues that
the western route of the CPEC can help integrate the least developed areas
into the mainstream economy of Pakistan. The CPEC, he argues, will create
opportunities for inclusive and equitable growth and so address long-standing
issues of regional disparity. The CPEC will connect distant areas to main
markets in Peshawar, Quetta, Lahore, Karachi and Gwadar through a network
92
of roads and railways and will link the country to the BRI and give Pakistan
direct access to global production networks.
Modern economic theory does not suggest that infrastructure improve-
ments will lead to balanced and equitable economic growth. The impact of
the CPEC will instead hinge on the interplay between reduced transport costs
and economic agglomeration effects. The CPEC will likely reduce transport
costs (even if the impact on costs and market efficiency seems exaggerated;
see later in this chapter). These reduced costs will help spread economic
growth to new areas by making it cheaper for those areas to produce and
transport goods and services to customers. This is where official Pakistan
thinking stops. This chapter will make the case that roads and railways run in
two directions. As well as increasing the ease of exporting, infrastructure also
increases the ease of importing. We are soon to hear the story of southern Italy
in the 1950s where new infrastructure facilitated the import of goods from the
north of Italy and the migration of people away from the south. Infrastructure
perpetuated relative poverty.
An extension of this story is that of ‘agglomeration externalities’ developed
by Alfred Marshall (1920). Agglomeration externalities work when firms and
workers derive benefits from being in close geographical proximity. Labour
may choose to migrate to be close to existing concentrations of industrial
activities. In Pakistan, for example, it makes sense for those with skills
in stitching footballs to live in Sialkot, the centre of football manufacture.
In turn, this means new football firms are more likely to open in Sialkot
where skilled labour is readily available. As well as goods and people, these
externalities can operate at the level of knowledge. A close location makes it
easier to exchange ideas about production and technology between firms in
similar industries and ancillary firms in marketing, research and design. When
there are agglomeration externalities, the impact of transport infrastructure
is more likely to stimulate forces of economic divergence, to make the rich
richer. The large existing markets, perhaps the richer and more populous areas
of Pakistan such as Lahore and Islamabad and the rapidly growing urban
areas of Xinjiang, are disproportionately attractive for firms and workers.
The combination of the importance of market access and the extra mobility
created by improved CPEC-inspired transport links could create a snowball
effect (a difficult analogy perhaps to sustain in the baking plains of Punjab).
Better transport connections will help firms relocate from other regions or
countries to take advantage of those large markets. This in turn will induce
93
intermediate firms, who supply inputs for production such as spare parts or
marketing services, to follow the migration (Puga 2008; Ottaviano 2008).
More production would put upward pressure on wages, which would attract
labour to migrate to the region. The arrival of new labour would boost local
demand and ease pressure in the labour market, which would in turn encourage
other firms to relocate. Whether or not such a chain reaction occurs depends
crucially on how costly it is to transport goods across regions. When transport
costs are very high, firms sell almost exclusively in their own location and
all that matters then are differences in local demand. Once trade costs are
sufficiently low, just a few extra firms and the additional workers they bring
with them can be enough to create a market size that will attract more firms
and workers and amplify differences further (Puga 2008; Ottaviano 2008).
At some point, rising costs of land or perhaps wages may induce firms to
start considering relocating to new areas, western Pakistan perhaps? Will the
CPEC cross-border infrastructure strengthen agglomeration externalities
in Xinjiang or spread economic development to western Pakistan? As the
discussion of Western China showed, this process may be long delayed due
to the cheap land, low-wage labour, inward migration from the rest of China
and massive government assistance to promote industrialisation in Xinjiang.
Historical and contemporary case studies also give us very clear evidence
on this question. One of the strongest empirical findings from studies of
historical and contemporary infrastructure projects is that infrastructure is not
a win–win solution but instead creates both winners and losers, and this is
true even as the net benefit is typically positive. This is just as predicted by the
theory of agglomeration externalities. There are many such examples.
During the early phases of nineteenth-century German industrialisation
that was spurred by railway-building, only in the region of Saxony and the
area around Berlin was there a significant increase in industrialisation–
urbanisation. There was even a slight fall in urbanisation in the east, which
increasingly specialised in the export of primary goods, especially cereals
(Lee 1988). In nineteenth-century Mexico, the initial impact of railways
was to make landholdings profitable as agricultural exports boomed and so
‘[r]ailroad construction precipitated land-grabbing on a scale unknown since
the Spanish conquest’ (Coatsworth 1979: 958). Tens of millions of hectares
of public lands in the sparsely populated northern states of the country in the
Yucatan Peninsula were sold cheaply or given away. The takeover of village
lands undermined subsistence agriculture and so reduced the cost of labour for
94
railroad construction projects and employment in the growing export sectors.

This formation of an agrarian and industrial proletariat and the sharp increase
in the concentration of wealth and income made a critical contribution to
the development of Mexican capitalism and to a more inegalitarian social
order (Coatsworth 1979). In India by 1914 around 18 per cent of agricultural
production by value was marketed over long or medium distances. This
set in motion a process of greater regional specialisation in the cultivation
of market-oriented cash crops such as cotton, sugarcane, indigo and poppy.
Across India cultivation shifted to high yield areas and away from areas close
to the rivers that had previously been used to transport agricultural produce.
Cotton cultivation came to an end in low yield areas in the United Provinces
(which grew a low quality, short staple variety) and central, northern and
eastern Oudh. Cotton became increasingly concentrated in the middle Doab
where yields were the highest and where the climate allowed an early sowing
and two crops a year (Derbyshire 1987). The United Provinces became the
great sugar-province of British India with a 30–50 per cent increase in its
sugarcane acreage being recorded between 1860 and 1895. During the same
period, sugarcane acreage fell in western and central India (Derbyshire 1987:
526). These changing patterns of specialisation led to growing inequalities.
Donaldson (2010) found that districts adjacent to newly constructed railways
experienced income increases and those further away saw actual declines in
income. For nineteenth-century Japan, Yamasaki (2017) finds that counties
that acquired early access to railways experienced much faster rates of
population growth and urbanisation in the 50 years to the 1920s, which in
turn stimulated regionally concentrated patterns of technology adoption and
industrialisation. Tang (2014) finds that the construction of railways in Japan
after the 1870s induced a considerable movement of manufacturing firms to
more populous regions and perhaps surprisingly to regions located away from
the coast. This latter effect was possibly to take advantage of labour released
from the traditional sectors located inland, for example, textile production
in central Honshu. These differential patterns in modern economic growth
were still visible by the 1920s (Yamasaki 2017). This 50-year process by which
inequalities were locked in is nothing compared to a finding by Dalgaard
et al. (2018). They argue that the construction of Roman roads up to 117
CE Europe was intended for military not economic motives, but did have
a profound economic impact. They find a remarkable degree of resulting
economic persistence in Europe. Areas with greater road density in 117 CE
95
were characterised with higher density of roads and economic activity in

2000 CE. Distributional effects of infrastructure were locked in for almost
2,000 years.
This pattern is exactly the same in stories of contemporary infrastructure
construction. After the 1950s Italian government efforts to promote
industrialisation of southern Italy through better infrastructure connections
with the north proved disappointing. Better transport links from north to south
increased the exposure to competition faced by firms in the south. Greater
fiscal transfers and welfare payments from the government to households in
the south boosted incomes and therefore consumer demand. This demand
leaked away into spending on imports from the north. Production in the south
failed to take off and the south remained stuck with a huge unemployment
problem. Fifteen million people migrated away from the south over the
twentieth century (Faini 1983). The US national highway system came into
being with the Federal Aid Highway Act of 1944. This led to a dramatic
transformation in the connectivity of the US. The legislation stipulated that
highways in the planned system should be located so as to connect the principal
metropolitan areas, cities and industrial centres. In 1947 the National Roads
Commissioner approved the final proposals from the states for 37,324 miles
of highways to be built to federal interstate standards. This plan represents
most of the systems we see in contemporary America, though some of it was
not completed until the 1980s (Baum-Snow 2007). Construction of the urban
transportation infrastructure played a key role in influencing changes in the
spatial distribution of the population in US metropolitan areas after 1950.
Between 1950 and 1990 the total city-centre population of urban areas in
the US declined by 17 per cent despite a population growth of 72 per cent
in urban areas as a whole. Better transport allowed people to move to the
suburbs and commute to work in the centre (Baum-Snow 2007). US highway
construction led to total earnings increasing (between 6 and 8 per cent) in
counties adjacent to the highways and falling (1–3 per cent) in non-adjacent
counties as firms moved to take advantage of the good access offered by the
highways (Chandra and Thompson 2000). Isserman and Rephann (1994)
find that the construction of rural highways in the US in the 1960s and early
1970s had a positive impact on counties that contained a city of more than
25,000 people. They find the effect to be concentrated on retail sales rather
than manufacturing, and prompted a migration from unconnected areas or
areas with smaller cities. Duranton, Morrow and Turner (2014) find that
96
distributional patterns also occurred within industry. The increase of within-

city highways in the US caused cities to become more specialised in the
production of heavy goods that required good transport links at the expense of
smaller and more locally oriented firms.
From the early 1990s, China embarked on an initiative to build and upgrade
its transportation infrastructure, especially highways. The construction of
China’s National Trunk Highway System (NTHS) was an effort to connect
cities with an urban population above 500,000 and also border-crossings on
a single expressway network. From a low level, spending on transportation
grew by 15 per cent per annum to about $200 billion in 2007, much of which
occurred in cities (Faber 2014). This was accompanied by rapid migration of
the rural populations to cities. Using satellite images of lights at night from
1992 to 2009, Baum-Snow et al. (2012) find strong evidence that the presence
of both radial highways and ring roads outside the central city reduced the
population density in the centre of cities (just as in the US in earlier decades).
They estimate that each additional radial highway in China displaced at least
4.2 per cent of the central city population to suburban regions. Each additional
radial railroad line also caused a displacement of about 17 per cent of the
central city’s GDP and 26 per cent of its industrial sector GDP to surrounding
regions. NTHS connections also reduced industrial and total output growth
in peripheral counties on the route between targeted metropolitan centres.
Highways failed to diffuse production from metropolitan regions to the
periphery (Faber 2014). As in the US, it was heavy industry in China that
benefited most from highway construction at the expense of light industry
(Xu and Nakajima 2013).
Railway investment in China by 2010 was around $120 billion per annum
or 23 per cent of all state investment; this was a 10-fold increase over 2003.
Much of this increase went into building high speed rail (HSR), the total
length of which reached 6,000 kilometres by 2007. Qin (2014) explores the
distributional impacts of HSR upgrades in China. To maintain speeds of 200
kilometres or more per hour, a high-speed train only tends to stop at populous
urban areas where there are significant demands for time savings rather than
in small cities and rural areas. Around 3,000 of the 6,100 passenger train stops
in China were abandoned in the 10 years to 2014 due to this need for faster
train travel. The results show that in those locations that watched high-speed
trains pass by but not stop, the GDP per capita declined by 4–6 per cent on
average. This was caused by a more than 10 per cent decline in investment.
97
Since HSR significantly reduces the transportation cost of passengers rather

than goods, its negative impact in the affected counties was more pronounced
in the service than in the manufacturing sector. In Europe regional policies
constitute around one-third of the EU budget and their focus became
increasingly skewed over time towards infrastructure, especially transport
networks in regions with GDP per capita below 75 per cent of the EU average.
Income convergence across European regions did not narrow after the 1970s
and even widened in the 2000s. Divergence was especially marked in terms of
employment (Puga 2008). This pattern is similar in the case of large energy
infrastructure. Duflo and Pande (2007) find that dam construction in India
boosts agricultural production and reduces vulnerability to rainfall shocks in
districts located downstream. In areas close to where the dam is constructed,
agricultural production increases but so too does its volatility. Poverty falls and
rises in the two areas respectively.
The response to distributional questions or pockets of deprivation is often
to construct even more infrastructure to try and ensure wider inclusion. It is a
fallacy that distributional issues can be overcome by even more infrastructure
construction, for example, that rural road building will offset rural disadvantage.
Asher and Novosad (2016) analyse labour market consequences of rural
transport construction in India. They study the Pradhan Mantri Gram Sadak
Yojana (PMGSY), or Village Road Programme, which was launched in 2000.
The scheme had the goal of providing all-weather access to unconnected
habitations across India. They estimate the causal effects of the $37-billion
PMGSY, which by 2015 had constructed over 400,000 kilometres of roads
to provide 100,000 Indian villages with paved connections to the wider road
network. They find that road construction led to a reallocation of labour out of
agriculture and into wage labour. This is evidence that poor rural transportation
infrastructure had been a major barrier to the efficient allocation of labour in
India. The roads also led to an increase in household earnings and a 20 per
cent increase in the share of households living in buildings with a solid roof
and walls. As well as these impacts on average income, there were profound
distributional impacts. Villages that were closer to cities experienced a much
greater shift out of agriculture than those located at a greater distance. Within
villages, income effects were most pronounced among those groups with the
lowest costs and highest potential gains from participation in labour markets,
that is, households with smaller landholdings and more male workers.
Bryceson, Bradbury and Bradbury (2008) examined rural road construction in
98
Zambia, Ethiopia and Vietnam and again showed a significant distributional

impact. They found that rural roads did improve the speed of travel but that
the main beneficiaries were the wealthier residents of villages, who were more
likely to own bicycles, motorbikes or cars and so were better able to utilise the
new road to its full potential. The dynamic effects were strikingly different
between the various country case studies. While motorbike use was spreading
down the social scale in Vietnam, this was less true in Zambia and Ethiopia.
This indicated that there were numerous country-specific constraints on road
development promoting wider social development.
One exception to the distributional impact was the construction of more
than 2,000 miles of roads in the poor Appalachian region of the US after the
mid-1960s. This resulted in the Appalachian counties growing faster in terms
of earnings and also per capita income. Here the distributional consequences
of infrastructure investment were mitigated by the wide-ranging remit of
state intervention in the region that was mandated by the 1965 Appalachian
Regional Development Act. These interventions included not just highway
investment but also massive welfare payments, land restoration, flood control,
water resource management, vocational education, sewage treatment and
other interventions (Isserman and Rephann 1995). The Appalachians are a
relatively small corner of an enormously large and wealthy national economy.
The US could afford lavish funding to intervene in this small pocket.
There is no prospect of a much poorer economy, Pakistan, affording similar
comprehensive interventions across backward areas over the entire country.
The distributional implications of CPEC-infrastructural investment
have been largely ignored by the CPEC literature beyond passing anecdotes.
There are frequently cited complaints, for example, that some groups, such as
fishermen displaced by the port construction at Gwadar, will miss out on the
benefits of the CPEC (Choudhry 2018). This discussion is highly relevant,
but extremely limited in its scope, for the case of contemporary Pakistan
which has long experienced striking regional inequalities in economic growth
(M. A. Zaidi 1992).
Economic theory takes the concern with distribution a step further and
suggests that the resulting patterns of inequality will also have long-run
implications for patterns of economic growth. An optimistic view is that
the entry of China into the global economy will induce changes in patterns
of production and export in other countries and regions and so offer new
opportunities for mutual benefits from trade. In reality, specialisation in
99
agriculture or low-technology production may lock in a region or country

to long-run patterns of slower economic growth (Deraniyagala and Fine
1999). It is upgrading and shifting to more high-technology manufacturing
that is beneficial for long-term economic growth. High-technology products,
for example, experience faster demand growth on the world economy over
time (Lall 2000). In this view the very efficiency of the railway system in
nineteenth-century India or Mexico locked the economy into long-run slow
economic growth. The impact of China on world trading patterns is discussed
further in Chapter 7.
A key safety valve for regional polarisation has historically been the
migration of people from the poor to the fast-growing areas of the national or
global economy. In contemporary India the population of the poor, landlocked
states in north central India, Madhya Pradesh, Uttar Pradesh, Rajasthan and
Bihar runs into the hundreds of millions of people. There is no safety valve
in contemporary India that permits mass migration to the rapidly growing
coastal states of Gujarat or Tamil Nadu. This is not a new phenomenon.
As far back as the census of 1881, 97 per cent of the population were living
in the province in which they had been born. It was marriage not economic
incentives that was the main reason for migration. Wage differentials persisted
across time and space in the nineteenth and early twentieth centuries and were
not being narrowed by migration (Collins 1999). Not much changed after
independence in 1947 (Cashin and Sahay 1996). Across the 1980s and 1990s a
representative sample of households in rural India found that, despite growing
wage inequality between rural and urban areas and across states, the likelihood
of male migration actually declined (Munshi and Rosenzweig 2005). In India
‘[t]he highest level of movement are recorded within the same district. The
flow of migrants across state lines is a trickle. Since 2001 there has been a
slowdown in permanent or long-term migration’ (World Bank 2009a: 163).
Many reasons have been put forward to explain the relative immobility of
labour. Many government welfare programmes are structured around location.
The rural employment guarantee programme launched in the mid-2000s
guarantees employment for people in specific (rural) geographic locations;
migration would mean losing this benefit. There are cultural prohibitions
on marriage outside specific caste groups, which may restrict the choice of
marriage partner choice to a local pool. In the advent of economic shocks
such as bad harvests or unemployment, local kinship and caste groups may
provide a source of informal insurance; migration would risk losing access to
100
this safety net. The enormous linguistic, religious and caste diversity of India
may also make the practicalities of migration harder. Migration in India is also
linked to constructed political constraints. The Shiv Sena party in Mumbai,
who have long controlled the city government, have an explicit and often
violent programme to keep the city and its public services a preserve of the
locally born.
In the mobile US by contrast, every decade a quarter of the population
changes its state of residence (World Bank 2009a: 46). Decressin and Fatas
(1995) find that in the US, region-specific shocks do not affect employment
or labour force participation; instead, they are absorbed by inward or outward
internal US migration. In the EU, regional shocks tend to have a much more
lasting effect and impact particularly on labour market participation rates.
Puhani (2001) finds that in the EU the reaction of migration to economic
shocks takes several years. It is harder culturally to migrate across international
borders. There is no migration impact in the EU even one year after a shock,
whereas in the US after one year, more than half of the impact will already
have been absorbed by migration. In China over the 1980s and 1990s, perhaps
100 million people moved from inland to rapidly growing coastal China.
A change in employment laws allowed firms located in SEZs to hire labour on
contract. These contracts gave firms flexible hiring and firing, bonus payments,
flexible wages and the ability to recruit migrant workers currently located in
rural areas. Less skilled contract labour often lived in company-managed
dormitories with four–six people per room (Sklair 1999). The proportion of
temporary migrants in the population of Shenzhen increased from 1 per cent
in 1979 to 72 per cent in 1994. The majority of the migrants were single and
42 per cent of them were in the 15–24 age group, and so required less child
care and housing costs. There was a preference for young women migrants.
By the time of the 1990 Chinese census, 75.4 per cent of male and 87.5 per
cent of female temporary migrants had less than high school level of education
(Liang 1999).
In contrast to India, Pakistan was a country born of migration. Soon
after independence, more than 50 per cent of the populations of major urban
areas such as Karachi, Lahore and Hyderabad were composed of migrants.
There are no directly comparable studies with India, but evidence for Pakistan
suggests that migration has continued at a high level over the subsequent
decades (Perveen 1993). Tension and conflict have been prevalent, for example,
in Karachi during the 1990s, over jobs and urban living space. This has not
101
hindered long-term migration. By 2000 ethnic Sindhis were a tiny minority

of the urban population of Sindh (Ahmar 1996; A. Khan 2002). There are
striking differences with the pattern of migration in China. Like China, the
majority of migrants in Pakistan are female, with the peak age of migration
at 15–24, as was the case in Shenzhen over the 1980s. Here the resemblance
ends. Women in Pakistan are not migrating over long distances; more than
80 per cent of women in the 1979 Pakistan Labour Force and Migration
Survey remained in their home province (Irfan 1986). The main motivation
for female migration in Pakistan was not the opportunity for employment,
but rather marriage. In 1979 more than 95 per cent of female migration was
linked to marriage (Irfan 1986). This pattern was essentially unchanged over
the intervening decades. The sample of over 89,000 people used in the 1996–7
Pakistan Labour Force Survey found that although women comprised nearly
53 per cent of migrants, 94 per cent of those migrating were doing so for
non-economic (marriage) reasons (Khan and Shehnaz 2000). The economic
slowdown in the mid-1990s seems to have slowed this trend. The share of
females in total migration (excluding marriage) fell from 8.5 per cent in
1996–7 to 6.8 per cent in 2006–07 (Hamid 2010: 4) and remained at low
levels according to the 2010–11 Labour Force Survey (Kanwal, Naveed and
Khan 2015). It also seems to be the case that it is not unskilled young women
(as in China) who are migrating for economic reasons in Pakistan and who
could provide a potential labour force for factories in the SEZs. While 33 per
cent of those with higher education left rural areas in 1979, only 5 per cent
of illiterate people from rural areas migrated, with a majority of them ending
up in another rural area (Irfan 1986). This was still true in 1996–7, when
migration was strongly linked to education for both men and women, with
the strongest impact coming from professional and post-graduate education
(Khan and Shehnaz 2000).
The reasons for the low level of female migration in Pakistan include the
cultural prohibitions on the free movement of women. The 2004–5 Labour
Survey showed that only 14.6 per cent of women of working age left the
household to work in the labour force (Ejaz 2007). This rate showed only a
marginal increase over data going back to the early 1970s (Afzal and Nasir
1987). With such low rates of participation, even in the local area, there is
little chance of women being permitted by the household to migrate over
long distances, as in China, to find work. Women’s labour force participation
increases with age and the number of other household members leaving the
102
household for work. Labour participation declines with marriage, the number
of children in the household and whether the woman lives in an extended
household and rises with her level of education. The ubiquitous prevalence of
marriage, the relatively young age at marriage for women, the high levels of
fertility and resulting larger families and widespread prevalence of extended
families all become structural constraints on the ability of women to work
outside the household (Kozel and Alderman 1990; Ejaz 2007; Fatima and
Sultana 2009).
‘COMMON PROSPERITY’: WILL THE CPEC

HAVE A TRANSFORMATIVE IMPACT?
We can think through some of the historical and contemporary examples of big
infrastructure investment and focus on those cases where that infrastructure
represented a transformative change, examine what impact that made and
then think about the relevance of such thinking for contemporary Pakistan.
A general rule in the historical literature is that when transport infrastructure
induces a big shift in the means of transport, it can have a transformative
economic impact. There are numerous examples of countries where pre-rail
transport was dirty, dangerous and slow.
In India pre-rail travel was constrained by slow speeds, seasonality and
the high cost of traditional transport technologies. Riverboat carriage was
risky, seasonal, with waterborne travel being difficult at the height of the
monsoon and during the early summer dry season in many places, and very
slow. Downstream from Allahabad to Calcutta could take 20 days travelling at
40 miles a day. The return journey was more unpredictable as travelling against
the current needed both rowers and towing from the banks, and so could
take between two and three months. Travel speeds rarely crossed 10 miles a
day. Overland transport was equally slow, seasonal and extremely costly. Only
some overland transport was done by carts along improved roads. The bulk
of overland travel was carried by bullocks. On the best surfaces and during
optimal weather conditions, bullocks could cover 20 miles a day; overall pack
bullocks were the least efficient mode of pre-rail transport. There was an
eight-month trading season between October and May which encompassed
the two principal kharif and rabi harvests. Bullocks could often not travel
during this season, when they and their handlers were required to till the fields.
The remaining hot-dry season and summer monsoons meant that roads and
103
waterways were difficult to use. In the United Provinces, accessible commercial

relations were limited to a semi-circle which extended just over 100 miles.
Railways, by contrast, offered speed and the possibility of monsoon shipment.
From the early years of their construction, railways showed themselves to
be clearly superior to road, river or coastal transport. Trains were able to
travel 600 kilometres a day and offered predictable timetables throughout
all months of the year. Railway freight rates were also much cheaper than
either road, river or coastal travel (Derbyshire 1987; Donaldson 2010). There
were complementary infrastructure changes that increased the potential of
rail travel even further. The opening of the Suez Canal in November 1869
shortened the distance from Britain to Bombay from 10,700 miles to 6,200
miles. By 1883–4 85 per cent of the total value of trade between Britain and
India was transported through the canal. The establishment of a telegraph
network linking India’s major towns with one another in the 1850s and then
linking India with Europe and other parts of Asia in the 1870s ensured that
information about supply, market demand and prices flowed easily from region
to region (Collins 1999). In nineteenth-century Russia, before the arrival of
the railways, interregional trade in grain was conducted by slow and expensive
wagon transport or waterways that were frozen across much of the country for
six months a year (Metzer 1974). In contemporary China, the HSR covering
the 2,252 kilometres between Kunming and Shanghai cut the travel time from
34 to 11 hours (Griffiths 2017).
A key research agenda in these historical case studies has been to show how
infrastructure has influenced the efficiency of markets. The implication being
drawn from neoclassical economic theory is that economic growth is best
promoted with freely functioning and competitive markets. Such orthodox
theory predicts that where a profit opportunity exists, somebody will take it.
Where prices and profits are high, commodities will be moved until prices and
profits decline to a common average—arbitrage.
In India, prior to the coming of the railways, price levels varied substantially
between different regions, particularly for low-value crops. Each area tended
to be self-sufficient in agriculture and so prices reflected wide variations in
local costs of production. Some regions were frequently short of grain, which
left them prone to famine, while others had a relative abundance of food.
In the 1860s the prices of grains in some districts were eight to ten times
higher than the prices in others (Hurd 1975). In India the railways caused
transportation costs to fall by approximately 80 per cent per mile, which made
104
trade in these low-cost and bulk goods possible. Entrepreneurs took advantage
of the prevailing price differentials. By 1910 almost 30 per cent of the wheat
crop (amounting to 3 million tonnes of wheat) and 14 per cent of the rice crop
were carried by rail (Andrabi and Kuehlwein 2010). Historical India has an
abundance of foodgrain price information. The British colonial state collected
very detailed data from 1861 onwards. Price data for wheat and rice shows
that a statistical measure of price variations between districts (the coefficient
of variation) declined from the mid-nineteenth to the early twentieth century.
This decline occurred alongside the years of railway construction. Price
variability was systematically lower in districts with the railways than those
without it. Price convergence slowed in the early nineteenth century as the
era of railway construction came to an end (Hurd 1975). In Bengal, across
the late nineteenth and early twentieth centuries, the same measure of price
variation declined across 70 subdivisions of Bengal and also showed a decline
in seasonal price variations (Mukherjee 1980). The most spectacular study is
that of Donaldson (2010) who, if you recall from the awed mention of his work
in Chapter 2, used seven million observations on district-level prices, output,
daily rainfall and interregional and international trade in India. He found that
the railways reduced trade costs, the responsiveness of prices to local weather
(rainfall) shocks, interregional price gaps and real income volatility, and
increased trade volumes and income levels. One problem with such studies
that purport to show that the railways were correlated with economic success
is that the railways were often built in areas already experiencing economic
growth. Haines and Margo (2006), for example, found that the railways were
often built along well-established trading routes that already had good river-
based transport. Making allowance for this problem reduces significantly but
preserves the impact of the railways on price convergence.
The evidence that infrastructure drives price convergence is consistent
outside of South Asia. Between 1885 and 1908, interstate differences in corn
prices declined in Mexico. Railways played a significant role in this decline
as prices converged at twice the speed in those states with railways in 1884
than in those states without railways (Dobado and Marrero 2005). In Russia
there was a clear decline in price differentials starting in the 1870s, between
wheat prices in Odessa and St Petersburg and other important regional
markets such as Riga–Moscow and Odessa–Moscow. The crop rye began
this era as a locally grown and consumed subsistence crop but experienced
rapid commercialisation as it became increasingly transported over distances
105
and sold in markets. These changes coincided with the first surge in Russian
railway construction. About 83 per cent of the decline in price differentials
have been attributed to the railway-induced decline in transportation costs
(Metzer 1974: 544).
The same impact was also felt globally. The opening of the Suez Canal
helped India to become a large exporter of primary products (cotton, opium, rice,
oil seeds, jute and wheat) and a large importer of manufactured goods (cotton
textiles and yarn, iron and steel, brass and copper, and railway equipment).
The difference in the prices of key commodities between Britain and India
declined rapidly in the late nineteenth century (Collins 1999). There is also
similar evidence from the contemporary era. The Golden Quadrilateral (GQ )
highway project in India had a significant impact on prices for manufacturing
firms in the mid-2000s. For districts located within 25 kilometres of the GQ ,
input prices declined by almost 60 percentage points more than in districts
further away (Asturias, Garcia-Santana and Ramos 2017).
In terms of the domestic economy, the relevance of these studies for
contemporary Pakistan is limited. There is good evidence that markets were
efficient in Pakistan well before the launch of the CPEC. Price data shows
that there was rapid convergence of prices across 15 commodities (especially
among more easily transported non-perishable commodities such as gram
pulse, gas cylinders, refined sugar and wheat), across 10 major cities between
2000 and 2011 (Alam and Bhatti 2014), across 35 cities between 2001 and 2008
(Mohsin and Gilbert 2010) and specifically among food commodities (but
less so other commodities) across 35 cities between 2001 and 2011 (Ghauri,
Qayyum and Farooq 2013). There is at most only a small relation between the
distance between Pakistani cities and the speed of price convergence (Alam
and Bhatti 2014).
In terms of international trade, these studies are of likewise limited
relevance for Pakistan. The estimates (see Table 3.1) showing the dramatic
reductions in distance that the CPEC makes for firms in central China who
wish to export to the Middle East or Europe are based on physical distance
and do not account for the cost of transport. The cost of travel is of crucial
importance. The enormous size of modern shipping vessels gives them a big
advantage. Containers travelling by rail from China to Europe can cost five
times that of a similar journey by sea. This implies that while rail lines may
take business from air routes, they are unlikely to account for more than a tiny
fraction (maybe 1 or 2 per cent) of maritime cargoes. In January 2017 a train
106
arrived in London from China pulling 34 containers from its starting point.
A small ship can carry hundreds of containers and a large vessel more than
10,000 (Frankopan 2018: 99). For the CPEC route crossing the mountains
between Kashgar and northern Pakistan, there are more serious cost and
geographical considerations. The tariffs or taxes needed to pay for the upkeep
of the route and to move freight over a 15,000-foot pass are likely to make the
CPEC land route non-competitive compared to sea routes. The transport cost
of a barrel of oil by sea from the Middle East to Shanghai at a shipping rate
of $75,000 per day at 23 kilometre per hour with 2 million barrels of oil cargo
is $0.9 per barrel and inland costs to Chongqing adding $1.23 extra. Moving
oil from Ras al-Tanura to Gwadar and then by rail to the heartland of China
would cost between $8 and $12.4 per barrel. It is likely that maritime shipping
routes will remain cheaper, simpler and more secure for transporting crude oil
and other goods into China (Collins and Erickson 2010).
Careful calculations show that the CPEC rail route into Western China
would be unlikely to be able to carry enough cargo to become a significant
transport route for Pakistan. In the 2000s a modern one-line rail track in the
US could handle around 16 trains a day. The Pakistan to China rail track is
likely to be one track each way and carry about 12 trains a day. US freight
trains carried an average of 2,800 tonnes of cargo in 2004. The CPEC route
crossing over the steep Khunjerab Pass would more likely carry loads of
around 2,000 tonnes. In aggregate annual terms, this implies that the CPEC
would be able to handle 8.75 million tonnes of cargo per year or 175,000
barrels of oil if all trains carried oil. The CPEC rail route would need an
enormous expansion, to three–four lines, to make a significant dent in the
existing reliance on sea transport. This would represent a vast increase in
its likely cost (Collins and Erickson 2010). The CPEC route into Western
China goes through areas that are subject to insurgency and natural disasters
in the Karakoram range, such as flooding, avalanche, landslides and seismic
activity. The Khunjerab Pass is closed during the winter months and trade
between China and Pakistan comes to a standstill. While ships can simply
reroute around trouble points, this is not possible with roads and railways
(Collins and Erickson 2010). The Karakoram Highway, though functional
since 1979, has remained underutilised for all of these reasons. These dangers
are reflected in existing trade patterns. Of the total existing trade between
China and Pakistan, only 1 per cent occurs through land-based infrastructure
while 97 per cent through sea and 2 per cent by air (Shafqat and Shahid 2018).
107
The same story is true of the proposed oil pipeline projects connecting Iran,
Russia, Burma or Kazakhstan to China either directly or through Pakistan.
Some look economically viable (the line from Kazakhstan) and others less so
(Burmese pipeline), but do not offer enough capacity to significantly reduce
China’s dependence on seaborne imports (China SignPost 2010).
Another important lesson we can learn from this historical literature
is that infrastructure, even of the transformative kind, is rarely sufficient to
create efficient markets and ensure the flexible allocation and reallocation of
commodities and factors of production. The evidence as reviewed elsewhere has
shown a consistent and positive impact of infrastructure on market efficiency.
The impact, however, is rarely transformative. The convergence of prices in
British India was due in part to the railways (20 per cent of the total according to
one estimate) but also to the replacement of bullocks with carts, the construction
of paved roads, the use of steamboats for river transport, the introduction of a
telegraph and postal service, after 1857 the establishment of internal peace in
India and the abolition of internal tolls and use of a single national currency
(Andrabi and Kuehlwein 2010). Turning to contemporary India, Van Leemput
(2016) asks how large are the internal versus external barriers to trade in India?
He found that internal trade barriers do account for a substantial fraction of
total barriers to trade. For international imports shipping from the rest of
the world to the ports of India accounts for 56 per cent of the total barrier
and shipping from the ports in India to local destinations accounts for the
remaining 44 per cent of the total barrier. Non-port states face trade barriers
that are around three times higher than the port states for both international
imports and exports. This is mainly driven by costs of trading cross-state in
India. These barriers do include infrastructure but also policy barriers such as
corruption and a burdensome tax administration (Van Leemput 2016).
While CPEC-optimists construct ever-longer lists of constraints on
economic growth in Pakistan that will melt away as the CPEC becomes
operational, the reality appears less optimistic. The Global Competitiveness
Report of 2017–18 reported a survey of executive opinions in Pakistan on the
‘most problematic factors for doing business’. On top of the list was corruption,
followed by tax rates, government instability, crime and theft, inefficient
government bureaucracy, poor work ethic among labour, access to financing
and policy instability. Inadequate supply of infrastructure was only 11th on
the list (World Economic Forum 2017: 23). For investors in Pakistan, there
is a significant problem of ensuring that they can profit from that investment.
108
Corporate and income taxes are low enough, so the benefits are not being
taxed away. The more important problems are related to poor property rights
and weak contract enforcement. The evidence for this can be easily observed
from looking at the various Global Competitiveness Reports produced by the
World Economic Forum (see Chapter 8).
Launching a small business was a long, expensive, and cumbersome procedure.

A poorly functioning legal system made banks reluctant to lend as they
faced a significant default risk from borrowers. Those borrowers could then
continue for years until being declared bankrupt by a corrupt and inefficient
court system, and be mandated to repay the debt; even then, once assets
were scheduled for auction to repay debtors, they would typically disappear.
Lending for property in Pakistan was hindered by inefficient, unclear, and
frequently disputed rights to land and land titling. The proximate constraint to
growth was low investment and its deeper causes lay in the lack of protection
afforded to potential investors. (McCartney 2015a: 72, 2015b)
Macroeconomic management in Pakistan remains poor. Pakistan has long

faced chronic constraints on its ability to mobilise tax revenue (see Figure 4.1)
which has remained stuck at around 10–12 per cent of GDP in the years
to 2017. Trade liberalisation in Pakistan in the late 1990s directly reduced
government tax revenue. The impact is very evident in Figure 4.1. The loss
of government revenue led in turn to lower public investment and finally
to reduced private investment. This was not surprising. Trade liberalisation
in developing countries has been found to consistently lead to a loss of tax
revenue and has forced reductions in infrastructure and education spending
(Khattry and Rao 2002; Khattry 2003). Pakistan has the structural features
which are typical of developing countries that make mobilisation of tax
revenue difficult. These include (a) the dispersed, low-income and fragmented
subsistence sector in agriculture and the small-scale informal sector in urban
areas, (b) the weak capacity of the tax administration and (c) the lack of good
practice in accounting procedures. These structural features make it difficult
to mobilise revenue from taxation of income and consumption taxes. Imports,
by comparison, mainly enter Pakistan through a few ports, airports and
international borders, and so are easier to physically verify and subject to taxes
than are the millions of income earners or consumers or hundreds of thousands
of small businesses (McCartney 2012). As Chapter 8 shows, state capacity in
Pakistan is poor and declining, which makes the process of tax mobilisation
109
Figure 4.1 Tax revenue as a share of GDP in Pakistan, 1970–2017

Source: Trading Economics (2019).
in Pakistan even more problematic. Figure 4.1 shows that tax revenue has
remained low in Pakistan as the government failed to raise revenue elsewhere
to compensate for revenue lost from reduced tariffs (McCartney 2015a).
The problems of low revenue mobilisation have been compounded by a
lack of control over government spending. The annual losses of public sector
enterprises are around 3.8 per cent of GDP. Pakistan International Airlines
(PIA) alone has losses running at around Rs 4 billion per annum. Power sector
arrears were reduced in the financial year (FY) 2014–15 but were allowed
to rise again in FY 2016–17 and the current stock of arrears soon after
reached around Rs 374 billion (about 1.2 per cent of GDP). Losses due to
theft and inefficiencies in the distribution of electricity by the public sector
remain high at around 20 per cent of generation. Government spending of
more than 20 per cent of GDP has generated a large (5 per cent or more)
budget deficit. The chronic budget deficit led to a current account deficit and
both domestic and international debt (Iqbal 2017: 10). The IMF returned to
Pakistan in May 2019 to negotiate a debt bail-out. The failure to mobilise
resources domestically has been compounded by a failure to export from
Pakistan. The reasons are many and include competition from other countries
such as Vietnam, regulatory burdens, the business climate, political instability
and the unavailability of skilled labour (Amjad et al. 2015). Between 2010
and 2017 there was also a consistent overvaluation of the exchange rate,
which reached the magnitude of between 10 and 20 per cent (IMF 2017).
Exports as a percentage of GDP declined from 10.3 per cent of GDP in
110
2013–14 to 7.9 per cent in 2017–18; over the same years, imports remained
stable at around 17 per cent to 18 per cent of GDP. The current account
balance correspondingly worsened from –1.3 to –6.3 per cent over these
years. Pakistan remains dependent on FDI and worker remittance income to
cover this trade gap, leaving its external finances precarious (Government of
Pakistan 2019: 5–6).
While the CPEC has promised to boost the supply of energy, quantity is only
a part of the solution to the crisis. A recent estimate suggests that distribution
and transmission losses amounted to 20 per cent of the total output. The energy
sector also faces problems stemming from poor institutional capacity, the use of
outdated technology, poor revenue collection, circular debt and weak monitoring
and accountability (Shafqat and Shahid 2018). With agriculture still accounting
for 20 per cent of GDP, and much of it being rain- rather than irrigation-
dependent, the threat of emerging water shortages is a significant concern.
The IMF in 2015 indicated that Pakistan’s per capita annual water availability
has dropped by around 82 per cent since 1947. Pakistan is ranked as among the
most water-scarce countries in the world. Pakistan’s use of water in cubic metres
per unit of GDP, which provides a measure of the water intensity rate, was also
among the highest in the world that year (Shafqat and Shahid 2018).
There are no studies for Pakistan that attempt to quantify the importance
of infrastructure relative to other constraints on economic growth. A typical
example in the existing literature is by Ahmad, Naz and Majid (2018), who
note that small and medium enterprises (SMEs) provide a great deal of
economic advantages, such as employment growth and that they require less
infrastructure and capital. They note that the growth of SMEs in Pakistan has
been relatively slow and that the CPEC will provide lots of opportunities such
as proximity to raw materials, middle-class consumers and markets in China;
therefore, they conclude that the CPEC will lead to the growth of SMEs.
The argument is a string of assertions lacking theory, evidence and any attempt
to quantify why SMEs have failed to grow and the extent to which the CPEC
is likely to alleviate the key constraints on growth.
The predictable failure of trade liberalisation (discussed in more detail in
Chapter 7) and improved CPEC infrastructure to energise export and GDP
growth in Pakistan can be seen from a study by Chaudhry, Jamil and Chaudhry
(2017). They examine firm-level and sector-level data from before and after the
FTA was signed with Pakistan in 2006. The FTA reduces trade costs between
China and Pakistan in a broadly equivalent manner to better infrastructure.
111
They examine the impact of these reduced costs of trade on various indicators
such as productivity, value added, trade flows, employment and the number
of firms. They focus on those sectors that faced greater competition from
Chinese imports and examine trends over time which allows a before-and-
after study of the impact of the 2006 FTA. They find that productivity has
fallen in most of the sectors that faced large reductions in trade costs induced
by the FTA. These included sectors that Pakistan has been successful in
exporting to the rest of the world, such as textiles, sports goods and medical
and dental instruments. The few exceptions where there was a successful
domestic response included leather, pharmaceuticals and rubber. There is also
some, though weak and non-significant, evidence that the number of firms
and employment in these sectors declined after the signing of the FTA.
This was not evidence that weak firms in Pakistan were being shaken up by
the exposure to Chinese competition. Firms that were impacted by lower
Pakistani tariffs on Chinese exports were initially more productive than other
sectors and this relative advantage declined after the signing of the FTA
(Chaudhry, Jamil and Chaudhry 2017).
This Pakistan-specific evidence is supported by much larger studies of the
global economy. Despite the dramatic headline numbers of trillions of dollars
of infrastructure investment, the available evidence does not show that the BRI
will have a transformative impact on the global economy. Zhai (2018) predicts
the likely impact of the BRI by quantifying the impact of infrastructure
connectivity through investments in railway, roads, ports, pipelines, airports
and energy. The Computed General Equilibrium (CGE) model he develops
forecasts a $1.4 trillion investment in infrastructure between 2015 and 2029,
25 per cent of which will be spent in China and the rest in BRI countries,
two-thirds funded by China, one-sixth by BRI countries and one-sixth by
other countries. This will lead to a slight, 0.3 per cent, increase in investment
in China. By 2030 the infrastructure stock will expand by only 6–7 per cent in
other BRI countries. The overall impact is forecast to be 1 per cent of GDP,
with the biggest beneficiary actually being India (despite not participating in
the BRI) because of the openness of its economy and greater opportunities
for trade with countries that are participating in the BRI. Global gains are
forecast at $1.6 trillion, which is equivalent to only a 5 per cent growth in
world trade. India is forecast to gain $332 billion, with Malaysia, Thailand
and Vietnam also being major gainers. Belarus, Ukraine, Moldova, Armenia,
Azerbaijan and Georgia are also forecast to gain significantly due to their
112
large trade spillovers with other BRI countries and high regional energy
dependence. Pakistan will experience the largest increase in investment,
but even this is not substantial. The total investment is forecast to rise by
5 per cent on average between 2015 and 2019.
IN SUMMARY
It is a difficult proposition: how do we evaluate the success or otherwise of the
CPEC, a project that is confusing in its coverage across space, time and sector?
This chapter has chosen to do so against the Government of Pakistan’s own
claims that the CPEC will have a complementary relationship with China,
it will generate mutual benefits and will transform Pakistan’s economy. The
economic evolution of Western China has been hitherto ignored by scholars
of the CPEC. Existing production of cotton and fruit and the rise of textile
production in Xinjiang mean Western China is becoming a more competitive
threat to Pakistan’s economy. The Chinese government claims that the CPEC
will represent a win–win outcome. The Government of Pakistan claims
that the CPEC will serve to integrate the backward regions of Pakistan
into a national development success story. Both of these claims ignore the
voluminous historical and contemporary case study evidence that shows that
big infrastructure always creates both winners and losers. The safety valve
of migration—people moving to take advantage of economic opportunities
created by the CPEC—may work in the case of Pakistan, which was a country
born of migration. Women, however, are unlikely to take part in this migration;
there are severe cultural constraints on their ability to work outside the
household and especially to migrate long distances in search of employment.
There are doubts about whether the CPEC has the capacity to transform
Pakistan into a ‘Falcon economy’. The CPEC will not transform Pakistan
from a bullock-travelling economy into railway-travelling as transformative
infrastructure has done in the past. Pakistan’s roads and railways are already
in a good enough condition such that they are rarely cited as important
constraints by business. Price differentials between cities indicate that Pakistan
already has fairly efficient markets. Evidence shows that there are other, more
significant constraints on economic growth, such as education and skills, lack
of long-term credit, enforcing contracts and political instability, which will
not be tackled by the CPEC. An important part of the CPEC are the nine
SEZs being planned to promote domestic industrialisation; the next chapter
evaluates the likely success of this effort. 113
5
The Dragon Uncoils

Special Economic Zones (SEZs) from
Shenzhen to Africa
A major headline feature of the CPEC are the nine special economic zones
(SEZs) that are planned to be established across Pakistan. The SEZs have
become a real symbol of the national nature of the CPEC, with one being
promised for every province of Pakistan. This is real concrete evidence,
claim the CPEC supporters, that the benefits from the CPEC will spread
out to encompass all of Pakistan. The problem with these claims is that they
are based on a corner of the CPEC that as yet only exists in the dreams
of economic planners and politicians. Much of what passes for analysis of
the likely impact of the SEZs is simply the economics of aspirational hope,
not rigorous thinking rooted in evidence and theory. This chapter outlines
the theoretical and empirical benefits of the SEZs as experienced in other
countries and in non-CPEC SEZs in Pakistan, what is promised for Pakistan
and how scholars have written about SEZs so far, and the lessons we can draw
from the experience of both SEZs in China and from Chinese-invested SEZs
in Africa.
THE BENEFITS AND COSTS OF EXPORT

PROCESSING ZONES (EPZs) AND SEZs
An export processing zone (EPZ) is a ‘geographically bounded zone in which
free trade, including duty free import of intermediate goods is permitted
provided all goods produced in the zone are exported’. An EPZ is an isolated
and cut-off corner of the global economy. An SEZ is also a geographically
bounded zone but lacks this exclusive export orientation. Firms receive
various incentives to locate in the zone, often via FDI, and may export or
sell their output onto the domestic market. The purpose of an SEZ is to
induce the global economy to set up and integrate with the local economy.
114
THE DRAGON UNCOILS
SEZs first appeared in Puerto Rico (1951) and then Shannon Airport in
Ireland (1959) but the main success story was China in the 1980s. By 2008
there were some 3,000 SEZs globally and by 2015 their number reached 4,300
across 130 countries. It has been frequently argued that SEZs played a crucial
role in promoting economic growth in 1970s East Asia and in initiating the
opening up and marketisation of the Chinese economy after 1979. Inspired
by these stories of success, 24 SEZs had been established across Africa by
2009. Most of these were unsuccessful and some had even been abandoned
or suffered from a dramatic decline. In the Dominican Republic, employment
in free zones reached 200,000 by 2000 which represented about 35 per cent
of the national manufacturing employment. This success did not last and
employment declined to 120,000 over the 2000s.
The optimistic view of SEZs is that they boost investment, particularly
through FDI. In economies that are largely agrarian or based on mineral
exports SEZs can pioneer manufacturing investment and boost employment,
and so can initiate the process of structural transformation. Deng Xiaoping
once said that an SEZ for China was a window to the outside world for
China. Indigenous firms may have little experience of exporting, using
modern technology and good management practices. The entry of an MNC
may provide a learning-catalyst effect. Encouraging foreign firms located in
an SEZ to source materials, inputs and components locally can help promote
wider domestic industrialisation. It may, however, take time to bring local
producers up to the level of quality, speed and consistency in production that
the foreign investor is used to from existing suppliers. The pessimists point
out that domestic investment in an SEZ may simply be a diversion effect
with investment relocating from elsewhere in the domestic economy to take
advantage of tax concessions. An SEZ may also end up as a low-wage assembly
operation utilising imported inputs with little local value added and so offer
few opportunities for learning ( Johansson and Nilsson 1997). As the world
economy more generally has been opening up to freer trade and investment
since the 1980s, the whole world has become more like a super-SEZ. In China
the initial small SEZs lost their pulling power as more SEZs were established
across China as the country pursued reforms (Wong 1987). In 1984 14 major
cities on the east coast were opened up as SEZs that encompassed 160 million
people. By 2008 92 per cent of the municipalities in China had SEZs (Wang
2013). In 2001 China joined the World Trade Organization (WTO) and the
whole country converged towards international openness (Yeung, Lee and
Kee 2009).
115
Evidence from 11 countries who established EPZs between 1980 and 1992
shows in general a strong and positive relation between EPZs and aggregate
exports, but the relationship was very diverse by country. It was positive and
significant for Hong Kong, Malaysia, Mauritius, Singapore and Sri Lanka
but had no impact in Egypt and the Philippines. The effect was more likely
positive in those countries that had an outward-oriented trade policy.
In Mexico, where the policy regime outside the EPZ remained one focused on
import substitution, the EPZ had no impact on aggregate exports. In Malaysia
the first EPZ was established in 1971 to promote a wider shift away from
import substitution. The EPZs had access to good infrastructure, were close
to large cities and were embedded in a favourable business environment. They
were widely perceived to be a success and by 1982 EPZs accounted for 52 per
cent of total Malaysian exports ( Johansson and Nilsson 1997). The creation
of backward spillovers varies widely between EPZs. In South Korea there
were tight links between EPZs and local manufacturing and an ever-growing
proportion of the value added as inputs, equipment and raw materials were
sourced from Korean firms. Net exports (a measure of how import-dependent
the EPZs were) varied widely in the early 1990s, from 51 per cent in South
Korea and 53 per cent in Indonesia to only 20 per cent in Bangladesh, 22 per
cent in Jordan and 26 per cent in Costa Rica (Schrank 2001: 232). Over time
the SEZs in East Asia led the climb up the value chain. Over two decades,
labour-intensive industries fell from 40–50 per cent of the turnover in South
Korean and Taiwanese zones to about 10 per cent in the mid-1990s, and by
then technology-intensive industries contributed over 80 per cent of turnover
(Brautigam and Tang 2014).
The adoption of the Indian SEZ Act in 2005 was spurred by the Chinese
example. In the years up to 2011, SEZs were widely perceived to have failed to
emulate the Chinese experience. Alkon (2018) finds that between 2006 and
2010, SEZs had no impact on 29 development indicators in the nearest village.
These indicators included power supply, mobile phone coverage, public bus
service, national highway, post office, treated water, presence of schools and
banking services. SEZs had been used by local politicians to manipulate the
allocation of resources to generate electoral support by targeting benefits on
the basis of caste and ethnicity. Politicians were also more interested in using
SEZs as part of real-estate speculation than promoting long-term economic
development (Alkon 2018).
116
THE DRAGON UNCOILS
In China SEZs started with labour-intensive processing and assembly

activities. Chinese SEZs never upgraded like those in South Korea and it is
still rare to find technology-intensive companies in the SEZs. There are some
exceptions, such as the telecommunication firms Huawei and ZTE who have
opened up production in the Shenzhen SEZ. China was very successful at
utilising the SEZs to promote domestic spillovers. An important strategy used
in China was to study the value chain of foreign investors located in a Chinese
SEZ, then attract related foreign and domestic suppliers to move into the
zone as well. After Motorola opened up production in the Tianjin Economic-
Technological Development Area (TEDA) zone, a number of its suppliers
also set up factories in the zone. Chinese SEZs have helped promote spillovers
through the use of targeted training. SEZ companies have usually provided on-
the-job training for local employees and provided opportunities for Chinese
managers to learn from experienced foreign managers. Sometimes this training
included overseas visits to parent companies by local Chinese employees.
Requirements for training were often written into the initial agreements when
MNCs invested in the SEZs. China also encouraged managers and skilled
employees to start their own firms, often initially as subcontractors, which has
helped diffuse skills acquired in the SEZs throughout the Chinese economy.
The Chinese assigned senior managers from state enterprises to work in joint
ventures in the zones for a period, and then moved them to other Chinese state
enterprises. In 1994 China established a joint economic zone with Singapore
in Suzhou in order to learn from Singapore’s expertise in zone management.
By 2008 more than 2,000 Chinese officials had been trained in Singapore
while about 20,000 officials from different parts of China had visited the zone
each year to study the ‘Suzhou model’ (Brautigam and Tang 2014).
SEZs AND THE CPEC IN PAKISTAN

Pakistan announced its own SEZ Act in 2012 and amended it in 2016 to
promote SEZs. The Act instructs that each province of Pakistan establish a
Provincial Investment Promotion Authority (PIPA) to assist each provincial
SEZ authority and facilitate zone development. The SEZ authority would
be responsible for the provision of infrastructure and utilities such as gas,
electricity, water and waste disposal. The new administrative structure is
supposed to facilitate and gradually move to single-window clearance for
117
SEZ investors. This would require the provincial government to either delegate
authority for implementing labour, environmental laws and tax collection to
the SEZ authority or depute representatives of the provincial government to
the SEZ authority.
The Act allows one-time exemption from customs duties and taxes on
imports of plant and machinery into SEZs for installation in that zone
enterprise (saving on average 22 per cent of the cost). SEZ enterprises are
given exemption from all taxes on income for five or ten years depending
on when they established commercial production. Other incentives include
cheaper credit for project financing, the ability to pay for land plots in
instalments and a freight subsidy at 50 per cent on the inland transportation of
plant and machinery for installation in any priority SEZ. SEZs are exempted
from all provisions of the Foreign Exchange Regulation Act and so allow
free repatriation of capital and profits. Obsolete or old machinery can be sold
into the domestic market of Pakistan after payment of applicable duties and
taxes. The Act was flexible and offered discretionary extra incentives if those
could be justified on the basis of an economic impact assessment. According
to the Act, the same law for all labour and employment shall be applied in the
SEZs as already exists in Pakistan (Abbas and Ali 2017a; Sherdil 2017;
HKTDC 2018b).
There is no doubt about the collective wisdom of the Government of
Pakistan; it has thoroughly imbibed all that is theoretically and empirically
positive about SEZs and is adamant that all those economic blessings will
be enjoyed by Pakistan. There are nine proposed SEZs under the CPEC,
including one in each province and one each in Gilgit-Baltistan, Kashmir,
Federally Administered Tribal Areas (FATA) and Islamabad. The Government
of Pakistan is clear that the CPEC will lead to local industrialisation and
will ‘[p]romote the quality, value addition, competitiveness and efficiency
improvement of the textile and garment industry, expand the size of the textile
industry, and increase the supply of high value-added products’ (Government
of Pakistan 2017: 16). There is no indication of the government using any
rigorous research on which to base these claims.
There is research on these SEZs but it is suffused with more hope than
rigour. A typical study of the CPEC SEZs is that of Abbas and Ali (2017a)
who list all nine proposed SEZs, describe how close they are to major urban
areas and transport links and give detailed lists of their resource endowments
in order to emphasise their potential for industrial growth. For example, the
118
THE DRAGON UNCOILS
Rashakai Economic Zone in Khyber-Pakhtunkhwa (KPK) is located in an

area that produces wood, tobacco, wheat, maize, rice and sugar beets and
has natural resources such as oil, gas, mines of marble, gem stones, emeralds,
uranium, electricity, salt, lithium and steel. The proposed SEZ is 65 kilometres
from a dry port, 25 kilometres from a railway station, 15 kilometres from a
city centre and is next door to a highway. The hope-plan is that processing
industries will emerge in the SEZ to utilise these endowments and transform
them into industrial output. No mention, for example, of the long list of
‘other constraints’ on manufacturing described in the section titled ‘“Common
Prosperity”: Will the CPEC Have a Transformative Impact?’ in Chapter 4. This
pattern of description, hope and assertion is common to other studies. Ahmad
notes that the northern zone of the CPEC due to its seasonal and elevation
advantages will produce higher-value crops such as apricots and grapes. Under
the CPEC, a fruit processing industry will be set up in Gilgit-Baltistan and
‘exports of fruit will get a boost as lead growing market in the region’ (Ahmad
2017: 93). He continues with an unbridled faith in Pakistani policymaking—
that government aspiration will inevitably translate into practical outcomes.
He writes that the fruits to be promoted include apricots in clusters/districts
such as Chanhe, Hunza, Nagar and Ghizer, which produces over 100,000
tonnes, representing more than half of national production, and also grapes in
Gilgit, Diamer, Astore and Ghizer, where more than 60 per cent of the national
production of almost 4,000 tonnes takes place. He argues that delivering the
benefits of the CPEC will require upgrading of infrastructure in regions along
the CPEC, and investment to provide local connectivity through feeder roads
to improve market access to the remote mountain regions and in creating
satellite markets with modern storage facilities. Other agricultural products
include mangoes, guavas, potatoes and onions in Punjab, dates and bananas in
Sindh, peaches and tomatoes in KPK and peaches, citrus, strawberries, apples,
melons and apricots in the Peshawar region (Ahmad 2017: 98). The method
in this study is to describe resource endowments, herald the coming of the
CPEC (and some complementary infrastructure) and jump to the assumption
that those endowments will then translate into output growth.
More optimism suffuses the work of Mehmood (2017), who writes
about what he labels a ‘win–win proposal’ to establish the CPEC SEZs.
Two problems are apparent from his optimistic discussion. The first is his
mention that there are now about 3,000 SEZs across 135 countries. Vietnam
alone has 326 industrial zones and EPZs where 7,500 foreign-invested
119
projects have contributed to a total FDI of $125 billion (Tang 2019: 7).
One may reasonably ask, why will investors choose to invest in those in
Pakistan rather than the global alternatives? The second problem is that SEZs
are not new to Pakistan. Mehmood (2017) notes that virtually every district
headquarters of Pakistan has an industrial estate or area offering infrastructure
and other incentives. During the 1970s, Pakistan established around
100 industrial estates to revive sick industries; they were failures. Punjab today
has 26 industrial estates, Sindh 30, Balochistan 7 and KPK 12. Many of these
are unsuccessful, have been established in remote areas and lack the necessary
skilled workforce or amenities for workers. In India, as a point of comparison,
there are approximately 430 registered SEZs; of these only 218 are operational
in some capacity, and of those working less than half are anywhere near capacity
(Daily Times 2019). There are also examples of extremely successful clusters in
Pakistan that are integrated into the global supply chains, including sports and
surgical clusters in Sialkot, ceramic pottery in Gujrat, textiles in Faisalabad,
fan cluster in Gujrat and engineering in Gujranwala. The football cluster in
Sialkot developed in the colonial era in response to the local endowment of
relevant skills and raw materials and local demand for footballs from a British
military garrison, not to special government incentives to promote the sector
(Atkin et al. 2015). There is no analysis why these newly proposed SEZs will
replicate the successful rather than the unsuccessful existing clusters.
One partial exception is Zia, Malik and Waqar (2017), who review the
operation of African and Asian SEZs to draw lessons for the operation of
SEZs in Pakistan. In practice, the study is based on very little research, the
entire African experience being captured, for example, from two studies by
Farole. This paper has no methodology and no discussion of how or why
Pakistan should be able to learn from the successful experience of SEZs in
Bangladesh, Cambodia or the Philippines or avoid the mistakes and failures
of SEZs in Africa. Despite these weaknesses, the conclusions are instructive
for Pakistan. They argue that the benefits from SEZs take up to 5–10 years
to emerge, that the host country needs to bear much of the cost of investing,
that tax holidays (a prominent part of the incentive structure in Pakistan)
have little impact and that SEZs need a determined developmental effort to
ensure that foreign investors purchase locally produced inputs (which is not a
planned part of the SEZs in Pakistan).
Looking backward to the experience of SEZs in other countries to make
a pronouncement about the likely future success of SEZs in Pakistan is
120
THE DRAGON UNCOILS
problematic. As noted above, as SEZs have spread and the world outside SEZs
has itself become more open, the bite of SEZ incentives has diminished.
On top of that, Pakistan has one of the poorest incentive structures in the
entire region. A regional comparison shows that it is more expensive to install
and maintain industries in Pakistan as compared to China, Vietnam, Sri Lanka
and Bangladesh (Sherdil 2017: 74). Other countries, for example, offer leases
on land for up to 40 years. For foreign investors, China also offers construction
land that is exempt from land-use fees for five years, then the fees are halved
for another five years (Sherdil 2017).
CHINESE SEZs AND REFORM AFTER 1979

We can learn enough about the working of Chinese SEZs after 1979 to be
reasonably convinced that Gwadar or Pakistan more widely will not become
‘a new Shenzhen’.
In 1979 China opened four SEZs, in Shenzhen, Zhuhai and Shantou in
Guangdong and Xiamen in Fujian. These SEZs offered incentives including
easy access to independent local planning authorities, tax breaks, free/low
duties on imported equipment and production materials, free- or low-rent
business buildings, flexibility in hiring and firing workers, depreciation
allowances, negotiated but limited access to the domestic Chinese market for
goods produced in the SEZ, and residence and work permits for foreigners
(Yueh 2013). Chinese SEZs were broader than EPZs; they were not just about
exporting but about the comprehensive development of a particular region
and so also included efforts to build infrastructure and a strong emphasis on
forming joint ventures to promote learning from foreign management and
the acquisition of foreign technology for the benefit of domestic firms (Wong
1987; Ge 1999). The initial aim was to attract investment in low-tech, light
industry. There was a later effort, after 1995, to establish high-technology
development zones (HTDZs). Here each zone was to include a university-
based research centre, an innovation centre to utilise applied technology for
product development, and a partnership with a commercial enterprise to
manufacture and market the products (Yueh 2013).
In 1978 Shenzhen was to some a ‘fishing village’ and to others a ‘small
border town’ of 20,000 people. When Shenzhen was designated as an SEZ
in 1979, it had only a few small industrial factories with a total output of
$10,000. By 1981 more than half of all inward Chinese FDI came to Shenzhen
121
(Wong 1987). Employment growth averaged 25 per cent per annum between
1980 and 1993. Labour-intensive manufacturing comprised about 75 per cent
of the total in 1980 and its share declined to 54 per cent by 1995 (Ge 1999).
The role of the state in production was surprisingly important. By 1993 state
enterprises still accounted for half of all employment (Ge 1999).
By the end of 1998, Shenzhen had seen $12 billion or more of FDI and
its population had grown to 3.5 million people. An annual economic growth
of 32 per cent during those 20 years saw the share of the primary sector in
output decline from 37 per cent to 1.2 per cent and the share of the secondary
sector rise from 20 to 50 per cent. Growth was initially in labour-intensive
light industries such as clothing, toys, shoes, packing and bicycles. Over
time, there was a gradual shift to higher technology output and FDI was
increasingly sourced from developed country firms rather than from those
in Hong Kong (Wei 2000). By 1998 40 per cent of Shenzhen’s industrial
output was high technology and the town boasted 14 per cent of the world
output of floppy disks, 6.2 per cent of personal computer motherboards and
8 per cent of hard drives (Wei 2000). Even as late as 2008, Shenzhen was still
experiencing 12 per cent annual GDP growth. By 2008 Shenzhen ranked
fourth in global rankings as a container port; for 16 years in a row, Shenzhen
had been the number one Chinese city for international trade and it had the
highest per capita income of any Chinese city. In 2007 high-tech industries
were producing output worth $111 billion, of which $80 billion was exported
(Yuan et al. 2010).
Some more excitable claims in the media suggest Gwadar is ‘on route to
becoming a replica of Shenzhen’ (Global Times 2018a), that ‘local officials
[in Pakistan] dream of a future where Gwadar becomes a second Shenzhen’
(Guardian 2016) and that the Government of Pakistan has established its nine
SEZs ‘inspired by the miracle of Shenzhen’ (Dawn 2017).
There are indeed some superficial similarities between Shenzhen–China
in 1979 and Pakistan–China in 2019. In 1979 wages in Shenzhen were only
10 per cent of those in Hong Kong which motivated firms to relocate in search
of lower-wage labour. This pattern is broadly true of contemporary Pakistan.
Comparative estimates of labour costs in textiles show them to be consistently
lower in Pakistan than in China and other competitors. In 2000, for example,
the average hourly wage (US$) was 1.8 in Mexico, 0.9 in China, 0.7 in India
and only 0.2 in Pakistan and Bangladesh (Tewari 2005: 28). Another estimate,
this time of ‘total textile industry operator costs’ per hour (US$) in 2000,
122
THE DRAGON UNCOILS
was 1.18 in Thailand, 1.13 in Malaysia, 0.69 in China, 0.58 in India and 0.37
in Pakistan. Only Indonesia at 0.32 was lower than Pakistan (Palpacuer, Gibbon
and Thomsen 2005: 416). In 2017 the International Labour Organization
(ILO) estimated wages in Pakistan to be twice those in Bangladesh ($121
versus $68 per month) but lower than India ($215), Vietnam ($280), Indonesia
($320) and China ($480) (Pakistan Business Council 2018: 39). This gap has
been widening in recent decades. Between 1999 and 2007, real wage growth
was about 1 per cent per annum in Pakistan and 13 per cent per annum in
China (International Labour Organization 2012). Between 2008 and 2017,
real wages grew by 5.5 per cent per annum in India, 4 per cent in Sri Lanka,
3.4 per cent in Bangladesh, 8.2 per cent in China and only 1.8 per cent in
Pakistan (International Labour Organization 2018: 122). In Bangladesh,
there are growing shortages of young women who are able to enter the
labour force. There is good evidence that real wage growth in Bangladesh
has accelerated since the early 2000s as ‘the supply of seemingly unlimited
labour was exhausted, the terms of trade in the labour market started to shift
in favour of workers, leading to a tightening labour market and an increase in
agricultural wages’ (Zhang, Rashid and Ahmed 2014: 274).
The problem for Pakistan and a crucial reason why Pakistan is not likely
to replicate Shenzhen–China in 1979 is that wage costs are only a fraction of
the total costs of production. Collier (2007) argues that the share of labour
costs in total costs of production in labour-intensive goods could be as low
as 16 per cent. This implies that, if wages in Pakistan were only 6 per cent of
those in China, this would yield only a 15 per cent cost advantage. The data
above shows that Pakistani wages were around 25 per cent of those in China.
For a fuller picture of costs in Pakistan, we do need to consider other costs,
such as transport, law enforcement, corruption, electricity, and the availability
of skilled labour. There is more extensive comparative cost evidence available.
The rather detailed Table 5.1 tries to quantify all the costs going into making
a simple men’s T-shirt. Notably, the fabric cost per kilogram (locally sourced)
in Pakistan at $2.89 per kilogram is significantly lower than the $3.37 in
Bangladesh and Cambodia, both of which source fabric from China. Overall,
the total cost of producing a men’s T-shirt in Pakistan is 7 per cent lower than
in Bangladesh and 19 per cent lower than in China (Nathan Associates 2009:
1–2). Other evidence confirms that electricity costs, particularly from gas or
captive power plants, are higher in Pakistan than its competitors (Pakistan
123
Table 5.1 T-shirt making costs
Garment making India India China Pakistan Bangladesh Cambodia

up by country
Fabric source India China China Pakistan China China
Main fabric
Fabric cost per 3.019 3.336 3.336 2.894 3.336 3.336
kilogram (US$)
Fabric shipping 0 0.069 0 0 0.06 0.069
cost per kilogram
(US$)
Fabric usage 0.235 0.235 0.235 0.235 0.235 0.235
per garment
(kilogram)
Fabric waste 5 5 5 5 5 5
(short pieces, end
of rolls, faults)
per cent
Main fabric cost 0.71 0.801 0.784 0.68 0.798 0.801
per garment
(US$)
Trim cost per garment (US$)
Thread 0.045 0.45 0.045 0.045 0.045 0.045
Labels, tags 0.037 0.037 0.037 0.037 0.037 0.037
Packaging per garment (US$)
Plastic polybag 0.018 0.018 0.018 0.018 0.018 0.018
Cardboard box/ 0.06 0.06 0.06 0.06 0.06 0.06
carton
Total materials 0.87 0.961 0.944 0.84 0.958 0.961
cost per garment
Labour hour $ 0.83 0.83 1.44 0.55 0.32 0.335
cost in making
up
Standard 6.12 6.12 6.12 6.12 6.12 6.12
minutes per
garment (cut,
make, trim,
finish)
Efficiency 25 25 15 30 50 70
adjustment
(Contd )
124
THE DRAGON UNCOILS
(Contd )
Labour cost per 0.106 0.106 0.169 0.073 0.049 0.058
garment (US$)
Reject garments 0.029 0.032 0.033 0.027 0.03 0.031
3 per cent
Manufacturing 0.026 0.026 0.042 0.018 0.012 0.015
overhead per
garment (25 per
cent on labour)
(US$)
Inclusive of
electricity, rent,
indirect labour
Sales and 0.011 0.011 0.017 0.007 0.005 0.006
administration
costs (10 per cent
on labour)
Total cost per garment—fabric, labour, overhead (US$)
Sales and 1.042 1.136 1.206 0.966 1.055 1.07
administration
Agent fees per 0.042 0.045 0.048 0.039 0.042 0.043
garment (4 per
cent on total
cost)
Factory gate cost 1.084 1.181 1.254 1.005 1.097 1.112
per garment
Shipping and insurance to Long Beach, CA (twenty-foot equivalent unit, or TEU)
Land transport 400 400 470 300 250 600
cost to port
(US$)
Ocean freight Mumbai Mumbai Shanghai Karachi Dhaka Sihanoukville
from X to Long
Beach
Ocean 2,100 2,100 1,800 2,000 1,900 1,900
transport cost
per container
including
insurance (US $)
Units per 18,000 18,000 18,000 18,000 18,000 18,000
container
(Contd )
125
(Contd )
Transport and 0.139 0.139 0.126 0.128 0.119 0.139
insurance cost
per garment
(US$)
Total cost 1.223 1.32 1.38 1.133 1.216 1.251
per garment
including cost,
insurance,
freight (c.i.f.)
(US$)
Tariff per cent 16.1 16.1 16.1 16.1 16.1 16.1
(import duty to
US)
Tariff per 0.197 0.212 0.222 0.182 0.196 0.201
garment (US$)
Quota cost per 0 0 0 0 0 0
garment (US$)
VAT percentage 12.5 12.5 17 15 15 10
VAT applied 0.15 0.16 0 0 0.18 0
(US$)
Cost per 0.35 0.377 0.222 0.182 0.378 0.201
garment—tariff,
quota and VAT
(US$)
Full landed cost per garment duty paid (US$)
Timescale (Weeks)
Fabric 4 4 4 4 4 4
production time
Delivery time 1 3 1 1 3 3
fabric
Making up 2 2 2 2 2 2
operations
Garment 5 5 3 5 5 4
shipping time
Total delivery 12 14 10 12 14 13
cycle
Source: Nathan Associates (2009: 1–2).
126
THE DRAGON UNCOILS
Other factors militate against any sensible effort to equate Pakistan and
Shenzhen. Hong Kong was only an hour away by road from Shenzhen (Wei
2000). By comparison, there is a long distance and inhospitable route across
the mountain separating Pakistan geographically from Kashgar in Xinjiang.
The closeness between Hong Kong and Shenzhen–Guangdong was also
evident in terms of language and culture. The province of Guangdong, in which
Shenzhen was located, has a long history of foreign trade and connections and
in 1948 accounted for 20 per cent of China’s foreign trade. The province was
also home to over 80 per cent of overseas Chinese. Shenzhen and Hong Kong
spoke the same language and were both closely linked to the same international
diasporic community (Zhang 1994). There are immense cultural, religious and
linguistic differences separating Islamic Pakistan from communist China.
The SEZs in China were combined with significant empowerment of and
decentralisation to local governments to run them. Moburg and Tarko (2014)
find that decentralisation of SEZ governance to local authorities is crucial
in explaining successful outcomes. Competition between SEZs to attract
investment and then promote internal SEZ growth was crucial in pushing
local governments towards pro-business policy and liberalising reforms.
In China, local governments were provided with the legal and administrative
authority to manage an SEZ without undue interference from higher levels of
government. Local governments were empowered to collect most of the taxes,
to administer resident state enterprises and control credit allocation by state-
owned banks. They could formulate laws to manage the SEZs and were able
to make decisions about the functioning of joint ventures by 1983 and with
wholly foreign-owned investments by 1988. Even a township government in
Guangdong was authorised to approve FDI. The existing tax system gave local
governments a strong incentive to promote the growth of SEZs. Guangdong
had to pass on a relatively low and fixed sum to the central government each
year whereas other provinces had to hand over a percentage of taxes collected.
Guangdong was able to retain 70 per cent of any export earnings above the
1978 level. Local governments were allowed to set up organisations in Hong
Kong and Macao for trade promotion and information-gathering (Zhang
1994). The combination of these strong incentives and real local administrative
freedoms gave the local government a powerful motivation to promote the
Shenzhen SEZ. The local state had the capacity to translate this motivation
into practical developmentalism. The role of the local state was important in
127
providing a local development strategy that sought to overcome the shortage

of local technicians and engineers. The local government set up firms to supply
skilled migrant labour to firms. The state also supplied a quality assurance
centre, state-funded laboratories, consultancy, information on technology for
local firms, and made some efforts to control intellectual property. The SEZ
required massive investment in infrastructure and improvement in governance
to lure FDI into manufacturing, which had initially been more cautious and
focused on real estate and tourism. The state emphasised the importance of
joint ventures to help acquire skills in production and in meeting international
orders (Wei 2000). The 18th Amendment to the Constitution of Pakistan
passed much power from the central state to the provinces. While the SEZ
Acts have talked about the need to decentralise further, to the level of the
SEZ developer, there have been no such reforms of the local state, nor the
empowerment of those bodies to run the SEZs. The SEZ programme remains
a centrally dictated one in Pakistan.
SEZs, even in China, had significant costs for the domestic economy.
The headline figures for exports from the SEZs were impressive. Between
1979 and 1994, the value of exports from Shenzhen increased from
$9.3 million to more than $18 billion, showing 75 per cent per annum growth.
However, the average import content of Shenzhen exports was around 80 per
cent. It was higher in other SEZs and, apart from Shenzhen, all the SEZs
were net users of foreign exchange. This was due to a combination of very
high levels of import dependent production and the fact that the SEZs were
able to import consumer goods (particularly electronic goods and clothing)
that were then diverted onto the wider Chinese domestic market. By 1985
the four SEZs had a combined trade deficit of $900 million. The high levels
of tax exemptions reduced the reduced revenue raised by the government
at the same time the state was financing large-scale infrastructure spending
(Sklair 1991). The ultimate gain of the SEZs was that they launched reforms
in China that were then implemented elsewhere. Shenzhen pioneered a new
labour system. Before 1979, urban areas in China were characterised by a
residence and workplace that was fixed for life; the vast majority of workers
had little job mobility. State enterprises were characterised as having excessive
employment. Together, these features of the labour market were known as the
‘iron rice bowl’ of lifetime employment. Contract labour was first used in
1982 in joint ventures and solely foreign-owned firms, and later extended
128
THE DRAGON UNCOILS
to state enterprises. This gave firms in Shenzhen flexibility in hiring and firing,
bonus payments and wages. In China-wide SEZs, contract labour increased
from negligible levels in 1978 to 10 million employees by 1990 (Sklair, 1991).
The proportion of temporary migrants in the population of Shenzhen
increased from 1 per cent in 1979 to 72 per cent in 1994. The majority of
migrants were young and single, with 42 per cent in the 15–24 age group.
These often-unskilled contract migrant workers lived in company-managed
dormitories, typically rented from zone authorities, with four–six people to a
room. This helped facilitate migration, particularly of young women, reduced
costs and increased company control over their labour force (Liang 1999).
In 1979 there was no private land in China, meaning land could not be
bought, sold or leased. The Shenzhen SEZ allowed land to be leased, and
in 1987 a lease for 50 years in Shenzhen was sold to a local public company.
The abundant land around Shenzhen was allocated cheaply, with around
90 per cent of land allocated for free. Rising property prices and government
allocation at low cost presented massive opportunities for corruption. Much
of this land was retained for speculative purposes and around one-third of
the allocated land remained idle (Zhu 1994). Other reforms that spread from
Shenzhen included commercialised housing for employees, price reform,
quota allocation of basic necessities, the presence of foreign banks and an
end to state monopoly in finance, the first stock exchange and FDI into state
enterprises (Yuan et al. 2010).
Viewed in isolation, SEZs may have induced inefficient resource
allocation but they provided a useful transition path to a market economy
by demonstrating success on a small scale and allowing reform to gradually
ripple outwards into the rest of China (Litwack and Qian 1998). This story
has no relevance for Pakistan. Pakistan started liberalising its economy in the
1980s and today has a market economy. There is no question of using SEZs
as a means to spread the market. Rather, the SEZs are more likely to become
ensnared in the existing poor functioning of Pakistan’s market and go the
same way as most of the other attempts to launch SEZs, industrial estates,
clusters and EPZs in Pakistan. Table 7.3 in Chapter 7 presents data from the
World Bank Doing Business Index. In 2019, according to this index, Pakistan
ranked 130th among 190 countries in terms of the ease of starting a business,
166th in dealing with construction permits, 173rd for paying taxes, 156th for
enforcing contracts and 167th for getting electricity (World Bank 2019a).
129
CHINESE SEZs IN AFRICA

There has long been a view that SEZs have failed in sub-Saharan Africa
(SSA). ‘With the exception of Mauritius and the partial initial success of
Kenya, Madagascar, and Lesotho, most African zones have failed to attract
significant investment, promote exports, and create sustainable employment’
(Farole and Moberg 2014: 3). The reasons seem clear enough and include
poor locations, poor planning, excessive red-tape, insufficient investment
to connect the SEZ to global or regional markets, policy instability and
political commitment (Farole and Moberg 2014). Why have these problems
persisted over 40 years when these reasons are now well documented and well
understood? There are deep political economy problems that have locked
in a cycle of failure; these include a tendency to use SEZs for the personal
enrichment of well-connected politicians or their political constituents (land
grabbing, construction and contracts) rather than to promote wider economic
development, and also a focus on the hardware (infrastructure), which offers
immediate gains, rather than the longer and more complicated software
(legal and regulatory environment) (Farole and Moberg 2014). Has Chinese
engagement in the SEZs in SSA (and Pakistan) made a difference?
In 1993 China became a net petroleum importer and soon after crude oil
represented 70 per cent of African exports to China. This led to what some
called ‘A New Scramble for Africa’. Total African exports to China soared,
rising from $5 billion in 2002 to more than $60 billion in 2010. Trade with
Angola alone increased from $1.1 billion in 2002 to $25 billion in 2008; by
then Angola was supplying 17 per cent of China’s total oil imports (Ovadia
2013). This changing pattern of imports was supported by Chinese FDI.
Between the mid-1990s and 2000, FDI primarily comprised efforts
to control infrastructure and natural resources, mainly by Chinese state
enterprises. According to one estimate, in 2011 51 per cent of the $68 billion
outward FDI was targeted towards natural resources. Africa held key deposits.
South Africa alone held 75 per cent of the world’s manganese reserves and
massive deposits of gold, copper and nickel. There were new oil discoveries in
Ghana, Uganda and Chad and there was copper in Zambia, the Democratic
Republic of the Congo (DRC) and Angola. China provided loans for
infrastructure with low interest rates, 5–7 years’ grace before repayment began
and 15–20-year repayment periods. The lending was often tied to the use of
services, goods and labour from China. The loans were secured against the sale
of specified commodities. Angola received $4.5 billion of loans by 2005–06,
130
THE DRAGON UNCOILS
which some estimates place at up to $14 billion in return for commodity,

mainly oil exports. Between 2007 and 2008, the DRC received $9 billion,
backed up by copper exports. Nigeria under President Obasanjo received
$12 billion of infrastructure (and some manufacturing investment) for
oil exports. This investment was concentrated in a few sectors of strategic
interest to China, especially in extractive industries, and done via state-owned
enterprises and joint ventures (Oyeranti 2010; Ovadia 2013; Alves 2013).
This early engagement generated a controversial debate. The US Council
on Foreign Relations charged that China was protecting repressive or corrupt
governments, undermining western efforts at improving human rights and
was engaged in unfair competition with US firms in contract bidding. Giving
aid to such countries is something that the US and European aid donors
were long criticised for, so there was nothing new here. Even the much-
criticised Chinese oil deal with Angola eventually led to Angola becoming
more transparent in its public accounts and providing audited financial reports
which together helped Angola to resume ties with the IMF in 2009. Here was
also a view that Chinese lending was entering countries in the space vacated
by debt-forgiveness made by developed countries. This meant that China
was free-riding on debt-relief from other countries. The Chinese Ministry
of Foreign Affairs argued instead that this was a strategic partnership and
represented win–win cooperation. African countries, they suggested, had
little or no foreign exchange or a sufficient credit rating to borrow with
which to repay loans. There was no other realistic way to create desperately
needed infrastructure in transport, health, education, agriculture and fisheries.
This engagement was sustainable because it was not contingent on ‘good-will’
from China. Sustainable goodwill from western Europe and North America
has been promised since 1970, to give aid totalling 0.7 per cent of the GDP,
and 50 years later remains just a promise. Rather, this was sustainable because
it was also a win for China, who in return received the resources it needed to
support domestic economic growth and could also win new contracts for its
construction companies. Throughout the 2000s, China was a net importer of
chromium, cobalt, copper, iron ore, manganese, nickel, petroleum, platinum
metals and potash. By 2010 mineral commodities accounted for $375 billion
or 64 per cent of Chinese imports, up from $40 billion a decade before (Ovadia
2013; Alves 2013). The sustainability of this engagement was made apparent
when, as the developed world drew back from foreign investments in the wake
of the 2008 Global Financial Crisis, such deals were expanded by China.
131
There is also little reason to think that China was anything but genuine
in proclaiming this as a win–win engagement. The infrastructure for
commodities exchange can be understood in terms of China’s own recent
history of rapid economic growth and escaping mass poverty. China perceived
itself to be exporting a model of development that had worked well in its own
recent history. In the 1970s, China was an agrarian economy with massive
reserves of natural resources such as oil, gold, and copper and little in the way
of foreign exchange to pay for the infrastructure needed to start exporting.
Japan began to import oil from China in 1973. In 1978 Japan signed a long-
term agreement to provide low interest loans to finance exports of $10 billion
of modern plant, industrial technology and materials and China paid by
exporting oil and coal. Fluor, a US-based engineering corporation, in 1978
constructed a copper mine in Jiangxi province. Bethlehem Steel and the
German Thyssen Company followed. In 1983 China awarded contracts to
five oil companies from Canada, Britain, Australia and Brazil. Each of these
deals involved an infrastructure-for-commodities exchange. China in the
1980s had little foreign exchange and these deals allowed import and delayed
payment until China could export the goods produced (Brautigam 2009: 46).
There were reasonable concerns; commodity-based production and exports
have traditionally been associated with generating problems of governance,
difficulties with macroeconomic management and corruption (Zafar 2007).
The exclusive infrastructure–commodities relationship did not last long.
Between 2000 and 2009, private companies became more engaged in FDI
and the emphasis shifted to manufacturing (Gu 2009). Between 2006 and
2010, China registered 293 projects in Ghana, 35 per cent of which were in
manufacturing, 32 per cent in general trade, 15 per cent in services, 9 per cent
in tourism and only 8.5 per cent in agriculture (Tang and Gyasi 2012). This
evolution still had the stamp of Chinese state policy intervention. There were
a range of general incentives for domestic Chinese manufacturers to shift their
production to Africa. More than 400 products exported from Africa were
granted tariff-free entry to China, including a wide variety of manufactured
goods such as motor vehicles, spare parts, diesel generators, gardening tools,
knit clothing and leather wallets (Brautigam and Xiaoyang 2011; Brautigam
and Tang 2014). As part of the 11th Five-Year Plan in 2006, China announced
it would establish 50 SEZs overseas. The intention was to help Chinese
companies to restructure and move labour-intensive capacity overseas.
Chinese engagement with Africa had diversified noticeably. China established
132
THE DRAGON UNCOILS
Confucius Institutes across Africa. There is no evidence to link these institutes

systematically to the presence of natural resources or wider aid flows from
China. The Confucius Institutes appear to be about increasing China’s general
economic and political influence, her soft power (Akhtaruzzaman, Berg and
Lien 2017).
Chinese companies were slated to take the lead in developing these zones.
By 2010, Chinese officials had authorised support for 19 zones worldwide,
6 of these zones being built in Africa (Egypt, Ethiopia, Algeria, Mauritius,
Nigeria and Zambia) and most focus primarily on manufacturing (Brautigam
and Tang 2014). The Ministry of Commerce held two rounds of tenders.
More than 60 Chinese companies sent in initial expressions of interest during
the 2006 tender round and half were invited to submit proposals. Of these,
12 were invited to present their proposals before a special tender board in Beijing
made up of professors and experts from China’s own SEZs. Eight proposals
were selected for funding and went into development; one was not completed.
The second round in 2007 found more than 50 companies applying, 20 were
invited to submit formal proposals and 11 proposals were selected. The panels
scored the proposals on the basis of their feasibility studies, the developers’
ability to finance the projects and their proven capacity to implement a major
construction engineering project (Brautigam and Xiaoyang 2011; Brautigam
and Tang 2014). The winning proposals received government financial support
including grants of up to $44 million and long-term loans up to $294 million.
Companies moving into zones from China could apply for reimbursements
of up to half of their moving expenses and reduced cost credit on Chinese
bank loans. Some provinces and municipalities added additional incentives.
Any locally registered small and medium-sized enterprises SMEs in the
zones were also eligible to apply for loan from a $1 billion Chinese fund for
African SMEs announced in November 2009. The zones were more diverse
than originally intended—some were 100 per cent Chinese-owned, some
joint, some mineral processing and some manufacturing. Developers include
both state and private companies. No conditionalities were imposed on the
host governments, who were free to negotiate directly with the implementing
companies regarding incentives and policies for the zones. The incentives
ended up being typical of such zones and included tax holidays, waivers on
import tariffs for raw materials and inputs, and restrictions on strike activities.
Some countries have developed a legal and regulatory framework for the SEZs.
Zambia’s Multi-Facility Economic Zone (MFEZ) policy was developed
133
with assistance from the Japan International Cooperation Agency ( JICA).

The policy allowed for VAT rebates and a graduated income tax schedule for
companies investing in the zone but also required that firms invest a minimum
of $500,000 before qualifying for incentives. Official data from China showed
that Chinese investment in African manufacturing from 2009 to 2012 totalled
$1.33 billion and accounted for 15.3 per cent of investment stocks by the end
of 2011 (Brautigam and Xiaoyang 2011; Brautigam and Tang 2014). It is easy
to exaggerate the impact of Chinese financial flows on economic outcomes
in Africa. Panel data for 1991 to 2010 showed that Chinese aid and FDI
had no significant impact on economic growth in Africa (Busse, Erdogan and
Muhlen 2016). Not surprising in many cases. China only accounted for
2.4 per cent of total FDI entering Ghana during those years (Tang and
Gyasi 2012).
The Chambishi MFEZ (Zambia–China Economic and Trade Cooperation
Zone) is located in a mining area 420 kilometres north of landlocked Lusaka and
is home to a major copper mining complex operated by the China Nonferrous
Mining Co. (CNMC). Firms in the zone have a 25 per cent customs duty
on imported equipment waived and will not have to pay Zambia’s 16.5 per
cent VAT, among other incentives. Firms are not exempt from the Zambian
labour or environmental code and must invest over $500,000 to be able to take
advantage of these incentives. The zone developers declared that they wanted
to build a whole industrial chain from exploration, concentrating and smelting
to the manufacture of copper end products. CNMC experienced problems,
including a series of serious safety violations and strikes. A 2005 explosion in
a dynamite factory on the grounds of the Chambishi mine killed more than
50 Zambian workers (Brautigam and Tang 2014).
The Ethiopian Eastern Industrial Zone (EIZ) is located 32 kilometres
from Addis Ababa on the main road linking Addis to Djibouti, which is about
550 kilometres away. The developer visited Ethiopia in 2006 and saw potential
for manufacturing against the security, political stability, comfortable climate
and developmental goals of the Ethiopian government. The EIZ is 100 per
cent Chinese-owned and was developed and is operated by the Qiyuan Group,
a private Chinese company. The Ethiopian government provided land at a
very favourable rate (around $0.05 per square metre per year for 99 years),
provided all the infrastructure outside the zone and covered the cost of
30 per cent of the internal infrastructure (Brautigam and Tang 2014). Between
early 2010 and July 2012, the start-up area of 100 hectares was levelled and
134
THE DRAGON UNCOILS
roads built; a cement company and a gypsum board manufacturer had built
their own factories and 11 new factory shells had been constructed for leasing.
The EIZ started out focusing on the production of construction materials and
light industries, including pharmaceuticals, electronics, chemicals and leather.
This was later diversified to include cement, packaging, steel pipes, agricultural
machinery, shoe manufacturing and motor vehicles. The zone was open to
investors from Ethiopia and the rest of the world. As of late 2012, only Chinese
enterprises had invested in the zone. One of those enterprises was a major shoe
manufacturer, Huajian, who brought 100 Ethiopians to China for training,
set up an export-oriented factory and eventually employed 1,750 mainly
Ethiopian workers in the zone (Brautigam and Tang 2014). Researchers have
found a voluminous list of factors constraining the growth of the EIZ. These
have included shortages of foreign exchange, inadequate electrical capacity
(low voltage), the slow pace of bureaucratic approvals, the weak regulatory
framework and inefficient customs administration (Giannecchi and Taylor
2018). The development of the EIZ seemed to pick up over time. Prior to
the creation of the EIZ, it could take developers up to eight years to obtain
a land lease certificate, with a great deal of corruption during the process.
Reforms associated with the establishment of the EIZ meant that developers
could get land through a transparent negotiation process and the land could
then be easily sub-leased to the investing companies at a fixed price (Tang
2019: 12). By 2018 the EIZ was employing an estimated 10,000–15,000 local
people in more than 30 companies. The Ethiopian government drew from
this growing success story ambitions to target the construction of 15 industrial
parks around the country and to target employment of 150,000 (Nicolas 2017;
Global Times 2018b).
The Nigeria Ogun-Guangdong Free Trade Zone is located in the Igbessa
Region of Ogun State, 30 kilometres from the international airport serving
Lagos. The project originated from a 2004 study of South China University
of Technology on the feasibility of setting up a Guangdong economic
trade cooperation zone in Nigeria. This report was used for the successful
bid by Xinguang International Group and a consortium in the 2006 tender.
The project was originally sited in Imo State but after several Chinese were
kidnapped from an unrelated project in the area they decided to relocate to
Ogun State as the government appeared better able to guarantee security
there. This delayed the project but developers were able to recruit the Ogun
State government as a minority partner. Construction began only in the first
135
half of 2009 but budget problems in Ogun State delayed some of the work
promised by the Ogun State government, including the paving of the road
leading to the zone. By June 2013 the main roads within the start-up had
been paved and a natural gas power plant was under construction. Thirty-four
enterprises were registered in the zone, coming from Nigeria, China, Lebanon
and India (Brautigam and Tang 2014).
The Egypt Suez Economic and Trade Cooperation Zone is located inside
the 21.85 square kilometre area of the Suez Canal Economic Area outside
Egypt’s new deep water Sokhana Port, just below the southern entrance of the
Suez Canal, and 120 kilometres from Cairo. In 2000 the Chinese launched
their own small zone of 1 square kilometre in a nearby area and in 2007 the
local partners participated in a second Ministry of Commerce tender. The
zone proposal built on the earlier investment and was planned around four
clusters—textile and garments, petroleum equipment, automobile assembly
and electrical equipment—with the possible addition of electronics and heavy
industries in a second phase. Despite early problems with the joint venture,
Egypt’s political transition in 2011 did not appear to have much effect
and by early 2013 38 companies had invested in the zone (Brautigam and
Tang 2014).
There are some general lessons we can draw from the China-led effort to
establish SEZs in Africa. All of the zones aspired to attract large numbers of
firms to invest and all of them had great difficulty in doing so. The Ogun Zone
developers stated a plan to attract over 100 enterprises to the zone within
5 years and 700–800 companies within 10 years. The Egyptian zone aimed
to host 50 small- and medium-sized companies by 2018. All the zones fell
well behind their initial targets. A flurry of media-reported interest by firms
was typically based on little more than signing non-binding memoranda of
understanding (MoUs) rather than actual investment. The most advanced was
the Egyptian zone, where, as of March 2012, about 23 of the firms that had
invested in the zone were reported to be productive enterprises. At Chambishi
in Zambia, 5 initial companies had grown to 26 by 2012 with some
$322 million worth of equipment and plans to invest over $1 billion. Three
of these were Zambian and the rest Chinese. In Ethiopia, by July 2011, the
headline investor was the leather shoe company Huajian Group, which had
opened an assembly line with more than 1,000 workers. Otherwise, the only
factories operating in the EIZ were a cement factory, a brick factory, a gypsum
board maker and a producer of plastic bags. In Nigeria, the Ogun Zone had
136
THE DRAGON UNCOILS
attracted the most interest from private Chinese firms. As of June 2013, all
of the 34 enterprises with commitments to invest in the zone were private
companies but most were fairly small (many only planned to invest around
$1 million). Seven factories were already in operation, including a ceramic
factory, a packaging material producer, a steel manufacturer and a plastic
product maker (Brautigam and Xiaoyang 2011; Brautigam and Tang 2014).
There is not much sign of mass employment being created by Chinese FDI,
but neither is there much evidence for the negative outcomes of Chinese-only
employment being created in Africa. Between 2006 and 2010 for example, in
Ghana Chinese FDI employed 10,048 Ghanaians and 1,828 Chinese workers
(Tang and Gyasi 2010). The ratio of Chinese to African employment has also
tended to decline once initial construction work was completed and factories
then hired more local production workers. At Chambishi in Zambia, by 2009,
3,300 Zambians and 900 Chinese were employed and, by September 2011,
7,973 Zambian workers and 1,372 Chinese workers, including the mining
workforce (Brautigam and Tang 2014). The comparison shows that local
policy can be important in influencing the share of local employment. Egypt
offers one foreign work permit for every nine Egyptians employed. In the
first stage of construction in the Suez zone by Egyptian companies, 1,800
local workers and only about 80 Chinese staff were employed (Brautigam
and Xiaoyang 2011; Brautigam and Tang 2014). The evidence shows that
SEZs do create jobs but survey evidence shows that the working conditions
are often poor. Capacity can be developed through training, yet there is little
evidence of this taking place at the firm level, where Chinese employers
generally hire unskilled African labour. Huajian in Ethiopia is unusual
in providing vocational training to its employees, including training local
technicians in China for 3–12 months (F. Nicolas 2017: 31). More generally
in the EIZ, while export-oriented firms from China do tend to invest more
time in training than firms catering to the local market do, the Ethiopian
government lacks any proactive SEZ training policy, and skill transfer from
Chinese to local employees continues to remain minimal (Fei 2018: 22).
Chinese attitudes brought to Africa suggest a high degree of paternalism,
high regard for hierarchy, low regard for trade unions and a relationship that
may disadvantage non-Chinese employees. Chinese MNCs have created
employment but they could do more in terms of contributing directly
to skills development, engaging with communities, mutual learning and
looking at how employment conditions might be improved ( Jackson 2014).
137
SEZs, argues Kim (2013), have been more successful in strengthening Chinese
soft power among African elites than among the masses, who remain sceptical
about issues regarding wages, working conditions and labour rights. It is easy
to exaggerate the importance of a specific ‘China Factor’. African views (across
20 countries using the Afrobarometer data—an all-Africa research network
that provides survey evidence related to economic and social matters) of China
are not that different from African views about western countries. China is
seen as a relative threat to employment from imports but a benefit in terms of
FDI reducing poverty. China scores less than western countries on civic and
political human rights (Hanusch 2012).
So far none of the zones shows any sign of clustering. The zones in Egypt
and Ethiopia included a plan for clustering in their original SEZ design but
in neither case did this materialise in practice. In none of the SEZs were there
a sufficiently long queue of firms applying to join that developers could pick
only those who fitted the cluster model. In the Ogun zone in Nigeria, for
example, developers had intended to focus on light manufacturing. In practice,
firms that located in the SEZ came from an eclectic mix of sectors, including
packing materials, ceramics, plastic, steel construction materials and detergent
powder (Brautigam and Tang 2014). Recently, five new textile and garment
manufactures have opened up in the EIZ, doubling the number of such firms,
and raising new hopes of clustering effects; the firms are only small producers
though and remain small relative to the overall size of the EIZ (F. Nicolas
2017: 30).
The creation of backward spillovers from firms located in the SEZs
to the domestic economy has in general proved disappointing. As of late
2012, few local enterprises had moved into the SEZs; there were no local
manufacturers or suppliers in the Ethiopian, Mauritian and Nigeria-Ogun
zones. The copper-focused SEZ in Zambia had to wait until 2012 to receive
its first manufacturer, who turned out to be based in plastic products. Local
enterprises in the Egyptian SEZ included a bank and a customs clearance
company (Brautigam and Tang 2014). Mauritius even prevented local firms
from joining the SEZ to prevent local companies simply relocating production
to take advantage of the SEZ incentives (Brautigam and Xiaoyang 2011;
Brautigam and Tang 2014). The failure of spillover creation can also be seen in
the activities of Chinese firms. Hazan is a major Chinese shoe manufacturer
from Wenzhou in Zhejiang province. By 2009 Hazan and at least four other
Chinese shoe manufacturers had moved their shoe-making assembly lines to
138
THE DRAGON UNCOILS
Nigeria. All the firms continued to import crucial inputs, including uppers
and soles, from China. Another example was the large Chinese shoe company
Huajian Group, which invested in the EIZ.in Ethiopia. The potential to
create backward spillovers in leather was obvious. Animal husbandry accounts
for 20 per cent of Ethiopia’s GDP. Huajian tried to source leather from a
new Chinese leather processing factory, China–Africa Overseas Leather Co.,
which had just started operation near Addis Ababa, outside the EIZ. The
link failed because the leather company would face the prospect of losing its
tax holiday if it sold to a local firm rather than export its production. As of
July 2012, Huajian was unable to source more than 30 per cent of its leather
inputs locally. Overall, very few of the enterprises in the EIZ have developed
durable backward spillovers. By 2011 an estimated 61 per cent of the total
material inputs and supplies used by Chinese firms in the Ethiopian EIZ were
sourced abroad (F. Nicolas 2017: 30). The EIZ has replicated the weaknesses
that characterised previous attempts in Africa to set up industry clusters—
poor infrastructure, fragile ties to sources of innovation and technology, and
a broad absence of state support (Giannecchi and Taylor 2018). The EIZ
remains keen to work with local companies in Ethiopia and continues to
actively encourage Ethiopians to invest in the zone. The recent establishment
of the Huaijian International Light Industry City is intended to encourage
domestic and foreign firms to establish capacity in locally produced inputs
(F. Nicolas 2017: 30).
China was drawn into copper mining in Zambia lured by both booming
demand from China and rising world prices. In 1998 the state-owned CNMC
acquired a small mine with support from the China Exim Bank. The presence of
the CNMC increased over the following decade as they purchased more mines
and invested $800 million into the Chambishi Zambia–China Economic and
Trade Cooperation Zone. This gave China a significant presence in Zambian
copper across mining, smelting and processing industries. The copper sector
in Zambia had been locked into a spiral of neglect and decline since it was
nationalised in the 1970s. Indicators started to look up. The export of copper
increased from $474 million in 2000 to $4 billion in 2008 and, with only a brief
interruption during the Global Financial Crisis, further to $5.65 billion in
2010. The growth of mining and some related production created a substantial
demand for inputs, including steel plates, rubber products, lubricants,
explosives, tyres, cement and personal protective equipment. Between 1998
and 2009, these backward spillovers leaked out of Zambia into imports from
139
South Africa. There was no realistic prospect of backward spillovers being

retained within the Zambian economy. The Zambian economy lacked the
capacity to meet these demands and there was little indication of sustained
Chinese efforts to change this. Compared to South African and European
companies, Chinese producers were less embedded in the local community;
they were not, for example, members of the local Kitwe Chamber of Commerce
and Industry. Traditionally, South African mining firms had made significant
efforts to construct local networks to source their inputs and other supplies.
Such efforts included advance payments, assistance with transport to help
delivery, technical help to suppliers to improve product quality, joint running
of quality tests and detailed quality feedback. This assistance was structured
around long-term contractual relationships between mining companies and
local suppliers. The relationship between Chinese mining firms and local
suppliers tended to be more market-based and revolved around short-term
considerations of cost. Chinese firms made little effort to help suppliers in
upgrading skills, technology or management procedures and the relationship
was limited to negotiating delivery contracts (Fessehaie 2012; Fessehaie and
Morris 2013).
Chinese firms have tended to rely on initiatives by the Chinese government
to promote backward spillovers in the local economy. One example of this is
the Chinese zone programme run by China’s Ministry of Commerce. This
programme runs hundreds of training programmes for government, private
and NGO officials from developing countries. These training programmes
have often been structured around learning about the Chinese domestic
experience of promoting investment through the creation and management
of SEZs. The EIZ in Zambia is linked to the Chinese-built Technical
and Vocational Education and Training (TVET) Centre in Addis Ababa
(Brautigam and Tang 2014). Efforts of host governments other than Ethiopia
to actively promote links, beyond establishing the SEZs, appear minimal.
In Zambia, for example, there has been no effort to utilise any form of industrial
policy to promote backward spillovers by tackling skill shortages, capital
market imperfections or high transport costs (Fessehaie 2012; Fessehaie and
Morris 2013). Sometimes the rules by which the SEZs operate hinder the
creation of backward spillovers. The Zambian government had mandated a
$500,000 minimum investment for firms to enter and enjoy the incentives
offered by the SEZ which proved to be a substantial hurdle for local investors
(Brautigam and Tang 2014).
140
THE DRAGON UNCOILS
Amendolagine et al. (2013) examine data from 1,400 companies across

19 SSA countries from 2010 to examine spillovers, the demand for local
intermediates and the cost share of local intermediates. They find that firms
with a foreign partner and those selling to the local market rather than
exporting have a higher degree of local interactions—an obvious problem as
most of these SEZs were established with the aim of boosting exports. Local
spillovers are higher when local management has more autonomy from the
company headquarters and especially when firms are standalone rather than
being part of a group—again a problem, as Chinese investment tends to be
from large MNCs or state-owned enterprises. After controlling for a set of
firm characteristics, not surprisingly they find that Chinese firms create fewer
backward spillovers than other firms. As well as the character of the investor,
host country characteristics are important. Domestic policy is important in
boosting the domestic spillovers generated by foreign firms. Auffray and
Fu (2015) examine spillovers from foreign construction firms in Ghana. In the
early 2010s, Chinese firms began to challenge established European firms in
bidding for and winning large construction contracts. Chinese construction
companies do have a reputation of using fewer local suppliers. The study
found this view to be exaggerated and that Chinese firms were sourcing
from domestic producers. One mechanism to promote local spillovers
worked through Ghanaian government regulations regarding employment.
Employment of foreign nationals is restricted in Ghana. Foreign companies
face a quota on visas and are compelled to employ Ghanaian nationals in any
business. Chinese contractors employ both more expatriate managers than
other firms and also more local, unskilled labour. There remains a widespread
perception that the Chinese language has created a communications barrier
between Chinese and locally hired managers. Chinese firms have initiated
a managerial localisation programme to boost the number of Ghanaians
in management positions whereby a local is paired with a Chinese expert
with training provided locally and in China. This has helped improve
communication inside and beyond the firm and reduced costs. A much more
general study of spillovers comes from Seyoum, Wu and Yang (2015) who
examine the impact of Chinese manufacturing FDI on the productivity of
Ethiopian manufacturing. The study is based on 1,033 large- and medium-
scale manufacturing firms operating in Ethiopia in 2011. The study finds
that foreign-owned firms are significantly more productive than their
local counterparts, indicating the potential for Ethiopian firms to learn.
141
The empirical results show that domestic firms with a smaller technology
gap with foreign firms experience positive spillovers, while for more backward
firms there are negative spillovers.
In part the absence of backward spillovers reflects the nature of the local
economy where the Chinese have invested. Chinese firms have proved willing
to enter where local manufacturing capacity is low, the existing supply chain
is weak and profit margins are slim (Dannenberg, Kimii and Schiller 2013).
Where western firms see risk, Chinese firms see opportunity, are thinking
longer term and accept lower profit margins in the short term. Using data for
44 African countries for 2003 to 2017, Miao et al. (2020) find that Chinese
FDI has no significant positive effect on growth except in those countries with
a better institutional environment. However, using data on registered Chinese
firms investing in Africa between 1998 and 2012, Chen, Dollar and Tang
(2018) find that Chinese FDI, unlike western FDI, does not favour countries
with good governance. China is more inclined to invest in politically unstable
countries. There is some evidence that backward spillovers were influenced
by a maturity effect and have increased over time as Chinese firms learned
more about the socioeconomic environment in which they were investing.
In a case study of Chinese FDI in the DRC, Parente et al. (2018) find that
the Chinese firm Weihei entered through their first construction project to
build a school. This was an aid project in which Weihei took the lead and
thereby built local legitimacy and relationships with officials. This was in 2008
and they imported most of their equipment and inputs from China. The long
delivery time for these imports gave Weihei an incentive to purchase local
supplies. Weihei took an active role in building such a supply network for
raw materials and inputs together with local and Chinese suppliers. Some
of those Chinese suppliers Weihei had itself introduced to Africa. There
has been a more widely observed tendency that once a pioneer Chinese firm
establishes production overseas, there is then a tendency for Chinese firms to
cluster together for mutual support and coordinated production. The Chinese
government and Chinese state-owned banks helped build Wehei’s reputation
with potential clients. This, together with a gradually strengthening and more
reliable domestic supply chain, allowed Wehei to subsequently start bidding
on larger, World Bank–funded projects.
Evidence from China’s own history shows that one way of encouraging
spillovers is to organise FDI policy through joint ventures. A joint venture
can include the signing of a technology transfer agreement at the outset of
142
THE DRAGON UNCOILS
an FDI project. In China, prior to joining the World Trade Organization

(WTO) in 2001, joint ventres accounted for around two-thirds of the FDI
from the early 1990s to 2000, but declined to 20 per cent when China joined
the WTO. WTO rules made it much harder to impose such restrictions on
the free flow of FDI. In absolute terms though, there was less change. FDI
governed by some form of joint venture obligation declined from $20–30
billion per annum in the 1990s to $20 billion per annum in the 2000s. Data
from a firm-level survey between 2000 and 2005 shows that joint ventures
are 73 per cent more productive than other firms, while technology transfer
agreements are more strongly associated with productivity than other forms
of joint ventures (Yueh 2013: 165–8). There is no indication that governments
in African SEZs are learning from the Chinese experience of joint ventures
beyond some tentative efforts by Ghana to push for locals to be hired as
managers.
The SEZs in Pakistan represent a delayed reaction to the new round of
industrial restructuring that happened in China after 2010. This was linked
to the Going Out strategy launched by the Chinese government which saw
some low-end export-oriented manufacturing activities shifting from coastal
China to low-income countries in Southeast Asia and Africa. The SEZs
represent little more than another attempt at a very traditional approach.
Pakistan is again trying to attract investment from China in the import-
substituting industry by providing an array of tax and other incentives. This
strategy may succeed in attracting production in consumer durables such
as televisions, refrigerators, air conditioners and washing machines. There
is a risk that the SEZ will repeat past failures, that the SEZ-style ‘strategy
has failed in the past and it is unlikely to do much better this time since it
will only attract investment for assembly plants producing for the domestic
market’ (Hamid and Hayat 2012: 278). Pakistan’s strategy could be to attract
Chinese investment into export industries, particularly those labour-intensive
industries that are likely to be relocating out of China in the next 10 years and
which are also Pakistan’s strengths. These include garments, textiles, leather
and footwear, surgical goods, cutlery and sports goods (Hamid and Hayat
2012). Chapter 7 discusses the Pakistan–China trade policy and Chapter 8
the importance of an industrial policy, both of which need to be thought about
carefully by Pakistani policymakers in order to maximise the potential benefits
of the CPEC.
143
IN SUMMARY
SEZs have been widely used as a means to promote industrialisation,
technology transfer, employment growth and exporting over the last 70 years.
Theory and empirical evidence show that they have a mixed record with both
positive and negative consequences. The Government of Pakistan has absorbed
the positive side of this story without much critical reflection. Nine SEZs are
planned as part of the CPEC investment. Why will FDI flow to these SEZs
rather than the thousands that exist in countries with better governance? Why
will the CPEC SEZs be a success when so many similar efforts have failed
in Pakistan’s own past? China’s own history with SEZs since they were first
established in 1979 does serve as an important marker but it is far-fetched
to think that Gwadar or Pakistan can become a ‘new Shenzhen’. Chinese
constructed and invested SEZs in Africa have been a mixed success. The most
successful examples have been in Ethiopia where they had been driven by the
proactive interventions of a developmental state. Recall as well from Chapter
4 the discussion of how a successful industrial policy in nineteenth-century
Germany helped ensure that railway construction generated links to domestic
German industry. We return to this debate in Chapter 8 when we ask whether
Pakistan can successfully use an industrial policy to maximise the benefits of
the CPEC.
144
6
The Dragon’s Embrace

Pakistan–China Trade Policy
The section ‘The Benefits and Costs of EPZs and SEZs’ in Chapter 5 made a
case that what is happening in China, particularly Western China, is of crucial
importance to the likely impact of the CPEC on Pakistan. While the CPEC
is about infrastructure and FDI, these processes are influenced by overall trade
relations between China and Pakistan. Infrastructure links that open up trade
between China and Pakistan will make little difference if the two countries
charge high tariffs on trade. This chapter outlines the emerging patterns of
trade between Pakistan and China, particularly in relation to the signing of
the 2006 FTA.
THE CHINA IMPACT ON WORLD TRADE:

THE IMPLICATIONS FOR PAKISTAN
In 1979 China had a mix of land shortages and an abundant and reasonably
well-educated labour force. The entry of China onto world markets after
1979 from a position of near autarchy had a global impact on relative factor
endowments and thereby the comparative advantage of other countries (Wood
and Mayer 2011). The value of having a great deal of available land to produce
agricultural products, natural endowments of oil, metals and industrial inputs
such as cotton, or the ability to produce skill-intensive manufacturing and
services in other countries increased as China sought more imports (Renard
2011). Those countries exporting low-skill labour-intensive manufactured
goods were set to face the brunt of Chinese competition in export markets.
Chinese potential was realised in the massive surge of Chinese exports.
China’s exports of manufactures increased from $48 billion in 1990 to
$303.5 billion in 2002, from a 3 per cent share of world manufactures in
1995 to 11 per cent in 2006. These exports were initially concentrated in
145
labour-intensive sectors such as textiles, clothing and wood products. Between

1990 and 2006, China’s exports of clothing increased from $9.7 billion to
$95.4 billion and its world market share from 9 to 31 per cent, with an extra
9 per cent when including Hong Kong (Wood and Mayer 2011). In many
cases, this had a devastating impact on other exporters of textile products.
The value of clothing exports from sub-Saharan Africa (SSA) to the
US dropped by 25 per cent between 2004 and 2006. For Lesotho, the fall
in export value was 15 per cent, mostly in 2005, but exports did stabilise in
2006. Madagascar saw a 24 per cent decline, Swaziland 22 per cent, Kenya
3 per cent, the Republic of South Africa (RSA) 54 per cent and Mauritius
62 per cent. In the same period, the value of China’s clothing and textile exports
to the US increased by 82 per cent (Kaplinsky, McCormick and Morris 2007:
23). At their peak in 2002, Lesotho’s clothing exports to the US accounted for
virtually all the country’s manufactured exports and were equivalent to 50 per
cent of the GDP. In Swaziland, the most severely affected, overall employment
almost halved. In Lesotho in the first half of 2005, 8 of the 47 garment-
exporting factories closed and employment fell by 26 per cent. In Kenya,
clothing enterprises accounted for the equivalent of nearly 20 per cent of all
formal sector manufacturing employment and clothing exports fell
2.5 per cent in 2005 and employment declined by nearly 10 per cent (Lall,
Weiss and Oikawa 2005; Kaplinsky, McCormick and Morris 2007; Morris
and Einhorn 2008; Giovannetti and Sanfilippo 2009; Sandrey and Edinger
2011; Jenkins and Edwards 2015). Using more rigorous statistical evidence,
Giovannetti and Sanfilippo (2009) ask whether Chinese exports crowded
out African exports from markets in other countries. They find that the rise
of Chinese exports displaced African exports, in machinery and equipment,
textiles, especially footwear, and in chemical products such as petroleum
refineries, coal, rubber and plastic products and also in non-metallic mineral
products. One specific effort to quantify the impact found that RSA exports
to SSA were 10 per cent lower, to the EU 8 per cent lower and to the US
6 per cent lower than they would have been without Chinese competition.
These lost exports included knitted apparel, prepared fruit and vegetables,
tinned peaches, pears and apple juice, iron and steel, non-electrical machinery
and vehicles and parts ( Jenkins and Edwards 2015: 917). More generally,
this competitive effect was found to have led to a reduction in exports of
labour-intensive manufactured goods of between 1.5 and 5 percentage points
for the average developing country (Wood and Mayer 2011).
146
THE DRAGON’S EMBRACE
The rapid growth of Chinese trade with Latin America dates from the late
1990s. At that time, China was facing a shortage of natural resources, which
were increasingly being met through imports, and China was poised to join
the WTO (in 2001). Latin America had undertaken extensive liberalisation
of trade, FDI and the private sector in the 1980s and 1990s, which meant it
was well positioned to respond to these changing global market conditions
induced by China. At first glance, it appeared that Latin America would face
less damaging competition than Africa from China. Using a measure of the
similarity of exports across 181 manufactured products, there appeared to be
no significant overlap between Chinese exports and those from Latin America
in the early 2000s. There were exceptions in the case of Costa Rica and
Mexico. For Costa Rica, this threat was in electronics, apparel, processed foods
and instruments. Mexico had a very similar production and export structure
overall to China. By the early 2000s, Mexico was suffering from the effects
of Chinese competition, was running a $5.7 billion trade deficit with China
and also losing third markets (especially the US) for its exports. Argentina
and Brazil by contrast experienced trade surpluses with China as they had
large and competitive primary and resource-based export sectors. There are
problems in using this evidence to jump to a generally optimistic conclusion
regarding China–Latin American interactions. This data was based on a
snapshot and says nothing about how the potential competitive threat from
China evolved over time. The data was showing that the intensity of the
Chinese threat to Latin American exports was increasing over time. This was
not good news for Latin America. Countries such as Argentina, Paraguay and
Peru were being pushed back into primary or resource-based production. For
Mexico, the decline in export overlap was due to the squeezing out of Mexican
exports to third markets (Lall, Weiss and Oikawa 2005). Various countries,
despite the protection of geographical proximity, were even squeezed out of
exporting to the US. One estimate is that for the 18 countries of Latin America,
exports to the US were 9 per cent lower than without China ( Jenkins, 2010,
2012; Jenkins, Peters and Moreira 2008). After the initial surge, there was
some hope that the competitive threat from Chinese exports was diminishing.
In the early 2000s, 90 per cent of imports from China were low technology
and labour-intensive. The technology level of Chinese imports increased
over time. By 2008 low-technology manufactured imports comprised only
20 per cent of the total, medium-technology 25 per cent and high-technology
40 per cent ( Jenkins 2012). Perhaps Latin America could rediscover a niche in
low-tech, labour-intensive manufacturing?
147
The flipside has been Chinese demand for imports.

Demand for oil and other minerals from China made all the difference for
SSA. If oil is excluded in 2005, SSA had a trade deficit of $7.3 billion with
China; if oil is included, SSA ran a surplus of $5.9 billion in the same year.
In 2004 China sourced 30 per cent of its oil from Africa. Chinese demand was
not limited to oil but permeated across global trade in minerals. Between 1998
and 2003, China’s share of increased global demand was 96 per cent for steel,
99 per cent for nickel, 100 per cent for copper and 76 per cent aluminium.
Between 90 and 100 per cent of oil exports from Angola, Nigeria, Sudan and
Congo went to China, 60 per cent of metal exports from Ghana and almost
100 per cent from the DRC, 42 per cent of wood exports from Gabon and
40 per cent from Cameroon, and 54 per cent of cotton exports from Tanzania
(Kaplinsky, McCormick and Morris 2007: 16).
For some countries in Latin America, China was emerging as a significant
source of export demand. By 2005, over 10 per cent of the total exports
from Chile, Peru and Cuba went to China. Between 2000 and 2005, up to
20 per cent of export growth from Brazil and Argentina went to China. As a
whole though, Latin America remained more marginal for China as a source
of imports. By 2010, for the region as a whole, China only accounted for
9 per cent of exports, but less than 2 per cent of exports from Mexico,
Dominican Republic, Ecuador, Paraguay and various other Central American
countries. For China, the region was negligible. Even the large economies of
Mexico and Brazil only accounted for 1–1.5 per cent of total Chinese imports
( Jenkins 2012).
Latin American exports to China were not very diversified and were more
heavily concentrated in primary and resource-based manufactures than for
the rest of the world. In the early 2000s, the main exports consisted of soya,
iron ore, pulp, fish meal, leather and copper. Seventy-five per cent of China’s
imports from Argentina and Chile were soya and copper respectively. This
trend was increasing over time as Chinese demand affected the international
terms of trade. Prices of agricultural goods increased relative to manufactured
goods, which changed global incentives to produce and export. Venezuela
benefited from higher oil prices, Argentina from increased agricultural prices
and Chile from higher copper prices. The share of total exports to China from
Latin America accounted for by primary products increased from 40 per cent
in 1990 to 72 per cent in 2008, while to the rest of the world it decreased from
49 per cent to 39 per cent. The share of manufactured goods in exports to
148
China from Latin America declined from 60 per cent to 28 per cent and to the
rest of the world increased from 50 per cent to 59 per cent. Some of this decline
was driven by price incentives and some by Chinese state intervention.
In soya, steel and iron, the Chinese government was promoting the expansion of
processing industries within China and switching to imports of raw materials
( Jenkins 2012: 1344). Chinese FDI in mining and oil comprised more than
90 per cent of Chinese investments in Latin America and reinforced this
restructuring ( Jenkins, Peters and Moreira 2008: 240).
China was clearly driving a distinct pattern of change in Latin America
but was a long way from becoming a new hegemonic power in the region.
In 2010 only 8 per cent of the region’s exports went to China, compared to
41 per cent to the US and 13 per cent to the EU; the pattern of imports was
similar. Despite rapid growth, by 2010 China accounted for only 0.3 per cent
of the stock of FDI in the region ( Jenkins 2010, 2012; Jenkins Peters and
Moreira 2008).
The level and pattern of international (Chinese) demand is also important.
Chinese consumers seem to be less concerned with the value of ‘variety’ than
EU buyers, and have a lower preference for environmental standards and so
make few related demands on international suppliers. Chinese buyers place
a greater premium on low price and large volumes. For Gabon, there was a
marked shift in logging demand from furniture makers in the EU to those in
China after the mid-1990s. The Chinese preference for importing unprocessed
raw materials led to a collapse in producer capabilities in sawing, cutting and
finishing, and a worsening of environmental standards and those governing
the working conditions of forest workers. There was a loss of employment
in domestic processing and value added. A similar outcome occurred in
cassava exports from Thailand, which experienced less domestic processing
as higher value added and technologically demanding production niches were
increasingly taken over by local producers in China (Kaplinsky, Terheggen and
Tijaja 2010).
There are some examples of a successful and dynamic response to Chinese
competition. In Ethiopia, domestic footwear firms, particularly the leather
shoe industry near Addis Ababa, responded by upgrading their production
capabilities (Morris and Einhorn 2008). In the early 2000s, Ethiopia was
flooded by Chinese shoe imports. The local industry first slumped, then
responded, with output growing by 25 per cent per annum between 2002
and 2004, and even started exporting. By 2006 Ethiopia had emerged as a
149
net exporter of shoes. A sector emerged containing around 1,000 factories

producing leather shoes with most employing less than 10 workers but some
employing hundreds of workers. New enterprises were established by former
employees and saw improvements in the quality of products, marketing
and management. The speed of upgrading in production management was
striking. In the two years after 2004, the rate of customer returns was reduced
by between 30 and 40 per cent. Firms also managed to reduce stocks of
inventories and reduce the lead time from order to delivery. Large enterprises
were able to purchase leather and soles in bulk from tanneries and established
connections with European enterprises to access information on advanced
technology and latest fashions (Sonobe, Akoten and Otsuka 2009; Gebre-
Egziabher 2009).
TRADE POLICY IN PAKISTAN SINCE THE 1980s

Beginning in the late 1980s, Pakistan substantially liberalised its economy.
These changes sought to increase competitive pressures on incumbents by
easing the entry of new producers and encouraging more imports into the
country. It was anticipated that this would compel producers to upgrade and
become more efficient and so enable them to expand and to export. To this
end, trade liberalisation, which began in 1987, continued deepening into the
late 1990s. The number of tariff slabs fell from 14 to 4, and the maximum
tariff fell from 225 per cent in 1986–7 to 70 per cent in 1994–5 and to
25 per cent in 2001–2 (S. A. Zaidi 2005). Liberalisation also encompassed the
complementary areas of FDI (Kemal 1999) and finance (Husain 2003;
B. A. Khan 1999; S. A. Zaidi 2005). The macroeconomic results were
disappointing. GDP growth, which had averaged 6.0 per cent between
1961–2 and 1991–2, fell to an average of 4.3 per cent between 1992–3 and
2010–11, and this was around a declining trend (Kite and McCartney 2017).
Trade liberalisation did not energise Pakistan’s trade performance. The
growth of exports matched only this slowing economic growth. Exports as a
percentage of GDP declined from 16 per cent in 1992 to 10 per cent in 2013
and then even fell in absolute terms from $25 billion in 2014 to $22 billion
in 2017. Manufacturing employment, which peaked at 16.3 per cent of total
employment in 1966–7, has remained between 10 and 15 per cent ever since
(Hamid and Khan 2015; Pakistan Business Council 2019: 6).
150
By 2017, textiles contributed 8.5 per cent of Pakistan’s GDP, 40 per cent
of industrial employment and 60 per cent of national exports (Pakistan
Business Council 2018: 11). Exports reached $8.6 billion in 2004–5, making
Pakistan one of the largest textile exporters in the world (Pakistan Business
Council 2018: 14). The US and EU accounted for 25 and 20 per cent of
Pakistani exports in the mid-2000s respectively (Shabir and Kazmi 2007).
World trade in textiles and clothing boomed in the 2000s, increasing from
$157 billion in 2000 to $250.7 billion in 2010 and to $454.2 billion in 2017.
Textiles remained Pakistan’s leading export sector throughout, but performed
poorly against this favourable backdrop. Exports from Pakistan of all textiles
increased from $11 billion in 2006–7 to only $12.5 billion in 2011–12, and
this around a fluctuating rather than rising trend. Domestically, the textiles
sector experienced growth of less than 1 per cent per annum in 2010–11,
2011–12 and 2012–13 (McCartney 2015a).
Pakistan has a narrow export base: ‘other made up textile articles’, ‘articles
of apparel and clothing accessors’ and ‘cereals’ make up 65 per cent of Pakistani
exports (Pakistan Business Council 2019: 7). There was no change in an index
of export composition between 1986 and 2004 (Hamid and Khan 2015: 120).
Within these broad averages, there was some sign of a welcome shift into
garments. By 2017, Pakistan was exporting $5 billion of garments, which
represented a doubling over the previous 15 years (Pakistan Business Council
2018: 14). As we shall see later, these garments represented the low end of
the market in terms of quality and price. Pakistan’s export basket remains
concentrated in sectors that are exported by even poorer countries which has
undermined prospects for faster export and economic growth.
CHINA AND PAKISTAN TRADE NEGOTIATIONS

There is a clear link between the China story and the Pakistan story.
The rise of exports from China coincided with the slowing growth of similar
exports from Pakistan in the 1990s. Part of this effect can be attributed to
China (Wood and Mayer 2011). Evidence for this can be found in measures
of the competitive threat of China to Pakistan. The competitive threat is
based on the similarity of production and export structures, their composition
and also volume of exports. Pakistan is vulnerable because it specialises in
exports that overlap substantially with those from China. Data for the early
2000s show that 80 per cent of exports from Pakistan are in sectors where
151
China is internationally competitive and 92 per cent of Pakistan’s exports are

in sectors in which China has a clear comparative advantage ( Jenkins 2008:
1360). The threat and opportunity of trade with China has steadily increased.
Until 2000, China’s share in Pakistan’s external trade was less than 6 per cent
and it crossed the 10 per cent mark after 2003. Pakistan’s imports remained
focused on the US, Japan, Saudi Arabia, Germany, UK and Malaysia (Shabir
and Kazmi 2007). Pakistan has steadily localised its trading patterns. By 2010,
about 25 per cent of Pakistan’s exports and 35 per cent of imports were from
neighbouring countries, including the UAE, China, Afghanistan, India and
Iran. As a group, these neighbours are now more important for Pakistan’s trade
than North America or Europe (Hamid and Hayat 2012). China now has the
largest share in Pakistan’s total imports (29 per cent) and the UAE second
(13 per cent) (Kamal and Malik 2017).
This shift in the geography of trade has been influenced by policy
changes. Since May 2001, China and Pakistan have engaged in high-level
political visits resulting in various agreements to boost tourism, technical
cooperation, infrastructure and especially trade. This culminated in the 2006
FTA. Under the agreement, which became operational on 1 January 2006,
China reduced tariffs to zero on 767 items and Pakistan 356 items (Shabir
and Kazmi 2007). There is good reason to suppose the FTA (and CPEC-
related infrastructure improvements) would lead to more trade between
Pakistan and China. Choudhri, Marasxo and Nabi (2017) have constructed
a data set encompassing 183 reporting countries for 2004–13 with more
than half a million observations to estimate a gravity model of international
trade to identify the effects of barriers on Pakistan’s bilateral trade flows with
China. The results show that barriers to overland trade with China have
substantially reduced bilateral trade. Using the estimated model, they estimate
that Pakistan’s exports to China would increase by 8 per cent for a modest 10
per cent reduction of trade costs and by 19 per cent for the larger 25 per cent
trade cost reduction. By other methods, the increase may be up to 50 per cent.
Unfortunately, the increased trade has been entirely due to more exporting
by China. The China–Pakistan trade balance has been consistently in favour
of China, except during the Korean War in 1952 (Shabir and Kazmi 2007).
Figure 6.1 shows that China exported more to Pakistan than it imported in
the years leading up to the FTA in 2006 and CPEC discussions in 2013. After
the CPEC agreement was signed, exports from Pakistan stagnated or even
declined as exports from China surged in the years up to 2017.
152
Figure 6.1 Pakistan: imports from and exports to China, 2003–17

The (official) data used in Figure 6.1 underestimates Pakistan’s trade deficit
with China. It has been relatively easy for unscrupulous Pakistani importers
and Chinese exporters to misclassify imports from China and understate
their value to evade import duties and taxes. As a result, the actual increase
in imports was likely even greater than that indicated by official figures. One
way to estimate the extent of this tax evasion is to compare the difference
between Pakistan’s reported imports from China and China’s reported exports
to Pakistan. In 2017 this was $2.8 billion, down from $5.42 billion in 2015
(Pakistan Business Council 2019: 4). The underreporting is probably even
greater since exports are reported on a free-on-board (f.o.b) basis and imports
on a cost-insurance-and-freight (c.i.f.) basis and the cost of insurance and
freight is generally between 10 and 20 per cent of the import value. Even
with a conservative 10 per cent adjustment for insurance and freight, the
underreporting is significant. This would mean that in 2010 actual imports
from China amounted to $7 billion as opposed to $5.2 billion in the official
data. This both undermines government tax and tariff revenue and also
provides unfair competition for domestic producers in Pakistan (Hamid and
Hayat 2012: 279).
China’s export boom has encompassed both low- and high-technology
exports. Productivity in China has risen fast enough to offset rising wages
and China remains competitive in labour-intensive production (Lall, Weiss
and Oikawa 2005). This is clearly reflected in the wide range of goods that
153
China exports to Pakistan. The top Chinese exports to Pakistan as of 2007

included machinery and parts, railway vehicles and parts, road vehicles and
parts, iron and steel manufactures, tyres and tubes, chemical materials and
products, fertilisers, yarn and thread of synthetic fibre, and medical and
pharmaceutical products (S. Kumar 2007: 765; Shabir and Kazmi 2007).
In 2017 26 per cent of Chinese exports to Pakistan consisted of electrical
and electronic equipment and machines, around 10 per cent iron and steel or
articles of steel, and 3.6 per cent man-made filaments, 2.2 per cent man-made
staple fibres, 1.1 per cent knitted or crocheted fabric and 1 per cent ceramic
products (Trading Economics 2019).
The rapid growth in imports from China has undermined a number of
industries in Pakistan. Between 2006 and 2017, there was a sharp reduction
in the output of sectors such as heavy machinery and equipment, bicycles, TV
sets, electric transformers, electric meters, sewing machines and deep freezers
(Kamal and Malik 2017: 17). These were industries that were dominated by
small to medium firms producing for the local market. Imported Chinese
products were often of poor quality and even imitations of established
local brands. They prospered because they were very cheap due to Chinese
competitiveness, bolstered by the evasion of taxes and import duties. In many
cases, small local producers were unable to compete with such imports because
the effective tariff (including sales tax) on the final product imported from
China was much lower than the effective tariff on the raw materials used by
small manufacturers in Pakistan. Small producers have to buy raw materials
from commercial importers who have to pay the import duties and a sales tax
as they are not eligible for the concessions that large manufacturers enjoy under
Pakistan’s Statutory Regulatory Order Regime (Hamid and Hayat 2012: 279).
In some cases, Pakistan benefited from the import boom. Chinese
imports have generated a huge gain in industries where Pakistan did not
have any local manufacturing. Pakistan would never have achieved the mobile
phone penetration it has if only full duty-paid and sales-tax-paid phones
were available in the market. The motorcycle industry was previously an
oligopolistic industry shielded by high levels of trade protection. The industry
was opened by removing entry restrictions on the assembly of motorcycles
and allowing the import of parts and components from China in 2006.
This resulted in the transformation of the industry. The domestic market
was boosted by a 40 per cent decline in real prices between 2006 and 2012.
Domestic production of motorcycles rose from less than 600,000 in 2004–5 to
154
over 1.6 million in 2010–11 (Hamid and Hayat 2012). The import of Chinese
machinery has helped local industry in Pakistan upgrade its technology in
some sectors. Pakistan has imported cheap filament yarn for use by the textile
sector, fertiliser and machinery in use by the agricultural sector, and cranes
and other heavy transport equipment for the construction sector. Imports to
Pakistan from China of electrical machinery and equipment increased from
$1.76 billion in 2013 to $3.65 billion in 2017 and of machinery parts from
$837 million to $3.3 billion over the same years (Pakistan Business Council
2019: 16). These benefits remain more marginal than they should be. Pakistan
has not made enough progress on reducing tariffs on imported intermediate
goods. India and Turkey by comparison made significant reductions in related
tariffs and saw an increase in imports of intermediate goods from $2 billion
in 2003 to more than $18 billion in 2011 in India and from $2 billion to
$5 billion in Turkey. Firm-level surveys in Pakistan show that access and cost
of intermediate goods remain a major constraint on firm-level output growth
( Jamil and Arif 2018). The import of low-cost consumer goods could benefit
local retail business. Various markets have opened across Pakistan selling
Chinese imported goods such as electronics, automobiles, toys and electrical
goods and accessories (Kamal and Malik 2017).
Eighty per cent of Pakistani exports to China consist of rice, cotton yarn
and fabric. This mix has been unchanged since 2000, with even signs of
downgrading (recall the discussion about Latin America). The share of cotton
yarn has increased from 41 per cent in 2000 to 51 per cent in 2010, much of this
to supply textile industries in China (and also Bangladesh) (Hamid and Hayat
2012: 276). The share of cotton yarn has continued rising and reached 59 per
cent in 2017. The remaining export items to China consisted of equally low-
tech, low-value items such as ores, slag and ash, cereals, salt, stone, plaster, lime
and cement, and raw hides and skins. Only 4.5 per cent of Pakistani exports to
China consisted of manufactured textiles and apparel of various types (Kamal
and Malik 2017; Trading Economics 2019). The 2006 FTA or the CPEC has
not prompted any diversification of exports. But then, why would we expect
any diversification? It is not surprising that trade liberalisation has stimulated
the expansion of exports of raw materials from Pakistan to China. As discussed
earlier in this section, this is exactly what happened in earlier years with Africa
and Latin America. The pattern of Pakistani exports faithfully duplicates
Pakistan’s comparative advantage. Pakistan has the world’s second largest salt
reserves and is ranked third in copper reserves, fifth in coal, iron ore and gold
155
reserves, fourth in cotton and milk production, eighth in wheat and eleventh
in rice production in the world (Abbas and Ali 2017b). Changes in the region
have pushed Pakistan further down the agriculture and raw material path.
Thailand dominated as the leading rice exporter to China until 2009. Export
prices of rice from Thailand increased in 2011 and Pakistan was able to benefit
from its low-cost production and emerged as the second largest supplier of
rice to China, after Vietnam.
There was no consistent pattern to the tariff reductions Pakistan had
gained in negotiation with China in securing the 2006 FTA. Goods in the
zero-rated category included cotton fabrics, marble, leather articles and
medical appliances, which were all sectors where Pakistan had well-established
production and export capacity. Sectors where Pakistan has no competitive
advantage and no prospects of exporting to China, such as telephone sets,
digital cameras, electrical machines and children’s toys, were also zero-rated
for export to China. Other well-established sectors in Pakistan, including fish,
cotton, paper, plastics and textile items, were not given any tariff concessions.
Some crucial inputs for Pakistani industry, such as electric and electronic
products, machinery, chemicals and other raw materials, were given zero tariffs,
while others, including woven fabrics, synthetic fibres, paper and paperboard,
and machinery products, were given no tariff concessions (Kamal and Malik
2017). Cotton yarn exports from Pakistan were subject to a tariff of 5 per cent,
which was not scheduled to be reduced during the first phase of the Pakistan–
China FTA. The polyester sector, including fabrics and garments, have been
put in the ‘No Concession List’ and no duty reduction took place for at least
five years after 2006 (Shabir and Kazmi 2007). Pakistani dried fruits faced a
tariff rate of around 25 per cent, semi-milled or wholly milled rice/broken rice
65 per cent, footwear with wood base/metal toe-caps 24 per cent and men’s or
boys’ garments of cotton 16 per cent (Kamal and Malik 2017).
There is no evidence that Pakistan was able to take advantage of the
tariff concessions it was granted by China. There were only 253 tariff lines
where Pakistan exported at least $500 of goods to China. This was around
3.3 per cent of the total tariff lines (7,550) on which China had granted
concessions to Pakistan (Ahmad 2017: 89). For denim and surgical goods,
Pakistan was subject to zero tariffs, which were phased in during 2007–10.
These are sectors in which Pakistan has been successful in upgrading and has
significant overseas markets. Even in these sectors, Pakistan has failed to make
inroads into Chinese markets. During 2006–12, denim exports increased from
156
$100,000 to $4.6 million and surgical goods from $700,000 to $4 million, in

both cases representing only a tiny fraction of output in Pakistan (Chaudhry,
Jamil and Chaudhry 2017). By 2017 the situation had hardly changed.
In surgical instruments, China imported $2.46 billion from the rest of the
world, while Pakistan exported $361 million to the rest of the world and only
$1.8 million to China (Pakistan Business Council 2019: 94).
China by contrast has been well able to make use of tariff concessions
given by Pakistan. In 2017, of the $15.4 billion of Chinese imports to Pakistan,
84 per cent of imports enjoyed zero or low tariffs or significant tariff reductions
(Pakistan Business Council 2019: 21). Products on which Pakistan has
granted zero tariffs for China include semiconductor devices ($659 million
of imports), mobile telephones ($509 million), flat-rolled products of alloy
steel ($402 million) and water-tube boilers ($326 million) (Pakistan Business
Council 2019: 22–3).
When China signed its FTA with Pakistan in 2006, it was one of only three
FTAs that were operational. Pakistan was nothing special, as by 2007 China
was negotiating further FTAs with another 27 countries (S. Kumar 2007: 768).
Pakistan’s margin of preference was being eroded from the moment it was
granted in 2006 as Chinese free trade spread to other countries. Pakistan lost
preference on 79 per cent of its exports to China after China signed FTAs with
other countries, especially with the ASEAN (Association of South-east Asian
Nations) countries (Abbas and Ali 2017b). Tariffs were reduced by 20 per cent
across a range of goods in which Pakistan has established production capacity,
including readymade garments, honey and some fruits. Despite reductions,
tariffs on these goods remained between 12 and 14 per cent, whereas the
same exports from ASEAN countries were zero-rated (Chaudhury, Jamil and
Chaudhry 2017). As of 2013, Pakistan’s exports were still subject to higher
tariffs than those from the ASEAN countries in sectors such as cereals, dairy
produce, meat, fruits, wool, cotton, knitted or crocheted apparel, and made-up
textiles (Abbas and Ali 2017b). China and Australia entered into an FTA
in 2015. The scale of their trade ties dwarfed those connecting China and
Pakistan. In 2017 China imported $116 billion from Australia and by 2017
had a stock of $65 billion of FDI in Australia. Australian exports that will be
zero-rated for tariffs entering China by 2026 and will be directly competitive
to Pakistani exports include dairy, wool, coal, copper, car parts, plastic products
and iron ore (Government of Australia 2018). Table 6.1 shows that tariffs
on Pakistani exports remained higher for crucial potential exports than
competitor countries across ASEAN, New Zealand and Australia.
157
Table 6.1 Tariff comparison of Pakistani exports to China, selected products, 2013
Chinese tariff rates (per cent)

Product ASEAN New Zealand Australia Pakistan
Cotton yarn 0 0 3.9 3.5
Rice and other varieties 20 n/a n/a 65
Fresh or frozen fish 0 0 6 8
Prepared leather sheep/lamb 0 0 4.8 5
Leather prepared after tanning 0 3.1 8.4 9.8
Woven fabrics of cotton 0 0 6 0
Dried fruit 0 0 15 24.2
Source: Shafqat and Shahid (2018: 77).
Rather than sitting passively and watching its trade preferences with China
being eroded, Pakistan should remember that the CPEC can be more than just
economic ties with China. Pakistan can and should proactively use the CPEC
to reach out to other economic partners. Some progress was made at the 7th
Coordination Committee ( JCC) meeting on the CPEC in 2017 when China
and Pakistan officially invited Afghanistan to join the CPEC. Despite its
geographical proximity and linguistic and cultural links, in 2016 Afghanistan
took less than 7 per cent of Pakistan’s exports. Greater Pakistan-promoted
connectivity with Afghanistan could be extended onwards to use the Wakhan
corridor to connect Pakistan to energy-rich Central Asia (Ahmad 2017).
At the moment, there is only a tiny volume of trade between Pakistan and the
Central Asian Republics (CARs); the largest is Kazakhstan, where the total
trade volume reached $27 million in 2016, which comprised only 0.13 per cent
of Pakistan’s trade (Ahmad 2017). The five CARs (Kazakhstan, Kyrgyzstan,
Tajikistan, Turkmenistan and Uzbekistan) have a combined population of
61 million and after prolonged stagnation started growing rapidly in the early
2000s. The region’s trade expanded by around 19 per cent per annum in the
decade to 2010, when CAR imports reached $56 billion with two-thirds
of these coming from neighbouring countries. Russia has been replaced by
China as the major source of imports; Iran and Turkey also saw gains. Imports
from Pakistan actually declined, by around 50 per cent, between 2000 and
2010. Trade potential is apparent. In 2010 Pakistan exported to the rest of
the world significant quantities of four from ten of the main items that the
CARs imported from China and Turkey. The most important of these were
158
clothing and accessories and textile yarn and fabric. Non-metallic mineral
manufactures such as cement and miscellaneous manufactured goods were
also potential export items from Pakistan (Hamid and Hayat 2012: 288).
Not surprisingly, there have been calls from Pakistan and beyond to alter
the structure of the FTA. The influential Pakistan Business Council estimated
in 2016 that there were over 250 products across four key sectors (textiles
and clothing, hides and skins, food products, metals) in which Pakistan has
a comparative advantage and so pushed for favourable tariff concessions
for these goods (Shafqat and Shahid 2018). A later report by the council
recommended a specific focus on securing better market access for men’s or
boys’ trousers, rice products, women’s or girls’ blouses, t-shirts, men’s or boys’
shorts, and footwear (Pakistan Business Council 2019: 47–8). There was some
good news for Pakistan when the second phase of the FTA negotiations was
concluded in September 2017. By the end of this phase, the FTA with China
covered more than 7,000 tariff lines with zero tariffs. Pakistan has secured
market access for immediate exportable products such as blended fabrics,
cotton fabrics, synthetic yarn and fabrics, home textiles, minerals, cutlery, sports
goods, surgical goods, mangoes and industrial alcohol. China has further eased
the export of Pakistani agricultural products, especially citrus and mangoes.
Agreements reached between the Chinese General Administration of Quality
Supervision, Inspection and Quarantine (AQSIQ) and Pakistan’s Ministry
of National Food and Security (MNFSR) mean that exports from Pakistan
can now proceed with time-saving pre-clearance. This measure could benefit
fruit growing areas, including Gilgit-Baltistan (Abbas and Ali 2017b). More
remains to be done to clear away some of the non-tariff barriers to trade. Trade
between China and Pakistan via the Karakoram Highway has suffered due to
logistical bottlenecks. Transport trucks travelling from Pakistan to China are
stopped at Tashkurgan for offloading onto different trucks to complete the
journey to Kashgar. Visas have remained a problem for Pakistani transporters,
causing delays (Shafqat and Shahid 2018).
The US–China trade war could also offer opportunities for Pakistan.
Pakistan has some export presence in Chinese export sectors hit by US tariffs,
including knitted or crocheted fabrics, carpets, synthetic yarn and woven
fabrics. Exports in these sectors to the US could be expanded (Pakistan Business
Council 2019: 56). Trade liberalisation and even CPEC-style infrastructure
represent at most expanded opportunities for firms to invest more, produce
more and export more. As discussed in Chapter 5, so far it is mainly Chinese
159
firms that are taking advantage of those opportunities. Figure 6.1 confirmed
this and demonstrates the boom in Chinese exports to Pakistan. Recall all
of those other constraints on firm growth that were discussed in Chapter 5.
Pakistan needs more than better infrastructure or freer trade to energise its
economy.
IN SUMMARY
China has had a massive impact on world trade. An explosive growth of
both Chinese exports and imports from the 1980s onwards transformed the
global trading environment. Production and exports in both Africa and Latin
America have been pushed out of manufacturing and towards raw materials
and primary products. This process has been largely ignored by the CPEC
scholarship, which focuses instead on the opportunities offered by Chinese
FDI and markets. The FTA signed by China and Pakistan in 2006 has helped
China boost exports of manufactured goods to Pakistan, but Pakistani exports,
which comprise mainly cotton yarn, to China have stagnated. This is not
surprising. Chapter 5 demonstrated that there are numerous constraints on
industrial growth in Pakistan that will not be tackled by the CPEC. This
chapter has shown that the 2006 FTA continued to deprive Pakistan of market
access in various sectors in which it had export potential. The margins of
preference given to Pakistan were significantly eroded by FTAs China signed
with other countries in the same period. The experience of nineteenth-century
Germany (Chapter 4) and contemporary Ethiopia (Chapter 6) illustrate how
an industrial policy may be utilised to maximise the domestic benefits from
big infrastructure investment. Chapter 7 now asks whether Pakistan has a
state that is capable of pursuing a developmental vision.
160
7
The Will of the Dragon

The Importance of an Industrial Policy
The review of historical and contemporary case studies in Chapters 3 and

6 revealed some important lessons for Pakistan. An industrial policy is
crucial to promote both domestic economic spillovers from infrastructure
investment and local industrialisation from SEZs. An industrial policy can
help ensure that spillovers remain within the domestic economy rather
than leak out abroad. This chapter examines the economic theory regarding
relevant market failures; reviews some historical and contemporary examples
of industrial policy; studies the recent calls in Pakistan for an industrial policy
to complement the CPEC investment; inspects whether the state in Pakistan
has the capacity to implement developmental-type interventions; looks at the
politics of infrastructure; and examines how the CPEC will be financed.
FROM MARKET FAILURES TO AN INDUSTRIAL POLICY

The economics of neoliberalism are structured around the assumption of
an efficient, spontaneous and self-organising market economy. The state
should preserve free markets from any inclinations towards monopoly, issue
a currency and protect property rights according to clear rules rather than
discretion. The government is likely to act according to political motivations
and will lack the necessary information to intervene efficiently (Wade 2012).
Neoliberal thinking does not pay particular attention to economic structure,
such as the share of industry in the economy, which it argues will reflect
the free functioning of efficient markets. The implicit view is that with
macroeconomic stability and well-functioning markets, structural change
will be governed by comparative advantage. When an economy is open to
international trade, comparative advantage directs resources to where their
161
contribution to national product is maximised and there is no rationale for a

policy which favour some economic activities over others (Rodrik 2006).
Contrary to the assumptions of neoliberal economics, market imperfections
are likely to be pervasive. Any entrepreneur who invests in a new area of
manufacturing provides a demonstration of its success or failure to other
potential entrepreneurs; the pioneer will generate learning in production
and, with new technology, train workers and managers and provide inputs
for other firms or retail outlets. The social value will exceed the private
return of such investment (Rodrik 2006). The training labour needs to work
in manufacturing (remember the shortages of skilled labour in Pakistan
discussed in Chapter 5) is unlikely to be solved in a free market. If one expends
time and energy in training labour, another firm can save training costs by
poaching those workers. The market-determined supply of skilled workers
is always likely to run below national need. Manufacturing activities tend
to be highly interlinked, and the establishment or expansion of one firm or
sector may require corresponding expansion elsewhere. Who would establish a
shipbuilding industry unless they could be assured of a supply of steel? But in
turn who would go to the great expense of establishing a steel industry unless
they could be guaranteed of a purchaser by the presence of a shipbuilding
industry. This problem is an example of a coordination failure. Neither the
steel nor the shipping entrepreneur would invest unless they could be assured
the other will invest. Contracts could be negotiated and signed between the
steel mill and shipbuilding firms to establish production and agree to buy/sell
a particular amount of steel, but such contracts are likely to be too costly to
draw up and monitor. The economy can become stuck in an equilibrium of
under-industrialisation. During the early stages of industrialisation, there is
a need for the state to coordinate complementary investments (Rosenstein-
Rodan 1943; Scitovsky 1954) so that interdependent investment projects are
implemented at the same time. In 1960s and 1970s Taiwan, the solution to the
coordination problem was state investment. State enterprises invested directly
in sectors where the scale of production was beyond the financial reach of
private entrepreneurs and where there were significant potential spillovers
with the wider domestic economy. These sectors included petroleum refining,
petrochemicals, steel and other basic metals, shipbuilding, heavy machinery,
transport equipment and fertiliser (Wade 1990: 179). These market
imperfections are not isolated instances but are characteristic features of what
it means to be underdeveloped (Rodrik 2008).
162
THE WILL OF THE DRAGON
The second key rationale for an industrial policy is the empirical evidence
that emphasises the importance of industrial growth and product diversification
in promoting economic growth (Rodrik 2006). Nicholas Kaldor (1967) made
an influential case for the ‘superiority of manufacturing’. Kaldor’s ‘First Law’
captured the relationship between industrial growth and total GDP growth.
The law states that industrial growth has a positive relationship with economic
growth. Evidence across 50 countries and 29 Indian states for the 1990s
supports the first law (Dasgupta and Singh 2005, 2006). Kaldor’s ‘Second Law’
argues that there is a strong and positive relationship between output growth
and productivity growth in the manufacturing sector. The rapid industrial
growth after c. 1960 in Japan, Singapore and South Korea, for example,
was closely linked to the rapid growth of productivity in those countries.
The slowdown in industrial and GDP growth across many developed countries
after 1973 was clearly linked to a slowdown in productivity growth. This
dynamic is driven by dynamic economies of scale. Dynamic economies of scale
imply that as producers accumulate experience in production, they improve in
skills and capabilities and so learn to produce with more efficiency. Over time,
average costs will decline and productivity will increase. There is evidence
that productivity growth by country and also across different Indian states
varies positively with the expansion of the industrial sector (Dasgupta and
Singh 2005). Kaldor’s laws suggest that rapid and sustainable growth of
GDP is likely to be associated with a structural shift of the economy towards
industry. These laws show that the industrial sector is the dynamic centre of
technical change and productivity growth. A more industrialised country will
experience an increasing economic lead over non-industrial countries.
Merely manufacturing simple, low-technology, labour-intensive products
such as cheap textiles is not enough. Such sectors are vulnerable to competition
from new entrants, and the intensely competitive markets will likely be
associated with declining product prices. China, India, South Korea, Taiwan,
Singapore and Malaysia have all sustained growth by being able to diversify
into more sophisticated, technically demanding activities (Rodrik 2006).
Since the 1980s, there has been a clear and positive relationship between the
degree of technological complexity of exports and the growth rate of such
exports on world markets (Lall 2000: 344). There is a robust and positive
relationship between the level of a country’s export sophistication in 1992 and
subsequent economic growth up to 2003 (Rodrik 2006: 9). These differential
growth rates resulted in significant changes in the structure of world trade.
163
The share of world trade accounted for by low- and medium-technology

exports in the 1980s and 1990s stagnated or declined while the share
accounted for by high-technology exports increased steadily (Lall 2000: 351).
Another reason why manufacturing promotes growth is that specialisation
based on manufactured goods provides a better platform for jumping onto new
economic activities with unexploited productivity potential. A country with a
broad-based manufacturing sector is more likely to be able to take advantage
of new opportunities than one that had specialised in a few primary-based
products (Rodrik 2006).
Many models of technological transfer assume that technology is freely
available to all countries/firms at a given market price. Countries with an
abundance of cheap labour will select more labour-intensive technology. Once
selected, the new technology can be easily absorbed through a predictable
and automatic period of learning (Lall 1992, 1994). In reality, markets
within which international technology transfer takes place are subject to
various market failures. First, the market for technology is characterised by
asymmetric information. Those supplying technology cannot reveal all the
details of the technology without giving away trade secrets. How then do
purchasers of technology realise its value to them and agree on a purchase
price? Second, intellectual property rights are likely to give owners of new
technology monopoly market power. Monopoly can be used to exploit buyers
of technology. Third, learning will generate externalities. As one firm adopts
and successfully utilises a new technology, other firms will be motivated to
utilise the same technology, as evidence of its success has been demonstrated.
These wider social benefits of the initial learning will not be taken into
consideration by the pioneer firm. Fourth, contrary to the simple models,
technology has to be physically used in order to master its functioning.
This implies that firms may only really understand their own technology
and have little idea about alternative technologies. Any effort to utilise new
technology will require a period of learning and adaptation (Lall 1992, 1994;
Hoekman, Maskus and Saggi 2005).
With market failures, structural change and rapid economic growth are
not an automatic process. The task of the government then is to promote
structural change, which involves producing new goods with new technologies
and transferring resources from traditional to new activities (Rodrik 2008).
An industrial policy encompasses ‘policies that stimulate specific economic
164
activities and promote structural change’ (Rodrik 2008: 3). Industrial policy
need not be about industry and could, for example, be targeted at non-
traditional agriculture or services. Public subsidies to promote the adoption
of new agricultural crops, for using high-yield variety seeds, or for promoting
tourism would all be examples of industrial policy (Rodrik 2008). Industrial
policy can refer to a joint government and private sector diagnosis as to discover
the origins of constraints to growth and the proposed solutions (Rodrik 2006).
Solutions can include promoting investment in new industries or making
credit more easily available for new technology, encouraging the provision of
complementary services and inputs such as electricity, transport infrastructure
and phytosanitary testing services, or training workers and managers (Rodrik
2008). The government does not have to ‘pick winners’ but can instead
create public–private institutions where information on profitable activities
can be shared and potential means of intervening to remove constraints can
be considered (Rodrik 2006). Even those who accept all these arguments
may be cautious over the potential for industrial policy, generally or in the
case of specific countries. Governments may lack the capacity to identify
with any precision the relevant firms, sectors or markets that are subject to
market imperfections. The effort to construct an industrial policy may leave
governments subject to political pressures to subsidise and protect firms or
sectors for political, corrupt and other non-economic reasons. This may then
divert the efforts of entrepreneurs to seeking favours from the government
rather than boosting sales and acquiring new technologies (Rodrik 2008).
J. Y. Lin (2010) listed six steps to help the state pick sectors for assistance.
He argued that the government should start by listing the goods and
services produced by successful developing countries over the previous
20 years that have similar endowment structures and an average GDP
100 per cent higher. For Pakistan, this list would likely include Vietnam, India
and Malaysia. The government should then select those goods and services
from the list in which domestic firms have already established some local
production capacity. Interventions should focus on removing constraints to
growth in existing firms and upgrading their technology and in helping other
firms establish new production capacity. If there is no domestic production
of particular goods or services, then the government should consider
making efforts to attract FDI from higher-income countries ( J. Y. Lin 2010;
Wade 2012).
165
HISTORICAL EXPERIENCES OF INDUSTRIAL

POLICY AND INFRASTRUCTURE
The classic cases of industrial policy are those countries who successfully
promoted rapid industrial (and usually export) growth after 1945. Between
the 1960s and 1990s, East Asian countries used a mixture of incentives and
discipline in their industrial policies. Tax incentives (Taiwan) and credit
subsidies (South Korea) were generous and were conditioned on performance,
and especially on export performance. Non-abiding firms were penalised
by withdrawal of subsidies and in other ways. Latin America, by contrast,
in promoting import substitution relied on incentives (trade protection and
cheap credit) but failed to exert discipline on the beneficiaries, which helps to
explain the generally lower investment and growth in that region. The more
germane historical examples for Pakistan are those case studies of industrial
policy that focused on trying to maximise the benefits of infrastructure
investment for promoting domestic industrialisation.
Until 1834 the German customs union (Zollverein) followed a policy of free
trade with regard to pig iron, which was treated as a raw material and therefore
not subject to import duties. As railroad construction began in the 1830s and
accelerated into the 1840s, factories producing inputs such as rails relied on
imported pig iron, mainly from Britain. In 1851 the tariff on imported pig
iron was raised to 51 per cent; the tariff on imported rails was also increased to
protect domestic producers against the higher costs of domestically produced
pig iron (Fremdling 1977). Simply changing tariffs to reduce imports of
industrial raw materials was not enough; the growth of industry would likely
have been undermined by expensive and poor-quality locally produced inputs.
The German government in the nineteenth century went beyond supporting
law and order and external security and utilised an industrial policy to help
ensure German industry could produce efficiently and at low cost. The state
supported a system of universal elementary education and a considerable part
of higher education. The government also took over the responsibility for
providing transportation and communications and in conjunction with the
private sector running banking and insurance and some industrial enterprises
such as coal and potash (Bowen 1950). The state took on a more directive and
guiding role in relation to domestic industry after 1840. The state facilitated
the international transfer of skills, technology and knowledge. It promoted
infant industries through tariff protection, state investment, public–private
166
cooperation, scholarships to promising innovators, subsidies to competent

entrepreneurs and directly facilitated the organisation of new machinery
and industrial processes (Chang 2002). Foreign technology was transferred
through leading German manufacturers travelling and studying in France and
Britain and also though employment of French, Belgian and British engineers
and skilled workers (for example, puddlers and roll-masters) and through
foreign investors founding firms, especially French ironmasters in the Ruhr
and Rhine area (Fremdling 1977).
In nineteenth-century Russia, the demand for track materials was largely
satisfied by imports of good quality and inexpensive rails from England.
Factories in the Urals developed a small capacity to supply high-quality iron
for the construction of locomotives and railway rolling stock. The construction
of the national railway system from the 1850s was met by imports of rails, and
tariffs were reduced to make those imports cheaper. After 1868 rails rolled
in Russia gained both a government subsidy and tariff protection; domestic
production soared and gained a 60 per cent market share by 1880. Trade
protection did not just create an inefficient and high-cost domestic industry
but had come on top of two decades of continuous technological innovation
after the 1860s. Government investment spending was a crucial driver in the
momentum of industrialisation. Domestic steel production in the late 1880s
saw a slowdown and a revival after 1893, mirroring patterns of government
spending, and another revival after 1908 linked to government spending on
rearmament. The government contract for 100,000 tonnes of rail given to
the Donets Steel Company in 1891 gave the firm confidence to invest in
new capacity and technology. The creation of new capacity after the early
1890s, including the three giant works, Kolomensk, Briansk and Putilov, gave
Russian producers a 50 per cent share of the total Russian ferrous metals
market in 1896–1900 and around 80 per cent of locomotives and rolling
stock during the late 1890s. While the industry was often under pressure,
the combination of technological progress and government intervention
had created an industry that was showing signs of self-sustaining growth.
The firm Briansk started producing its own pig iron after opening a blast
furnace at the Alexandrovsk factory in 1887. Soon after, the factory was forced
to discontinue the production of rails due to cutbacks in government orders
for railway equipment. The capacity already created allowed the factory to
switch to other steel products for non-rail markets instead (Blanchard 2000).
167
The case of contemporary Ethiopia is even more relevant for Pakistan.

The government of Ethiopia sought to boost domestic industry in the context
of massive Chinese investment in infrastructure and growing exposure to
competition from Chinese imports. The successful revival of the footwear
sector (and growth of floriculture and cement) in Ethiopia despite the
pressures of import competition from China can be explained in reference to
the ‘developmental state’ in Ethiopia. Between 2000 and 2013, the economy
experienced near 10 per cent annual GDP growth, not based on oil or other
commodity exports, but by a shift into manufacturing. The local leadership
had a clear and politically driven desire to promote rapid economic growth.
The drive came from the developmental ideology of Meles Zenawi, ‘one of
the most remarkable leaders and original thinkers in independent
Africa’ (Clapham 2018: 1153). This vision has been expressed in a series
of developmental plans, including the Plan for Accelerated and Sustained
Development to End Poverty (2005–10) and the Growth and Transformation
Plan (2010–15) (F. Nicolas 2017). There was a strong state capable of
implementing this vision. Ethiopia has a long-established history of statehood
that has been both hierarchical and intolerant of dissent and held together by
a long history of and strong sense of nationalism. The state was dominated
by a military leadership transitioning to civilian rule. This was similar to the
state in Taiwan and South Korea during the 1960s onwards. The Ethiopian
state was able to successfully mobilise resources through domestic savings, the
domestic banking network and external aid from both China and the west. Aid
came on favourable terms as Ethiopia was able to diplomatically project itself
as a regional force for stability despite human-rights questions, as a leading
partner in the war on terror, supplying peacekeeping forces, as a mediator in
Somalia and Sudan, for success in achieving improved measures of human
development and efficient governance in the use of resources. The relative
abundance of resources gave the Ethiopian government a clear autonomy in
pursuing its own development agenda and to pursue its activist industrial policy
(Clapham 2018). Ethiopia has not let China dominate. Between 1992 and
2016, although China was the largest cumulative foreign investor (measured
by licensed FDI projects), it only comprised 21.6 per cent of the total. Saudi
Arabia (18.8 per cent), Turkey (8.3 per cent), India (6.3 per cent) and Britain/
Netherlands (5.3 per cent) also made significant contributions. The total
amount of employment generated by British investment was significantly
higher than Chinese investment (F. Nicolas 2017: 18).
168
Does Pakistan have a capable and developmental state? Has Pakistan

been able to play a clever diplomatic game to maximise its own freedom
of manoeuvre and mobilise resources and other assistance from a variety
of sources?
PAKISTAN: THE NEED FOR AN INDUSTRIAL POLICY

In reference to the theory of industrial policy, there is evidence from a
diagnosis of economic development in Pakistan that the country is in need of
an industrial policy. The arguments of Chapter 2 (and Table 1.1) showed that
economic growth has been both stable and relatively rapid. But, as this section
shows, Pakistan has experienced slow and declining productivity growth, slow
rates of structural transformation and little technological upgrading.
Productivity growth during the 1960s was quite rapid in sectors such as
electrical machinery, tobacco, textiles, printing and publishing, and rubber
and textiles, and only in the paper industry was productivity growth negative
(Cheema 1978). An index of productivity growth (total factor productivity,
or TFP, measured in two ways) for Pakistan confirms this rapid growth from
the mid-1950s to the early or mid-1970s, then continuous decline until the
early 1990s (Wizarat 2002: 76–7). Table 7.1 shows that between 1970 and
2011, TFP growth in Pakistan has been low; of particular concern is the sharp
slowdown of TFP growth in Pakistan in recent years while TFP growth has
tended to accelerate in comparator countries.
A crucial part of this general failure of productivity has been the slow
growth of labour productivity. Table 7.2 shows that labour productivity growth
in Pakistan has long been much slower than comparator countries and has
slowed down in recent years.
Table 7.1 TFP growth, 1970–2011
Country 1970–2011 (%) 2005–11 (%)

Pakistan 1.4 0.6
Thailand 1.8 1.3
China 3.2 4.2
India 1.4 3.9
South Korea 1.7 2.4
Source: Asian Productivity Organisation (2013: 76–7).
169
Table 7.2 Labour productivity growth, 1990–2011
Country 1990–2000 2000–11

Pakistan 1.9 1.3
China 8.9 9.3
India 3.0 5.5
Thailand 3.9 2.2
Bangladesh 3.3 1.0
South Korea 5.2 4.2
Source: Asian Productivity Organisation (2013: 66).
Specifically in manufacturing, Raheman et al. (2008) find that TFP growth

was very slow (0.9 per cent per annum) between 1998 and 2007. Importantly
for Pakistan, the key textiles sector suffered negative TFP growth in weaving,
spinning and composite sectors (Burki and Terrell 1998). Another study finds
that technical efficiency did improve slowly over the 1990s in diverse sectors,
including textiles, food manufacturing, industrial chemicals, iron and steel,
drugs and pharmaceuticals, electrical machinery and non-electrical machinery.
But the study also found big gaps in efficiency remained in these sectors by
the early 2000s. In other sectors, such as glass and glass products, transport
equipment, tobacco manufacturing, non-metallic mineral products and other
chemical products, technical efficiency continued to decline (Din, Ghani and
Mahmood 2007). Productivity in textiles was constrained by the absence of
skilled workers in design, planning and production, the lack of experienced
and trained mid-level managers, the small-scale and fragmented structure of
the industry reducing economies of scale, and access to finance to acquire new
technology (Pakistan Business Council 2018: 40–1).
The second symptom suggesting the need for an industrial policy was the
slow rate of structural change. By the mid-2000s, the service sector accounted for
about 50 per cent of Pakistan’s economy, and agriculture and industry about 25 per
cent each. The share of employment in agriculture had only declined slowly over
time, from 60 per cent in the 1980s to 40 per cent in the mid-2000s. Employment
growth was slower than output growth in services and by the mid-2000s, the
sector employed around 30 per cent of the labour force. While a cross-sectional
regression showed that Pakistan’s output share in industry in 2004 was about that
expected for a developing country of Pakistan’s GDP per capita (Felipe 2007: 7),
Pakistan needs industrialisation to sustain a higher trajectory of GDP growth.
170
The share of manufacturing (a subset of industry) in GDP did increase

significantly in countries such as Indonesia, Malaysia and Thailand over the
last 30 years. In Pakistan, the share has remained at around 15 per cent since
the 1970s. The manufacturing sector in Pakistan is heavily dependent on food
and beverages and textiles. These two categories comprised about 58 per cent
of the sector’s value-added in the 1970s, with industrial chemicals only
about 11 per cent, and electrical and non-electrical machinery plus transport
equipment only 8 per cent. By the 1990s, food and beverages plus textiles
still represented 48 per cent of the total manufacturing output. The share
of industrial chemicals had increased to 15.5 per cent, and that of electrical,
non-electrical machinery plus transport equipment represented 11 per cent
of the total. These were very slow structural shifts within manufacturing
to areas of higher technology. In Malaysia, the share of these last three had
increased to about 40 per cent of the manufacturing output (Felipe 2007: 15).
The manufacturing sector in Pakistan is relatively low-technology.
The share of medium- and high-technology is well below 10 per cent of the
total and has been stagnant since the 1970s (Felipe 2007). The export mix
in textiles has long been dominated by low-end products like t-shirts and
vests. The average value of Pakistan’s top five readymade garment exports was
$5 per unit in 2017, significantly below its competitors; in Bangladesh it was
$15–20 and India $5–10. There was some good news in that the share of (low
value added) cotton yarn in Pakistan textile and garment exports declined from
19 per cent in 2012 to 10 per cent in 2017, while the share (slightly higher value
added) knitted apparel increased from 17 per cent to 20 per cent and woven
apparel from 14 per cent to 20 per cent (Pakistan Business Council 2018: 24).
FDI may lead to technology transfer to local firms through imitation
or copying of that technology. Workers trained by an MNC may transfer
knowledge to a local firm or start their own firms and take with them the relevant
technological know-how. Research has found that the magnitude of positive
spillovers from FDI depends on local endowments of skills, the capability of
local educational and research institutions, the technological capability of local
firms and the provision of infrastructure (Lall 1992; Gorg and Greenaway
2004). There is also ample evidence that successful developing countries have
long leveraged FDI as a source of technology transfer through active industrial
policies. In post-war Japan, government approval was required for the import
of technology and almost always had to come through the formation of a
joint venture between a Japanese firm and the foreign technology supplier.
171
The government exerted pressure to lower royalty payments, ensure Japanese

nationals acquired a thorough understanding of the new technology and that
Japanese managers attained senior positions in the joint venture ( Johnson
1982). South Korea learned much from Japan as it launched its own rapid
growth drive from the mid-1960s. Firms wishing to invest in Korea were
usually compelled to enter into a joint venture with a Korean firm. Joint
venture agreements would specify levels of capital investment, output and
export targets, the type and level of technology transfer, the provision of raw
materials, the access to foreign markets that the foreign investor will provide,
and an agreement on the sale of foreign-held equity to the domestic partner
to enable a transfer of financial and operational control over the enterprise.
After learning relevant skills, further expansion in that sector would be carried
out by the domestic partner and that sector would then be closed to FDI
(Mardon 1990). In Singapore, the government Economic Planning Board
(EPB) maintained overseas offices to monitor foreign firms from which FDI
could provide important gains for the local economy in the form of value-
added, skill content and capital intensity. The EPB would approach such
firms and negotiate directly about the incentives necessary to attract them to
undertake FDI (Huff 1995). Throughout the 1980s, the EPB pushed foreign
firms in Singapore to upgrade the technological complexity of production
(Huff 1995; Ermisch and Huff 1999). During the 1990s in Ireland, the state
used a mixture of generous tax incentives, a world-class telecommunications
system and an ample supply of technical and technologically educated labour
to attract FDI and to later upgrade production. Spending on research and
development (R&D) among IT and telecommunications companies such
as Digital, Amdahl, IBM, Siemens Nixdorf, Ericsson and ATT/Lucent
Technologies increased steadily throughout the 1990s (Ó Riain 2000).
Many authors agree that developing countries such as Pakistan are today
severely constrained in their ability to leverage FDI to promote technology
transfer. Wade argues that developing countries are now more ‘tightly
constrained in their national development strategies by proliferating regulations
formulated and enforced by international organisations’ (2003: 621). Chang
(2002) argues that the ‘good governance’ agenda of the IMF, World Bank and
WTO focuses on reducing deficits, opening up to free trade and privatisation,
and so prevents contemporary developing countries from promoting
industrialisation and upgrading into new technologies. The Agreement on
Trade-related Investment Measures (TRIMs) was a product of the Uruguay
172
Round of the General Agreement on Tariffs and Trade (GATT)–WTO trade

negotiations in the mid-1990s. TRIMs aimed to remove those ‘trade and
investment distortions’ which, argue Chang and Wade, had historically been
used to great success by the now-developed countries to boost investment and
technological acquisition by national firms. One important example was the
local content agreements that compelled FDI firms to source an increasing
proportion of their industrial inputs over time from local firms.
These concerns are overstated and there remain various other routes to
promoting domestic industrialisation and technology acquisition (Di Caprio
and Amsden 2004). Developing countries can support domestic enterprises
in the name of ‘promoting science and technology’. Various subsidies are
permitted for R&D that leads to innovation and technological upgrading,
for targeting backward sub-national regions and for environmental reasons
(Weiss 2005). The main constraint is not international law but domestic
politics—‘the most coercive part of the new international economic order
is informal’ (Amsden and Hikino 2000: 110). Rarely do contemporary
developing countries (certainly including Pakistan as we shall discuss in the
next section) have a committed faction among the political and civil service
elite capable of articulating and promoting an industrial policy to promote
technology absorption.
A related debate concerns the ‘race to the bottom’, wherein developing
countries compete to host FDI by trying to offer better tax, labour or
environment concessions than rival hosts. This leads to a dynamic in which
much of the benefit of the investment is gained by the foreign MNC. The
success of countries such as Ireland and Singapore in attracting FDI has masked
the lack of benefits to the hosts. Recall as well the concerns about whether
those SEZs established in China after 1979 generated net benefits to China.
In Singapore, the Jurong Town Corporation, a state-owned development
corporation, leased land for 30 years at a very low cost to FDI firms, frequently
with an easy option to renew for a further 30 years. The government also
provided modern infrastructure including a port, airport, telecommunications
and a mass rapid transit system and sometimes even purpose-built factories.
From the early 1980s, the government began to spend heavily on education
and training geared to the labour/skills needs of FDI firms. In return for this
largesse, foreign firms gave little back. Ever-extending tax concessions for FDI
in Singapore meant that, by the 1990s, no tax was paid on the profits from
almost two-thirds of Singapore’s manufactured output, and three-quarters of
173
direct exports (Ermisch and Huff 1999; Huff 1995). MNCs have used their
global presence to exploit loopholes and reduce their tax burden even further.
In 2012/13, Amazon UK had sales of £4.2 billion on which they paid taxes of
only £3.2 million. Amazon.co.uk has classified itself as a service provider to
the Luxembourg-based business Amazon EU, to which it pays large fees, thus
reducing its UK profitability and tax liabilities. The large profits earned by the
Luxembourg business are subject to much lower rates of taxation than those
prevailing in the UK.
Pakistan in the recent past has had some opportunities to leverage FDI to
acquire new technology. Figure 7.1 shows there was a boom, if temporary, in
inflows of FDI to Pakistan after 2003.
As noted earlier, technology transfer from FDI has been found to be
dependent among other things on the stock of educated and skilled labour.
In 2003–04, as the boom in FDI started, only 52 per cent of the population
of Pakistan aged 10 or more years were literate, including 40 per cent of
females (S. H. Khan 2009). There are long-standing concerns with the quality
of education beyond this starting point of basic literacy. The 2003 Learning
and Educational Achievement in Punjab Schools (LEAPS) project found that
by grade 3, less than 20 per cent of their sample of 12,000 children could
understand a simple written sentence in the local language and less than
10 per cent could graphically represent simple information with bar charts.
More than 70 per cent internationally were able to reach these standards.
Figure 7.1 Pakistan, foreign direct investment, net inflows (current US$)
174
The summary of education in Pakistan was dire, and ‘close to one-half of

the three million born in Pakistan will leave school unable to add, subtract,
multiply or divide, unable to read and write simple sentences in Urdu; and
unable to read a short word like “BALL” in English’ (Das, Pandev and Zajonc
2012: 232). Pakistan also fails with regard to the vocational education that
could have created the practical skills needed to absorb technology from FDI.
By the mid-2000s, Pakistan had 3,125 technical and vocational institutions
with a total enrolment of 256,000. The curriculum at these institutions had
little relevance to the needs of an evolving labour market and practical training
was nearly absent. The oldest institute in Pakistan was the Government
Polytechnic Institute in Karachi, which offered training in 17 different trades
but by mid-2009 was in a state of disrepair (S. H. Khan 2009).
INDUSTRIAL POLICY, PAKISTAN AND THE CPEC

Motivated by the economic failings related to productivity, structural change,
technological upgrading and learning (from FDI) failures, there has been a
general call for Pakistan to utilise an industrial policy; some of this has come
about completely independently of thinking about the CPEC. Speakman et al.
(2012) show that there is substantial variation in productivity levels between
firms in the same country across a sample of developing countries. Pakistan is an
extreme case with almost double the variation as the next country, Philippines,
and far more than the average for other countries in the sample including
Croatia, Bangladesh, Chile, Brazil, India, Indonesia, Mexico and South Africa.
They find that a few Pakistani firms are highly productive and competitive but
the large majority, even in the same industry, are not. Spreading technology
from leaders to laggards provides an obvious and enormous opportunity for
Pakistan. The costs of discovery are already borne as the leading firms have
shown that it works in Pakistan. Chaudhry and Andaman (2014) find that
Pakistan differs from comparable Asian countries because they have continued
to expand exports in simple low-technology textiles rather than raising quality
and moving from low to higher value added exports. The policy difference,
they argue, was the use of a coherent industrial policy in other Asian countries.
They point to a wide range of policies, including exchange rate depreciation,
incentives for FDI, expansion of higher education, and investment incentives
via cheap credit and tariff policies to ease access to imported capital goods, that
were used to promote exports and upgrade technology, and in doing so urge
175
that similar thinking be adopted by Pakistani policymakers. Noman (2015)

argues for an industrial policy in Pakistan specifically to raise investment from
current levels of 15 per cent of the GDP, and to direct investment towards
promoting industrialisation and other activities that lead to learning and
technological upgrading. His most explicit recommendation is to establish
or revive development finance institutions (DFIs) to provide access to cheap
finance to help stimulate investment. In the 1950s and 1960s, two DFIs,
the Pakistan Industrial Credit and Investment Corporation (PICIC) and
the Industrial Development Bank of Pakistan (IDBP), played a vital role in
creating a class of industrial capitalist-entrepreneurs that led a decade of rapid
industrialisation. The case for mobilising development finance to stimulate
investment, he argues, is made more compelling by severe constraints on public
investments on account of the great difficulty Pakistan has had in raising tax
revenue (See Figure 4.1). These recommendations raise concerns about how
revived DFIs can be protected from influence by the corrupt or politically
influential. Such negative influence marred the functioning of the nationalised
commercial banks, especially in the 1980s and 1990s. Aside from lessons from
Pakistan’s own past in the heyday of PICIC and IDBP, Noman points to
the idea of learning from the Brazilian Development Bank (BNDES).
The BNDES managed to implement successful interventions and was an
oasis of relatively efficient developmentalism surrounded by a more corrupt,
inefficient and politically compromised Brazilian state. The Corporaction
Andina de Formento in the Andean region and the Development Bank of
Ethiopia are other examples of possible emulation. The Pakistan Business
Council (2018), which represents the leading private sector firms, argues for
an industrial policy to support the garment sector in particular. They call for a
focus on increasing labour productivity, reducing production costs and boosting
productivity. This, they suggest, should be done through more demand-based
training, better accounting for the risks associated with technology acquisition,
incentives for quality certification, export credit, making customs more
business-friendly and enhanced roles for industry associations. The fact that
the private sector is so explicit in its calls for government intervention (rather
than liberalisation) is interesting, but the list of interventions does feel more
like a wish list of benefits rather than a coherent and feasible industrial strategy.
Others have called for an industrial strategy specifically to complement
and maximise the benefits from the CPEC. Chaudhury, Jamil and Chaudhry
(2017) provide a long list of what that industrial strategy should comprise.
176
The CPEC-related industrial activities, they argue, should have well-defined

local stakeholders such as joint ventures; policymakers need to make pragmatic
decisions as to the sectors Pakistan should focus on based on productivity
and export potential; the CPEC should allow Pakistani firms to move up
the technology ladder; the government should create firm-level incentives
for investment in advanced machinery; the CPEC industrial zones should be
subject to a minimum local content requirement to ensure inputs are sourced
from local firms; Pakistan needs a labour policy to ensure manufacturing
switches from low- to high-skilled labour. Focusing on international trade,
Ahmad (2017) argues that Pakistan needs to utilise the opportunities of the
CPEC to diversify out of cotton textiles to more high value added sectors.
To do so, Pakistan, he argues, should invest in sectors where world demand
is growing and should shift from import substitution to export promotion.
Mehmood (2017) argues that the development of the CPEC SEZs should
be made part of the overall growth strategy of Pakistan. He argues that
Pakistan should make a careful choice of industries to be invited into SEZs,
develop a system where targets with a timeline are effectively monitored to
meet agreed export and local employment targets, encourage Chinese firms
to produce intermediate inputs to be exported internationally or to non-SEZ
companies in Pakistan and production should be primarily for export. Further,
the SEZs should promote good practice in policymaking such as having a
single-window clearance and customs procedures that cut delays, bureaucratic
hurdles, corruption and trade costs for export-oriented industries.
The SEZs should also be supported by good infrastructure and service
provider firms, with the government active in improving transport, electricity,
water, telecommunications, waste disposal and other infrastructure to link
the SEZs with global and local markets. Every zone should have its own
power generation facility and provision of gas, petroleum and other utilities
at internationally competitive prices. The SEZs should also be linked to skills
and technology development institutions of Pakistan and the SEZs should
establish such activities as day-care centres, school facilities, clinics, housing
colonies, shopping malls and restaurants.
Such lists go on and on—they represent little more than aspirational wish
lists and lack any analytical substance about the feasibility of such ambitious
agendas. There is a well-established body of literature that explores the criteria
necessary for a state to be ‘developmental’ in the style of nineteenth-century
Germany or Russia or twenty-first-century Ethiopia and thereby ensure
177
that the benefits from a project like the CPEC promote domestic economic
growth rather than leak out overseas. This literature has been entirely ignored
in the aspirational wish lists of much of the CPEC writing.
The emphasis in industrial policy on firms attempting to acquire and
efficiently utilise new technology or discover new lines of production is an
inherently uncertain process and will generate a significant failure rate. The
question then is less about the ability of the government to pick or promote
winners but rather whether it has the capacity to let a failed firm or investment
project go. Industrial policy needs a means to either make continuing
government support conditional on some pre-defined measures of success or
to have a mechanism to recognise when investment projects are failing and
ensure there is the ability and political will to phase out government assistance
(Rodrik 2008). There are various ways to identify and drop failed investments,
including strictly negotiated conditions attached to state assistance, sunset
clauses—meaning that assistance expires after a defined time period—and
rigorous and regular monitoring and reviews against clear benchmarked goals
(Rodrik 2008). In 1960s South Korea, technology was mainly transferred to
large firms (chaebols) who received various forms of subsidies (often cheap
credit) and protection from imports to give them an opportunity to expand
production. Increased production was closely linked to learning-by-doing,
as subsidies and protection were given conditionally on firms successfully
reaching export targets, reducing costs and absorbing new technologies.
The state maintained a credible threat of removing these incentives should
firms fail to meet their targets. The balance of power between the state and
chaebols was such that inefficient firms were not able to protect their subsidies
if the state wanted to withdraw them (Khan and Blankenburg 2009).
The preconditions that enabled South Korea to do this are usually framed in
terms of it having been a ‘developmental state’.
The literature on the role of the state in economic development falls
into two schools. The first is the ‘economic school’. This effort starts with
theory and empirical evidence to identify a range of policies for which state
intervention can be justified and have proven to be successful. Examples are
often drawn from the East Asian experience. The examples mentioned earlier
show that there is a great deal of such writing regarding Pakistan and long
lists of what the state should be doing. This identifies the economic potential
of a developmental state. The second is the ‘political school’ and focuses on
the capacity of the state to identify and implement such policies (Fine and
178
Stoneman 1996). The ‘political school’ has all but been ignored in Pakistan’s
case. The politics of the developmental state are defined as ‘states whose
politics have concentrated sufficient power, autonomy and capacity at the
centre to shape, pursue and encourage the achievement of explicit development
objectives, whether by establishing and promoting the conditions and direction
of economic growth, or by organising it directly, or by a varying combination of
both’ (Leftwich 1995: 401). These political preconditions include: that leaders
have a politically driven desire to promote growth; that state institutions are
autonomous; the bureaucracy is competent and insulated from politics; that
civil society is weak; and that the state enjoys widespread legitimacy, whether
of the democratic variety or others (Leftwich 1995, 2000).
A key feature of a developmental state is ‘bureaucratic power’. The
ability of the state to promote long-term growth is shaped by the creation
of a ‘powerful, professional, competent, insulated, career based bureaucracy’
independent of the vagaries of short-term politics and able to formulate and
implement economic goals through long-term planning (Leftwich 2000).
Such a bureaucracy is characterised by promotion on merit, good salaries
in comparison with private sector alternatives, usually life-time tenure in
office, clear sanctions for corruption and is often dominated by a planning
agency standing outside and above individual ministries—commonly
known as a ‘pilot agency’ (Leftwich 2000; Doner, Ritchie and Slater 2005).
This point is discussed further in Pakistan’s case in the section titled
‘The CPEC, Governance and the Future of Industrial Policy in Pakistan’.
Often the developmental bureaucracy created an esprit de corps through a
common training process; in France this was through recruitment from the
Grandes Ecoles (Loriaux 1999); in Korea from the Korean Military Academy
(Cotton 1991); and in Japan from the Tokyo Law School (Leftwich 1995).
There are corners of developmental optimism in Pakistan’s bureaucracy.
A wonderful and ingenious study by Aman-Rana (2019) gives a rigorous and
quantitative measure of the degree of meritocracy in one corner of Pakistan’s
bureaucracy. She finds out that among the elite Pakistan Administrative
Services (PAS) in Punjab, junior workers who display excellence in the early
stages of the career (measured in terms of their success in mobilising tax
revenue) are more likely to be subsequently promoted by senior colleagues who
have observed this meritorious performance. Seniors wish to pull meritorious
workers into their team because a good team will act on the performance of
their own office and reflect well on them. Meritocracy is here driven by the
179
self-interest of senior officers when making promotion decisions—the

bureaucracy is structured to ensure officers have incentives to promote
developmental success. This simple finding required a vast amount of data,
including the tax collection performance of PAS new recruits on their first
job between 1983 and 2013, the universe of personnel records of PAS civil
servants that have ever worked in Punjab from 1953 to 2013, the recruitment
exam ranking of civil servants by the Federal Public Services Commission
from 1973 to 2012 and the historical tax collection records from the Board
of Revenue from 1983 to 2013. The study also required a deep understanding
of the process of promotion and career progression in the PAS (the author
herself is a civil servant in Pakistan). The deceptively simple result was a
hugely impressive research effort. The results showed that a civil servant in
the top 10 per cent of their cohort in the recruitment exam collected 3 per
cent more taxes and was 10 per cent more likely to be awarded an ‘outstanding’
by their immediate bosses (2019: 3). As former senior colleagues themselves
are promoted and have greater influence in the promotion decisions of junior
colleagues, the top exam performers and those who are successful in collecting
tax revenue in their first jobs are significantly more likely to be promoted
than other civil servants. This is heartening but the result only applies to a
tiny cohort of elite civil servants—about 20 each year. The reality of Pakistan’s
government below this meritocratic apex is very different. Even a meritocratic
bureaucracy can be sidelined or overruled by politicians.
Over the 1990s in Pakistan, political pressures weakened the institutional
capacity to efficiently manage public investment. Politics in command meant
that formal approval procedures were often bypassed for work, roads and
energy expenditures. As a result, the remaining smaller amount of spending on
development projects proved considerably less productive. By the end of 1996,
this had led to Rs 700 billion worth of questionable projects being started, when
only Rs 85–90 billion per year was available to complete them (McCartney
2011b: 183). This political pressure emanated from a civil society that was ever
more influential over the state. The relationship between the state and society
in Pakistan is one of patronage relations between politicians and supporters
or dependents, ‘people gain access to patronage by using their position within
a kinship network to mobilise support for a politician who then repays them
in various ways in office, or by using kinship links to some policeman or
official to obtain favours for relatives or allies’ and the process can be likened
to state fiscal resources being ‘nibbled by a plague of mice’ (Lieven 2011: 213).
180
The state fails to provide public services such as water, education and power
because it is both too weak to raise tax revenue and to control corruption
among state officials. Corruption is not just about individual gain but is also
for patronage whereby state resources are recycled by politicians to win, retain
and reward supporters and kinship groups.
We can also look at the capacity of the Pakistani state both comparatively
and over time. There are a number of indices produced by various organisations
that measure aspects of state capacity and have been producing their indices
over time.
The Doing Business report produced by the World Bank collects both
quantitative and qualitative data on business regulations and the protection of
property rights relevant for doing business and uses the data to rank countries.
Those regulations include starting a business, dealing with construction
permits, getting electricity, registering property, getting credit, protecting
minority investors, paying taxes, trading across borders, enforcing contracts
and resolving insolvency. Table 7.3 shows Pakistan’s doing business ranking
Table 7.3 Pakistan: World Bank Doing Business ranking
Criteria 2009 Report (181 2014 Report (189 2019 Report (190
countries) countries) countries)
Overall ranking 77 110 136
Starting a business 77 105 130
Dealing with construction 93 109 166
permits
Getting electricity n/a 175 167
Employing workers 136 n/a n/a
Registering property 97 125 161
Getting credit 59 73 112
Protecting investors 24 34 26*
Paying taxes 124 166 173
Trading across borders 71 91 142
Enforcing contracts 154 158 156
Closing a business 53 71 53**
Source: World Bank (2009b, 2014, 2019a).
Notes: * In 2019 this changed from ‘protecting investors’ to ‘protecting minority investors’.
** In 2019 this changed from ‘closing a business’ to ‘resolving insolvency’.
181
across various of these measures in 2009, 2014 and 2019. As is typical of these
reports, the number of countries surveyed increases over time, so the ranking is
not completely comparable.The data shows that Pakistan’s relative performance
has been declining over time. Pakistan’s overall ranking dropped from 77th
(out of 181) in 2009 to 136th (out of 190) in 2019. By comparison, the Modi
government in India since 2014 has made a clear and well-publicised effort to
improve India’s doing business ranking. The national Indian media reported
with excitement that India’s ranking had jumped 23 places to 77 in the year
prior to the publication of the 2019 report (The Hindu 2018). Pakistan rates
particularly poorly in dealing with construction permits, registering property
and paying taxes. Two other poorly performing areas, getting electricity and
trading across borders, may be expected to be influenced by the CPEC and the
FTA with China. Neither has shown any significant improvement in the years
since the CPEC or the FTA were signed.
The Global Competitiveness Report is compiled and published by the
World Economic Forum. It endeavours to capture in a composite index the
factors that determine a country’s level of productivity. By 2018 the index was
based around 98 individual measures grouped into 12 ‘pillars’: institutions;
infrastructure; technological readiness; macroeconomic context; health;
education and skills; product market; labour market; financial system; market
size; business dynamism; and innovation. The Global Competitiveness Reports
compile indices, ranging from 1 to 7 (7 being the best) to measure various
aspects of governance. Table 7.4 compares the reports from 2006–7, 2009–10,
2014–15, 2015–16 and 2017–18. The data reveal a U-shaped progress in the
Table 7.4 Pakistan: measures of competitiveness
Measure of governance 2006–7 2009–10 2014–15 2015–16 2017–18

Quality of institutions 3.5 3.3 3.2 3.3 3.5
Judicial independence 3.3 3.1 3.8 3.6 3.6
Favouritism shown 3.1 2.8 2.6 3.0 3.2
in decisions of
government officials
Wastefulness of 3.5 3.3 2.6 2.6 3.4
government spending
Reliability of police 3.1 3.0 3.1 2.8 3.3
Source: World Economic Forum (2006, 2009, 2014, 2015, 2017).
182
overall quality of governance in Pakistan. The measures of favouritism shown

in government decisions and wastefulness of government spending show a
worsening from 2006–7 to 2014–15 and an improvement since, with no overall
improvement over the entire time period. Perhaps as expected, the Lawyers’
Movement in 2007 and restoration of democracy in 2008 have helped lead to
a slow and steady improvement in judicial independence. Perhaps related to
this, there also seems to be a very steady improvement, from very low levels, in
the reliability of the police.
Transparency International first launched their corruption perceptions
index (CPI) in 1995 and it has had a huge influence in publicising the
issue ever since; countries often debate their ranking on the front page of
national newspapers. By 2018 some 180 countries were covered in the survey.
The survey is conducted among experts and businesspeople and asks about
their perceptions of corruption in the public sector. As such, the index is
limited by missing the everyday corruption experienced by ordinary residents
or corruption in the private sector. Table 7.5 shows that Pakistan by 2018
ranked about two-thirds down the list, worse than many comparator countries
in Asia but similar to rapidly growing Vietnam and less than the human
development success story of Bangladesh.
Table 7.6 shows that Pakistan has some grounds for optimism; the position
two-thirds down the national league table has been generated by a steady if
slow improvement in Pakistan’s corruption perceptions score over time and a
corresponding rise in its ranking, from near the bottom in 2003 to its current
level.
Table 7.5 Corruption perceptions rank and score, 2018
Country CPI score, 2018 CPI rank, 2018 (/180)

Pakistan 33 117
India 41 78
Indonesia 38 89
Sri Lanka 38 89
Thailand 36 99
Vietnam 33 117
Bangladesh 26 149
Source: Transparency International (2019).
183
Table 7.6 Pakistan corruption perceptions over time
Year CPI score CPI rank

2018 33 117/180
2017 32 117/180
2016 32 116/176
2015 30 117/168
2014 29 126/175
2013 28 127/177
2012 27 139/176
2011 25 134/183
2010 23 143/178
2009 24 139/180
2008 25 134/180
2007 24 138/180
2006 22 142/163
2005 21 144/159
2004 21 144/159
2003 25 92/133
Source: Transparency International (2019).
The Country Policy and Institutional Assessment (CPIA) assesses the

quality of a country’s present policy and institutional framework, with ‘quality’
seeking to provide a measure of how conducive that framework is to fostering
poverty reduction, sustainable growth and the effective use of development
assistance. The CPIA rates countries against a set of 16 criteria grouped in
four clusters: economic management; structural policies; policies for social
inclusion and equity; and public sector management and institutions. The
CPIA ratings are used in the allocation of development funding assistance
and several other corporate activities. The criteria accounted for in the CPIA
have evolved over time, reflecting lessons learned by the study and practice of
development. There was a major overhaul and review of the CPIA in 2004
and another round of changes in 2011which limits the comparability of the
index over time. Table 7.7 shows that measures of policy and institutions in
Pakistan (1 = low, 6 = high) tended to decline between 2005 and 2011, before
showing either a marginal improvement or stagnation at the lower level.
184
Table 7.7 Pakistan: Country Policy and Institutional Assessment (CPIA)
2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016
Business regulatory 4 4 4 4 4 3.5 3 3 3 3 3 3.5
environment rating
Debt policy rating 4.5 4.5 4.5 4 3.5 3.5 3.5 3.5 3.5 3.5 3.5 4
Macroeconomic management 4.5 4 3.5 2.5 3 2.5 2.5 2.5 2.5 3 3.5 3.5
Quality of budgetary and 3.5 3.5 3.5 3.5 3.5 3.5 3.5 3.5 3.5 3.5 3.5 3.5
financial management
Fiscal policy sustainability 3.5 3.5 3.5 2.5 3 2.5 2.5 2.5 2.5 3 3 3
Quality of public 3.5 3.5 3.5 3.5 3.5 3.5 3.5 3.5 3.5 3.5 3.5 3
administration
Property rights and rule-based 3 3 3 2.5 2.5 3 2.5 2.5 2.5 2.5 2.5 2.5
governance
Efficiency of revenue 3.5 3.5 3.5 3 3 3 3 3 3 3 3 3.5
mobilisation
Transparency, accountability, 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 3.0 3.0
and corruption in the public
sector
The World Justice Project rule of law index measures how the rule of
law is experienced and perceived by the general public in 126 countries
and jurisdictions, based on more than 120,000 household and 3,800 expert
surveys. The index collects data on countries’ rule of law performance across
eight factors: constraints on government powers; absence of corruption; open
government; fundamental rights; order and security; regulatory enforcement;
civil justice; and criminal justice. Table 7.8 shows that Pakistan performs
terribly on measures of order and security, which attempts to measure the
experience of crime, civil conflict and presence of violent redress. Pakistan
remains resolutely near the bottom of the world rankings despite the large
increase in the number of countries included in the index (from 35 in 2010
to 126 in 2017). This was caused by a decline or stagnation in Pakistan’s
score.
Pakistan also does badly in terms of the regulatory enforcement index.
This index tries to measure whether the regulatory framework is properly
enforced, if there is improper influence and delay in the legal process, respect
for due process, or expropriation without compensation. Table 7.9 shows that
Pakistan’s score has fluctuated somewhat around a stagnant trend and again,
its global ranking remains resolutely near the bottom.
Table 7.10 shows the results for constraints on government power. This
seeks to measure constraints on the judiciary and the executive, independent
auditing, sanctions for official misconduct, non-government checks on
governmental power and lawful transitions of power. There are more grounds
for optimism here. Both Pakistan’s score and its relative global ranking have
improved between 2010 and 2017. This is not surprising; these years saw
Table 7.8 World Justice rule of law index: Pakistan, order and security
Year Score Global ranking

2010 0.53 24/35
2011 0.33 66/66
2012–13 0.29 97/97
2014 0.3 99/99
2015 0.3 101/102
2016 0.29 113/113
2017 0.36 124/126
Source: World Justice Project (2019).
186
Table 7.9 World Justice rule of law index: Pakistan, regulatory enforcement

2010 0.33 33/35
2011 0.41 59/66
2012–13 0.36 88/97
2014 0.35 95/99
2015 0.36 99/102
2016 0.34 109/113
2017 0.38 116/126
Table 7.10 World Justice rule of law index: Pakistan, constraints on government power

2010 0.26 35/35
2011 0.37 60/66
2012–13 0.46 69/97
2014 0.46 73/99
2015 0.49 67/102
2016 0.52 72/113
2017 0.52 74/126
the establishment of democratic government, a general election and smooth

transition of the ruling party, and the 18th Amendment, which devolved
more power to the provinces.
The evidence from these various efforts to measure Pakistan’s governance
shows that the developmental capacity of the state is extremely weak and
has even been declining over the last decade. Recent efforts to implement
developmental-type interventions offer further evidence that Pakistan lacks
the political underpinnings of a developmental state.
One such example is the Medium-Term Development Framework
(MTDF) of 2005 to 2010, which aimed to provide basic infrastructure to
promote sustained economic growth. The effort was evaluated by the Asian
Development Bank (2013). The evaluation found that efforts at inter-industry
187
cooperation and coordination were a failure and the work was eventually
implemented separately by line ministries; there was a significant lack of
government capacity to identify projects, from feasibility studies all the way
through to the bidding stage, the government was unable to ensure contract
and licensing enforcement, and the absence of a long-term debt market and
therefore long-term financing was an enduring constraint on infrastructure
financing. Not surprisingly, almost none of the targeted developmental
outcomes was achieved in practice. The literature on the developmental state
shows that bank credit, subsidies and other forms of intervention can be used
to drive economic change, such as technology acquisition and upgrading,
structural change and diversification. There is no evidence that the state of
Pakistan has proved capable of utilising bank lending for developmental
purposes. The evidence shows that Pakistan allocates credit to the politically
well-connected, not to actual or potentially dynamic industrial firms. Khwaja
and Mian (2005) use a novel dataset that combines detailed loan-level
information for every corporate loan made in Pakistan from 1996 to 2002
and data on electoral outcomes for the two elections that overlap the loan data
period. They show that politically connected firms receive loans that are 45 per
cent larger while showing 50 per cent higher default rates. This preferential
treatment is driven entirely by loans from government banks, which supplied
64 per cent of domestic lending during the sample period. There was no such
political bias shown by private banks. The preferential treatment to politically
connected firms is not just due to government banks selecting firms with
worse default rates possibly motivated by social concerns or a desire to ‘rescue’
them. This is about politics and politically connected firms being able to
leverage that influence to get access to more bank credit. Firms with ‘stronger’
politicians on their boards, as measured by votes obtained, electoral success of
the politician or their political party, obtain even greater preferential access to
credit from government banks. Politicians from constituencies with greater
voter turnout receive lower preferential treatment, possibly due to checks
imposed by electoral participation and political accountability. Either winning
or being in the winning party increases preferential treatment, indicating
the exercise of political power. Politically powerful firms obtain credit from
government banks by exercising their political influence on bank employees.
This influence stems from the organisational design of government banks
which enables politicians to threaten bank officers with transfers or removals
or reward them with appointments and promotions. The top hierarchy of
188
government banks including the chairman, president and board members are
all appointed by the government. The board, in turn determines the credit
and personnel policies of the bank, appointments and removals of officers and
employees.
A recent example of a very specific developmental intervention (and
failure) was the effort by the Ministry of Textiles to promote competitiveness
and technology acquisition by textile firms in Pakistan. This was the Pakistan
Textiles Policy 2009–14. The policy made a good effort to understand the
many constraints faced by the textiles industry. These included the regulatory
burden on businesses, the lack of suitably trained workers and the poor state of
infrastructure. The proposed policy interventions were voluminous and read
more like (yet again) a long list of aspirations than any practical suggestions.
They included: develop state of the art infrastructure facilities, increase the
supply of efficient human resources, evolve a legislative framework that sets
standards for each stage of processing with a view to increasing productivity,
improving quality, ensuring optimum utilisation of resources, promote R&D
to achieve product diversification, technological advancement, increased
productivity throughout the value chain, and specifically in the quality
and diversity of fibres, and encourage exports by meeting the demands of
competition, technology and higher labour productivity. The programme was
estimated to cost $8 billion over five years, to be largely funded by the private
sector and supported by a government-sponsored ‘Textiles Investment Fund’.
The government promised to provide finance to support the acquisition of new
machinery, IT and technology, improve infrastructure and boost the supply of
skilled labour, and to help with marketing. Government assistance was also
promised to help attract FDI into the sector and to support mergers and joint
ventures with foreign firms. The rising fiscal deficit of the central government
and related IMF pressure led to the policy fading away. The private sector
failed to invest, as the industry was hit by macroeconomic instability, high
cotton prices, inflation and interest rates increasing to 35 per cent. Another
Textile Policy (2014–19) was launched soon after, again, with a familiar list of
goals: to double textile exports to $26 billion per annum, add $5 billion of new
investment in machinery and technology, create 3 million new jobs and boost
value addition. The policy changes promised were rapid refunds of any sales
taxes, duty drawback for garment exporters, easy access to low-interest finance,
vocational training, duty-free import of textile machinery and a dedicated
technology upgradation fund. The policy promoted some investment,
189
mainly in spinning but not in the targeted garment sector; there was little
new value addition and nearly all of the policy goals were missed (Pakistan
These recent failures have not deterred other efforts at implementing an
industrial policy in Pakistan. In March 2016, the Ministry for Industries and
Production released Pakistan’s Automotive Development Policy 2016–21.
This strategy increased import duty on sub-components from 5 to
10 per cent and reduced import duties on non-localised Completely Knocked
Down kits from 32.5 per cent to 30 per cent, and on localised kits from
50 to 45 per cent. These efforts were aimed at reducing the cost of imported
inputs and creating an incentive to conduct more assembly operations within
Pakistan. In January 2017 the government announced an export package to
support the textile industry worth Rs 180 billion. Policy measures included
the removal of sales tax on imported textile machinery, and the removal of
customs duty and sales tax on cotton imports (HKTDC 2018b).
THE CPEC, GOVERNANCE AND THE FUTURE

OF INDUSTRIAL POLICY IN PAKISTAN
External pressures from the US, World Bank, Asian Development Bank
(ADB) and IMF, especially since the 1990s, have not made much impact on
the quality of the institutions of governance in Pakistan. The CPEC may be
different. The US, IMF and others have an established record of opposition
to industrial policy in preference to promoting the role of the free market.
China by contrast has a long history of successfully utilising industrial policy
to promote output and export growth, structural change and technological
upgrading. China has also promised that its relationship with Pakistan is
not a two–three-year IMF style intervention but one based on a long-term
partnership. Can governance in Pakistan improve by learning from the
Chinese model? Some authors agree!
Husain (2017) argues that the CPEC provides a chance to improve
governance in Pakistan. He argues that the ministries, provincial departments
or executing agencies have to deliver the CPEC projects within the time-
bound, resource-specified plan agreed with the Chinese government. So, he
argues, governance must improve. There is no explanation though of how this
will happen. There is no acknowledgement that every failed IMF agreement
was also a time-bound, resource-specified plan agreed with the IMF.
190
There is no analysis rooted in a political economy understanding of Pakistan’s

state; rather, Husain provides a wish list of improvements and reforms.
He lists variously that officials responsible for planning, coordination,
regulation and execution of the CPEC projects should be: selected on merit,
technical and/or managerial competence and integrity rather than on loyalty
and connections; assured security of tenure to avoid whimsical transfers
while the project is being implemented; provided necessary resources and
autonomy to operate without too much interference, monitored regularly and
performance evaluated against pre-agreed indicators; given full support
and protection against frivolous accusations and character assassinations; and
held accountable for results and outcomes.
Recall that a development bureaucracy is characterised by promotion
on merit, competitive remuneration, long tenure in office, clear sanctions
for corruption and leadership through a pilot agency standing outside and
astride individual ministries (Leftwich 2000; Doner, Ritchie and Slater 2005).
Such pilot agencies have had real power, authority, technical competence and
insulation in shaping development policy. Examples include the Economic
Development Board in Singapore and the Economic Planning Board in
Korea, which were both established in 1961 (Leftwich 1995). The central
nodal agency is typically dominant over other institutions such as the Ministry
of Finance and the Central Bank and also the stock market and domestic/
foreign banks (Levi-Faur 1998). The agency coordinates between other centres
of government to drive through a developmental vision. It is good then to
learn that something similar seems to be in the offing for the CPEC. A joint
cooperation committee has been established co-chaired by Pakistan’s minister
of Planning Development and Reforms and the Chinese vice-chairman of the
National Development and Reform Commission. Under this umbrella, five
working groups have been established for planning, transport infrastructure,
energy, Gwadar port and industrial parks/economic zones (Ahmad 2017).
The optimism about this (latest) effort ignores the history of recent past
failures in similar efforts. To implement an expanded public investment
programme between 2005 and 2010, a Programme Co-ordination Unit
(PCU) was established in the Ministry of Finance. The PCU was supposed
to closely coordinate with the Infrastructure Project Development Facility
(IPDF), Private Power and Infrastructure Board (PPIB), National Highway
Authority (NHA) and others. The PCU was planned to comprise high-level
officials from the Ministry of Finance, the Planning Commission and other
191
state bodies relevant to policy reforms. The expected inter-agency cooperation

did not materialise. Eventually, most of the work was implemented separately
by the line ministries, the IPDF and the PPIB, with outside support from
the ADB. The projects experienced delays in implementation. There was a
lack of capacity within the government agencies concerned, especially in the
provinces, for identifying projects and bringing these projects to the bidding
stages as well as for defraying the initial costs of project preparation such as
conducting feasibility studies (Asian Development Bank 2013).
Early analyses of the new CPEC coordination committee are negative,
The existing CPEC Secretariat in the Planning Commission of Pakistan and

another similar set up in the Ministry of Finance are inadequate, disjointed
and suffer from inefficiencies of coordination. Similarly, the Ministry of Inter-
Provincial Coordination, Federal Interior Ministry and Home Departments
also lack mechanisms of effective coordination, monitoring and delivery of
services through the CPEC projects. (Shafqat and Shahid 2018: 59)
In response to their pessimism, the critical authors fall back upon a well-tested
theme, calling for ‘better reform’. They call for the creation of an ‘umbrella
authority to integrate and coordinate all the multidimensional facets of the
project’ and suggest this will ‘help facilitate the process of project design and
implementation with the mandate of coordination and regulation’ (Shafqat
and Shahid 2018).
There is a large body of evidence, not referenced anywhere in the CPEC
literature, that seeks to explore the implications on governance of a large
increase in foreign aid. Collier (2007) finds that foreign aid helps create a
rentier state, where rulers derive revenues from aid and so have little incentive to
provide accountability and efficient governance in exchange for the legitimacy
to tax the domestic populace. Isaksson and Kotsadam (2018) examine the
impact of Chinese development projects on Africa between 2000 and 2012.
They note that the lack of detailed project-level financial information about
its foreign aid makes any evaluation of Chinese aid notoriously difficult.
Their survey examines perceptions of corruption across 98,449 respondents in
29 African countries using the Afrobarameter. They compare the experience
of those located close to a site at the time of the interview with those near
a site of future implementation. They found consistently more perception
of corruption around Chinese project sites. This impact did not occur with
World Bank projects. This effect was not simply due to higher economic
192
activity stimulated by Chinese investment. The use of satellite data on light

intensity shows that Chinese aid (unlike World Bank aid) did not seem to
stimulate local economic activity. A different finding comes from Presbitero
(2016), who studies the difficulty of absorbing sudden, sharp increase in public
investment in developing countries that initially were lacking absorptive
capacity, skills, institutions and management. He uses a World Bank data-
base of 6,750 projects from 1970 to 2007 across more than 100 countries.
He asks whether investment project outcomes worsen when public investment
accelerates compared to a historical pattern (as the CPEC is projected to do
for Pakistan). He finds that the impact is small and depends on the quality of
policies and institutions (measured with the CPIA rating of the World Bank)
and human capital. Pakistan, as we have seen earlier, rates poorly on both
these measures, and so would be at risk, though the impact, argues Presbitero
(2016), is only small. He argues that the problem is associated with donor or
investor fragmentation. This problem will not be relevant in the case of the
CPEC, which obviously has a monopoly donor, China. Media reports have
started exposing corruption in CPEC projects. It is alleged that the CPEC
coal plants Huaneng Shandong Ruyi (Pak) Energy (HSR) and the Port Qasim
Electric Power Company Limited (PQEPCL) inflated their start-up costs
to gain millions of extra dollars. The bulk of the power-sector corruption,
however, occurred outside of the CPEC projects (Haqqani 2020).
INFRASTRUCTURE AND THE POLITICS OF

NAIVETY AND REALITY
Pakistan’s Express Tribune (2019c) wrote of efforts to ‘forge a consensus’ among
stakeholders with an interest in the CPEC. They wrote of a series of meetings
with leaders of political parties including the Pakhtunkhwa Milli Awami Party
(PKMAP), Pakistan Tehreek-e-Insaf (PTI), Qaumi Watan Party-Sherpao,
Jamiat-e-Ulema Islam-Fazal ( JUI-F) and Awami National Party (ANP).
The idea that a large development project such as the CPEC can ever
achieve a consensus among different stakeholders acting in the national
interest is a dangerous and naive fallacy. This can promote the belief that any
person or group that remains opposed is acting out of either self-serving special
interest or treacherous and anti-national motivations. This fallacy ignores the
real facts that economic growth along with projects that stimulate growth
193
such as infrastructure is a profoundly political process that generates winners,

losers and political conflict and cannot proceed on the basis of consensus.
Waiting for politicians, voters, bureaucrats and businesses to somehow come
to a consensus will lead to permanent delays. History shows us that losers
have instead been included through compensation or else excluded through
various forms of repression. Building infrastructure and industrial policy more
generally is an intensely political process.
The presumption that China will somehow be immune from domestic
Pakistani politics and so be able to take decisions over infrastructure and
other investments in the national interest is also not correct. In Sri Lanka,
for example, political favouritism in Chinese lending was clearly evident.
A 35,000-seat Rajapaksa International Cricket Stadium (with a local
population of 11,000 inhabitants), an international airport and a new deep-
water port were built at Hambantota, which just happened to be the home
village of the then president, Mahinda Rajapaksa. Pre-construction feasibility
studies all pointed to the financial non-viability of the projects (New York
Times 2018).
The exemplar of successful industrial policy to link infrastructure
investment with domestic industrialisation in nineteenth-century Germany
was only made possible by profound domestic political changes. The lack of
interest in industrialisation in Germany before the 1840s can be explained by
the fact that before 1848 Prussia was an absolute monarchy whose political
affairs were controlled by a coalition of the king, a landed aristocracy and
an aristocratic bureaucracy. This ruling alliance preferred to promote their
own interests in agriculture and land than promote the dangers of urban-
industrialisation. The fear of working-class radicalism after the European
revolutions of 1848 pushed the aristocratic class into a political alliance with
the middle classes. Only once in government were the middle class and
business class able to reorient government policy towards a pro-industry
stance. The subsidies and investment guarantees, paid for by a new income tax
(1851) and expanded government debt, generated the railway investment of
the 1850s and 1860s (Tilly 1967).
There is fascinating and more contemporary evidence regarding the politics
of infrastructure. Hodler and Raschky (2014) use information about the
birthplace of political leaders and satellite data on night-time light intensity.
They ask whether regions have more intense night-time light when they are
the birth region of the current political leader. Other studies have shown
194
that light intensity does have a positive relationship with GDP while data
on public spending or governance indicators are not usually available at the
sub-national level. They use a panel dataset with almost 40,000 regions across
126 countries from 1992 to 2009. They find that light intensity is higher in
the birth region of the current political leader. A leader who implements good
policies generates economic development (more night-time light intensity) and
may win national political office (an endogenous impact). The effect persists
even when corrected for the fact that the leader may emerge from a better
performing region. The effect does not outlast the time in office of the political
leader, indicating this political allocation of development resources does not
generate sustained economic impacts. Stronger political institutions and more
education both reduce the impact of regional favouritism. Higher aid inflows
are associated with more regional favouritism but not in countries with strong
political institutions. These findings give good reason to be pessimistic for the
case of Pakistan and mean that political favouritism will have more freedom
to operate. For the specific case of Kenya, Burgess et al. (2015) find strong
evidence of ethnic favouritism in road-building between 1963 and 2011.
The 41 districts that share the ethnicity of the various presidents in power
received five times the length of paved roads built and double the expenditure
on roads. The effect disappeared during periods of democracy. This shows that
even imperfect democratic institutions like those in Kenya during the 1960s,
1990s and 2000s imposed constraints on the executive, as leaders were forced
to share public goods across the wider population. Democracy was associated
with an increase in political choice and participation and increasingly vocal
civil society groups, and also less repression of popular expression through
a reduction in press censorship and a more active role of parliament and the
judiciary in holding leaders to account. As leaders came under more scrutiny,
they reduced their ethnic biases. The politics of infrastructure also works on
the provision of public services. Franck and Rainer (2012) find that patterns
of primary education provision and infant mortality of ethnic groups across
18 African countries are impacted by the ethnicity of the countries’ leaders.
The effects of ethnic favouritism are an important determinant of education
and infant mortality in SSA over 50 years. The impact is large relative to the
trend improvements in both variables. In countries with a single dominant
religion, the impact of ethnic favouritism is reduced. The effect is equally
present in African democracies and autocracies. It is not evident what policies
are used to generate this impact.
195
The politics of infrastructure is also evident among contemporary

developed countries. Data from post-1945 Italy shows that when districts
elect more powerful politicians affiliated with governing parties, these districts
secure more infrastructure investments. When a district gives a lower vote
share to the governing parties, it receives more investment as an inducement
to change its political mind (Golden and Picci 2008). French data shows that
a disproportionate share of infrastructure construction is targeted towards
areas in which ‘swing voters’ are important. This is modelled by the absolute
value of the difference between the scores of the right-wing and left-wing
coalitions in recent elections and is a measure of how heated the most recent
electoral race was. The level of unemployment (a measure of economic need)
does not appear to have a significant impact on the allocation of infrastructure
spending. Roads and railways are not built to reduce traffic jams; they are built
to get politicians re-elected (Cadot, Roller and Stephan 2006). Using panel
data for 81 German cities for 1980, 1986 and 1988, Kemmerling and Stephan
(2002) find the opposite, that cities with a prevalence of marginal voters do
not spend more on public infrastructure, nor receive more investment grants
from higher-level governments. They also find that the larger the local
government majority, the more is spent on infrastructure; the effect is not
felt among unstable or small majorities. More is spent on infrastructure if
there is the same party or a political affiliation between the local city council
and the state government. A study of planned central government transport
infrastructure investment in Sweden between 2010 and 2025 shows that the
allocation is driven both by welfare and politics. The presence of politically
influential large firms has no impact on provision, but areas run by centre-
right political parties get less investment. There is no impact for municipalities
ruled by the party in alliance with the central government and there is some
evidence that more productive projects get higher allocations (Hammes and
Nilsson 2016). In Spain between 1987 and 1996, regional parties in alliance
with the central government and areas of strong leftist political influence
received more infrastructure investment (Castells and Sole-Olle 2005).
US data shows that term limits, citizen-initiative and budgeting procedures
were significant determinants of the flow of new public investments
during the 1980s. State electoral rules that strengthened the durability and
predictability of governance, such as unlimited, four-year gubernatorial terms,
biennial budgets and separate capital budgets, tended to reduce the growth of
the capital stock (Crain and Oakley 1995). Comparative studies demonstrate
196
the importance of politics but show that it is very difficult to isolate any
general results. Infrastructure goes more often to poor regions in France but
not Germany and Spain. Political strongholds of the central government do
not receive more infrastructure in France or Germany but do so in Italy and
Spain. Infrastructure is allocated to areas where the political race is close in
Germany but not in Spain. Stronger left parties are associated with more
infrastructure investment in France and Italy but not in Germany and Spain
(Kemmerling and Stephan 2008). It is difficult to draw any general lessons
for Pakistan from this disparate literature. We have no real idea of how the
CPEC-inspired infrastructure projects were planned in Pakistan. There is
much official rhetoric about equity. Every province was allocated an SEZ in
the interest of fairness and that the three transport routes of the CPEC have
been designed to transit through every region of the country. In particular,
the western route was placed to ‘ensure’ the integration of backward regions.
The rhetoric about the CPEC is all about politics and legitimately makes us
wonder about the economic rationale of the investments. The section titled
‘Industrial Policy, Pakistan and the CPEC’ showed how Pakistan lacks the
state capacity to drive efficient developmental interventions, which tend to be
instead driven by politics, corruption and patronage. There is evidence from
wider studies and the Pakistan context to have doubts about whether the
CPEC projects will be efficient, productive and generate an economic profit.
MOBILISATION AND CONFLICT IN PAKISTAN

The ability to mobilise tax revenue and direct it towards developmental
purposes (as noted previously, Pakistan has failed on both these counts) has
long been regarded as a key indicator of the developmental role of the state.
As Rostow, for example, noted of medieval Europe:
How much they [medieval states] could mobilise depended on the power of
the sovereign vis-à-vis landowners and the powerful towns and guilds of the
West, on administrative efficiency, and on the pools of income and wealth
generated by agriculture, commerce, banking, and industrial activity to which
the sovereign could, in one way or another, gain access. (Rostow 1975: 40)
There is a wide range of more recent evidence to support the central

importance of state-led mobilisation in driving economic growth. A sample of
32 countries reveals a significant and positive impact of public savings on GDP
197
growth (Krieckhaus 2002). The stagnation of Brazil and the rapid growth of
Singapore and South Korea are all linked to distinct stories about savings.
The slowdown of growth in Brazil after c. 1980 is directly linked to a weakening
of the central state, a rise in populist and politically motivated public spending
and hence lower public saving and public investment (Weyland 1988).
In Singapore, the state mobilised public savings by charging high prices on
the monopoly services provided by public utilities such as telecommunications
(Huff 1995: 745; Ermisch and Huff 1999). The level of consumption was also
squeezed in Singapore to mobilise resources from savings and tax revenue.
In 1995 private consumption reached only 40% per cent of GDP, a lower
share than the lowest ever reached by the Union of Soviet Socialist Republics
(USSR) (55 per cent). This squeeze allowed the total savings rate in Singapore
to increase from 6.7 per cent of GDP between 1960 and 1966 to over
40 per cent of GDP in the 1980s (Huff 1995: 737). In South Korea, the
government was successful in mobilising tax revenue, which increased from
7 per cent of the gross national product (GNP) in 1964 to 16 per cent in the
1970s. This was combined with a tight control of current expenditures and
hence allowed government savings to increase from 0 per cent of GDP between
1961 and 1965 to 5.5 per cent between 1966 and 1970 (Kohli 2004: 103).
The government also restrained private consumption through deliberate
policy, such as controls on consumer loans and high rates of indirect taxation.
Imports of luxury goods were banned or subject to high taxes (Chang 1993:
139). The share of private savings in GDP increased from 6.8 per cent between
1961 and 1965 to 18.0 per cent between 1976 and 1980 (Kohli 2004: 103).
The experience of contemporary Pakistan is strikingly different.
Figure 7.2 shows that savings has been steadily declining after 2004, from an
already low level. Investment has consistently exceeded savings, by a margin
which reached around 10 per cent of GDP in 2008 and 2015. This implies that
Pakistan is dependent on generating a surplus elsewhere to fund investment.
Unlike the cases of Singapore and South Korea, the state in Pakistan is
not mobilising a surplus. Figure 7.3 shows that since the mid-2000s, a long-
standing surplus of tax revenue over government consumption spending
turned into deficit. This means that the government has to borrow to fund all
of its investment and even some of its consumption spending. This widened
again (the World Bank data are not available) after 2013 and government
deficits increased sharply.
198
Figure 7.2 Savings and investment in Pakistan, 1988–2018

Figure 7.3 Government consumption and tax revenue in Pakistan, 1990–2015

199
The Pakistani state, as was documented in the section titled ‘Industrial

Policy, Pakistan and the CPEC’, is too weak and non-developmental to use
force as a means of enforcing developmental projects against the wishes of
well-mobilised losers. An example of failing to tax and enforce regulations can
be seen in the case of the bazaar traders of Pakistan.
The bazaar traders of Pakistan are engaged in both retail and wholesale
sale of goods in a designated marketplace, known as a bazaar. It is not
uncommon in contemporary Pakistan for such traders to have annual sales
exceeding $50,000. The Labour Force Survey of 2014 showed there were
some 285,000 traders of this scale across all-Pakistan and 165,000 of them
were in Punjab alone. Despite the slowdown in national economic growth
after the mid-2000s, the bazaar sector saw a growth of 6 per cent per annum
and by 2013 sales reached more than $150 billion across around 1.6 million
establishments. Around 94 per cent of these enterprises operate in the informal
sector and have long fought off attempts to regulate or tax them. The political
influence of bazaar traders is not just reactive or defensive. After winning
the 2013 general election, the re-elected chief minister of Punjab, Shahbaz
Sharif, held one of his first post-victory meetings with representatives of
various bazaar associations. During the course of this meeting, he declared
that ‘bazaar traders formed the backbone of this country’, and ‘resolution of
their problems is the responsibility of this and every subsequent government’
( Javed 2019). The Pakistan Muslim League (Nawaz), or PML(N), led by
the Sharif family is widely acknowledged to have deep ties with the bazaar
community. Bazaar traders are commonly picked as electoral candidates in
urban areas and all political parties rely on their networks to mobilise finance
and political support ( Javed 2019). In local government elections held in
2015, 184 out of the 268 candidates of the ruling PML(N), and 167 out of
the 256 candidates of the main opposition party PTI were traders or other
commercial entrepreneurs ( Javed 2019). The bazaar–mosque relationship
is also central to the political incorporation of marginalised groups, such as
informal labourers, new urban migrants and slum dwellers. Bazaar traders play
an integral role in administering local religious charities and offering alms and
assistance, especially around the holy month of Ramadan. These networks are
translated into an ability to mobilise votes. This political influence has clearly
been translated into influence over policymaking. Liberalisation in the 1990s
and 2000s removed restrictions on the import and trading of consumer goods.
In Punjab alone, a six-year consumption and building boom between
200
2001 and 2007 saw the number of retail and wholesale establishments grow by
nearly 40 per cent ( Javed 2019). Despite the evident accumulation of wealth,
bazaar groups used their close links to the state to evade taxes (Ahmed 2010;
Piracha and Moore 2015) and enclose government land in commercial spaces
(Ezdi 2009). This political influence is structured through traders’ associations.
In June 1998 the central government attempted to extend the general sales
tax (GST) into the retail sector. Using the associational networks functioning
at the marketplace, district and provincial levels, bazaar elites acted effectively
through the legislature, conducted more informal lobbying and carried out
general strikes to weaken and ultimately overturn the government’s resolve.
In 2000 and 2001, the military government of General Musharraf launched
a documentation drive aimed at assessing the actual net turnover value of the
retail sector net. Despite the regime’s insulated, authoritarian character, it was
also unsuccessful in the face of successive bazaar strikes in Punjab, including
one that lasted for 11 successive days. A 2010 order from the government
to close markets and shops in Karachi and Lahore by 8 p.m. was ignored by
bazaar traders. A 2015 government policy to enforce the closure of markets
and shops in Islamabad by 8 p.m. (to reduce electricity consumption during
nationwide shortages) was denounced by representatives of the traders, who
were supported by the Islamabad Chamber of Commerce and Industry
(ICCI). They prevented the policy from being implemented (Dawn 2015).
More recently, traders across Pakistan, including Karachi, Lahore, Peshawar,
Rawalpindi and most other major urban areas, shut shops to protest increases in
taxes (Samaa TV 2019; Express Tribune 2019b). Wider political mobilisations
have included those related to state service delivery failures in electricity, gas,
and law and order, religious causes such as defence of the blasphemy law, and
national political issues such as Indian ‘aggression’ in Kashmir. The bazaar’s
ability to mobilise effectively and utilise their ‘shutter-power’ is a significant
feature of contemporary politics in Pakistan ( Javed 2019).
While China is able to use its military to acquire land to further the
growth of industry and infrastructure, Pakistan lacks any equivalent ability.
An example of failed ‘developmental repression’ occurred in Okara in the early
2000s. The conflict took place between small tenant farmers and the Pakistani
military, who were the landlords. The tenants were in practice tenants of the
state, not private landlords. In June 2000 the farm managers announced that
the contract system was to be replaced by a new system with rent paid in
cash rather than a division of produce and also increased management control
201
over the cultivation practices on the farms. This was aimed at restoring rental
incomes and encouraging modernisation and marketisation of production.
Both the large-scale capitalist farmers, who used machinery and hired
labour, and small-scale farmers and labourers mobilised collectively to protest
the change. After failures in negotiation, in October 2000 a rally of about
1,000 people culminated in a four-hour sit-in at the lawns of the Deputy
Commissioner’s Office in Okara, an event that marked the beginning of the
tenant protest movement. Within months, the mobilisation spread across
other districts in Punjab. There were around 100,000 residents on Okara
military farms and the movement was able to mobilise 30,000–40,000 at any
one time (Akhtar 2006).
Around 2003, there was a sudden upsurge in economic growth, to over
7 per cent per annum for several years. This was related to a surge in investment
from 16–17 per cent of GDP in 2003 to 23–24 per cent of GDP in 2007.
The macroeconomic space for more investment was created by efforts to control
consumption. Figure 7.3 shows that consumption as a share of GDP stabilised or
declined in Pakistan around 2000. The 1999 coup and incoming Musharraf-led
military government had strengthened the central government at the expense
of civil society and allowed them to restrain consumption and boost productive
investment (McCartney 2015c). In India and even more so China, similar
economic booms were supported by a more determined control of consumption
which permitted investment to rise to 35 per cent or more of GDP.
The relative autonomy of the Pakistani state was brief and after 2002
society began reasserting itself. Musharraf was forced under growing political
pressures to boost populist consumption. In Pakistan, in the 2002 elections,
an Islamic alliance (the Muttahida Majlis-e-Amal, or MMA) came to power
in the North-West Frontier Province (NWFP) and won 62 national assembly
seats. The PPP emerged as the largest party with 26 per cent of the vote.
The pro-Musharraf Pakistan Muslim League (Quaid e Azam Group), or
PML(Q), won around 25 per cent of the vote. The vote marked an extreme
fragmentation of politics, with each province being won by a different political
grouping. The failure of the PML(Q) demonstrated that the Musharraf
government had failed to incorporate or win over the bulk of the civil society
organisations, such as labour, the middle classes and religious groups. After
the 2003 US invasion of Iraq, support for the government declined further.
Musharraf was forced to resign as army chief of staff in 2004. The pace of
reforms slowed and ‘on most key issues he backtracked under pressure from
202
his own right wing allies and the mullahs’ (Z. Hussain 2010: 183). By 2007,
opposition to Musharraf intensified with the emergence of the Lawyers’
Movement. The movement originated in an attempt by Chief Justice Iftikhar
Chaudhry in early 2007 to place limits on President Musharraf ’s power.
The chief justice was then dismissed, which led to a protest movement among
lawyers (Lieven 2011). The focus until c. 2002 of the Musharraf government
on reducing deficits and boosting investment dissolved into a populist effort
to buy off growing political opposition. Bank credit was increasingly diverted
to consumption rather than industrial investment. Public sector employees
received large wage increases. In the early 2000s, petroleum imports had
been liberalised and a pricing formula adopted that linked domestic prices to
international prices (Husain 2003). By the mid-2000s, consumers were being
protected from rising international oil prices through the generous expansion of
subsidies. Efforts at both incorporation and repression had failed. The result of
all this populist largesse was rising consumption, falling investment and a sharp
increase in the government fiscal deficit. Economic growth dropped back
to its long-term average of around 4–5 per cent. The newly elected democratic
governments under the PPP (2008) and PML(N) (2013) were for a time
again able to boost investment at the expense of consumption. The PML(N)
aimed to boost spending on infrastructure. There were 30 per cent increases in
public investment in 2014–15 and 2016–17. The share of public investment
increased from 3.2 per cent of GDP in 2013–14 to almost 5 per cent of GDP
by 2017–18. This could not be sustained. Figure 4.1 shows that the government
was unable to mobilise sufficient tax revenue and fiscal deficits worsened to
6.5 per cent of GDP by 2017–18 (Government of Pakistan 2019). Figure 7.4
shows how consumption in Pakistan once again resumed its upward trend in
the run-up to the 2018 general elections.
The story of the CPEC has some clear parallels with the story of the
Musharraf years—an inability to stick to a developmental vision without
expanding subsidies to ever-widening groups. After the announcement of
the CPEC in 2015, there was immediate opposition from regions outside
Sindh and Punjab, who perceived that they were being marginalised.
In response, the government extended the CPEC into a western route
that ran through the more backward areas of Pakistan and emphasised
the socio-economic dimensions of the project. The inclusion of the
Orange Line urban train project in Lahore was of doubtful viability but
was clearly targeted at the political heartland of the then ruling PML(N).
203
Figure 7.4 Household final consumption expenditure as a share of GDP

The apparent opposition of industrialist Abdul Razak Dawood, an advisor

to the incoming PTI government in 2018, was neutralised by the award of a
Rs 300 billion contract to his company, Descon, to construct the Mohmand
Dam in northern Pakistan (E. Hussain 2019). These are repeated examples of
a story highlighted by Lieven (2011). Here the ‘poverty and weakness of the
state’ in terms of the resources it is able to mobilise through taxation means that
‘there just isn’t enough patronage to go round’ in Pakistan’s hyper-politicised
patronage society. On being elected, the government hands out favours in the
form of ministerial posts, ‘tax breaks, corrupt contracts, state loans (which are
rarely repaid) and amnesties for tax evasion and embezzlement’. This ‘keeps
the state poor’ and, as time goes by, more of the political elites are disappointed
and unable to pass on favours to their followers and voters, which increases the
likelihood of not being re-elected (Lieven 2011: 205).
The PTI, led by the incoming prime minister, Imran Khan, won the
national election in 2018. The party showed no sign of strengthening its
organisation such that it could avoid relying on the politics of patronage.
The party has been built on the basis of a long mobilisation by a charismatic
national figure. The party was formed in 1996 and only won a single seat in
the 2002 general election with 0.8 per cent of the popular vote. The PTI came
204
into power by winning 16.9 million votes to form the national government and
also had a strong base of regional support by forming the state governments of
Punjab and KPK. Although the PTI claims a mass membership (of 10 million
members), the party does not have a strong organisation and relies instead on
the charismatic appeal of former Pakistan cricket captain Imran Khan to win
elections. In 2012–13, the PTI held internal party elections among members,
US style, to positions on its National Council. It was the first Pakistani party
to hold such a large internal election among its mass base. This potentially
would strengthen the legitimacy of the party and its office bearers. Most of
the PTI central leadership was elected. The PTI made efforts to incorporate
a wide range of civil society groups within its fold, and had separate wings for
students, women and the youth. Whilst in office, the PTI revealed its internal
confusion and regressed back to the Pakistani political norm. In June 2019, all
office bearers except the chairman (Imran Khan), the vice chairman and five
others ceased to hold office; all PTI-related organisations were dissolved and
new bodies were planned to be formed in line with the party’s newly approved
constitution. The new office bearers would be initially nominated. Elections
were promised for some time in 2021 under a federal election commission to
which office holders could not be members (Dawn 2019).
THE NEED FOR GOVERNMENT INTERVENTION

IN FINANCING INFRASTRUCTURE
Figure 7.2 showed how Pakistan has long been dependent on external flows of
financing to finance its investment and other spending needs and why external
financing for the CPEC is a necessity. There are also important reasons, related
to various market failures, why government-to-government financing is likely
for a big infrastructure project such as the CPEC.
In infrastructure, investment has long been characterised by close
government involvement. In the nineteenth century, governments often
borrowed themselves to invest in infrastructure, provided direct subsidies to
private investors or offered guaranteed rates of return for private investors.
The economic rationale for such interventions rests on the externalities
generated by infrastructure investment. Fogel (1960), for example, estimated
that the social returns of the Union Pacific Railways in the US at 30 per cent
were two and a half times higher than the private returns. The sources of these
externalities were many. In some cases, they were created through increased
205
value for the land adjoining the railway projects. The profits of private textile
mills were boosted by the cheaper and more predictable delivery of raw
materials to the factory and finished products to the market. Infrastructure
projects are often large relative to a developing country economy, take a long
time to construct (long gestation) and generate an uncertain economic return
in comparison to more predictable and familiar sectors such as traditional
farming, real estate and money-lending (Eichengreen 1995). A common form
of intervention historically was for the state to offer guaranteed rates of return
to private investors. In nineteenth-century India, if a railway company did
not attain a 5 per cent minimum rate of return, the government made up the
difference from general taxation revenues. In Canada during the same era, the
guaranteed return was 6 per cent (Eichengreen 1995: 84). When governments
lack a strong tax base, they may be able to use assets to help finance expensive
infrastructure investments. In nineteenth-century US, land grants were a
common form of government subsidy. Approximately 150 million acres of
land was granted to the western US railways between 1850 and 1870 as a
reward for building the railways. Land grants were also used from the 1870s
in Canada. Acquiring land close to the railway lines allowed railway promoters
to recover some of the externalities generated by their investments. Both the
US and Canada had large amounts of cheap and lightly settled land that was
formally vested in government hands. Land grants, unlike financial subsidies,
did not require prior government taxation or create government debt
(Eichengreen 1995). This close government involvement did, of course, create
opportunities for corruption and other distortions. The guaranteed rates of
return on investment gave railway promoters more incentives to invest but
weakened incentives for investors to hold the management accountable.
An inefficient construction effort, by inflating costs, would be compensated for
by higher, guaranteed, revenues. There is some evidence that the construction
of the Canada Great Western Railway was burdened by extra lines being added
at inflated running costs. The construction company was closely connected to
the top levels of Canadian politics. In the extreme, this could even encourage
the construction of railways lines where there was no prospect of freight or
passengers, simply to recoup the guaranteed return (Eichengreen 1995).
The implication for Pakistan is that foreign capital will be required to
fund an extensive investment in infrastructure (hence China). Guarantees on
foreign investment are common historically and are likely to be useful in indu-
cing long-term foreign commitments to infrastructure (think the 17 per cent
206
guaranteed returns on energy investments), but Pakistan lacks international

credibility in its ability to commit to such long-term contracts and governments
with severe budgetary constraints such as contemporary Pakistan are likely to
resort to land grants to help pay for infrastructure (think the 40-year lease on
Gwadar). What many have interpreted as a contemporary manifestation of
predatory investment by China actually reflects deep historical norms.
MANAGING THE CPEC FINANCING IN PAKISTAN

Since 1951, domestic investment has systematically exceeded domestic savings
in Pakistan and investment has been substantially funded by foreign capital
inflows. In the 1970s, this structural imbalance was met through remittance
transfers from Pakistanis working overseas, mainly in the Gulf region, in
the 1980s from foreign aid linked to Pakistan’s alliance with the US against
the intervention of the Soviet Union in Afghanistan, and in the 1990s via the
accumulation of expensive foreign debt. In the 2000s, this deficit was initially
met through sharply rising levels of FDI and remittances and to a lesser extent
debt forgiveness linked to Pakistan’s support for the US ‘War on Terror’ in
Afghanistan and then, after 2008, increasingly foreign borrowing (McCartney
2015c). Together Figures 7.2 and 7.3 show that Pakistan continues to be
structurally dependent on external resources to fund domestic investment.
Many have placed the ebb and flow of capital into Pakistan as the key
driver of both growth rates and external political influences. The reliance on
US capital implied that ‘US priorities determined Pakistan’s domestic and
foreign policies from 1951 onward’ (Ali 2008: 251). The most frequently
discussed example in the literature is the ‘Decade of Development’ (1958–68)
under Ayub Khan. In response to Ayub Khan’s pro-US foreign policy stance
during the Cold War, many argue that a surge of capital inflows promoted the
investment-led boom until 1965. The war between India and Pakistan was
opposed by the US and the resulting decline in capital inflows led to economic
slowdown in the second half of the 1960s (Amjad 1983; Griffin 1965). This
long-standing narrative continues into discussions about the CPEC, where
sceptics claim that Pakistan in the future will be increasingly locked in to serve
the geopolitical interests of China.
The claim that China is responding to Pakistan’s own needs seems belied
by the relatively limited input Pakistan had in the planning for the CPEC.
The detailed initial planning for the CPEC came entirely from China.
207
The long-term plan was initiated when the National Development and
Reform Commission (NDRC) of China asked the China Development Bank
(CDB) to compile a detailed plan to guide the engagement with Pakistan.
The CDB worked with the NDRC, the Ministry of Transport, the National
Energy Administration and China Tourism Planning Institute to develop a
schedule of activity for 2015 to 2030. The report was first ‘transmitted to
the Government of Pakistan in 2015’ and the immediate input from Pakistan
was that it ‘gathered dust for a few months’ and ‘[u]nder prodding from the
Chinese government, a team from Pakistan met their Chinese counterparts
in Beijing on November 12th 2015’; the plan was finalised by December of
that same year (Dawn 2017). Compared to the extensive involvement of civil
society organisations (for example, the case of Ghana discussed in the section
titled ‘Chinese SEZs in Africa’), the initial CPEC planning was done behind
closed doors. The CPEC long-term plan was only revealed when it was leaked
to the media—‘Dawn has acquired exclusive access to the original document,
and for the first time its details are being publicly disclosed here’—where the
media source made it clear that this was not a Pakistani initiative, as ‘[t]he
plan lays out in detail what Chinese intentions and priorities are in Pakistan
for the next decade and a half, details that have not been discussed in public
thus far’ (Dawn 2017). Some have taken this argument further and argued
that the CPEC is designed for the benefit of the Chinese and their eventual
economic domination of Pakistan. It is argued that Pakistan being a fragile
and high-debt economy with weak exports, dependent on foreign assistance
and prone to periodic external payments crisis, would not be able to meet the
additional debt obligations and repatriation of profits created by the CPEC.
If it is not able to pay, China will have the opportunity to take over the Gwadar
port, land and assets in Pakistan, as they have done with the Hambantota port
in Sri Lanka (Husain 2017), and also influence Pakistan’s future domestic
and foreign policy. This echoes the debate recently re-popularised by Shashi
Tharoor, who argued that the railways built by Britain in nineteenth-century
India were a ‘colonial scam’ intended to benefit Britain at the expense of India
(Tharoor 2016).
A similar argument was raised on 3 August 2018 in a letter to US Secretary
of State Mike Pompeo and US Treasury Secretary Steven Munchkin signed
by a bipartisan group of 16 US Senators. The letter communicated concerns
about ‘predatory’ Chinese infrastructure financing and about the potential use
of IMF financing to repay those Chinese loans. There were concerns that the
208
$1.6 billion bailout of Sri Lanka in 2016 had gone to repay loans from China.
The letter stated that the goal for the BRI was the creation of an economic
world order ultimately dominated by China. At the same time, the former US
Secretary of Defence Jed Babbin, in a discussion of the proposed IMF bailout
package for Pakistan, argued that China was engaged in de facto colonisation
of Pakistan through the CPEC, which amounted to debt-trap-diplomacy.
There was, he argued, no rationale for IMF dollars to bail out CPEC loans
(Choudhury 2018; Washington Times 2018; NDTV 2018). There are frequent
lessons being drawn for Pakistan by various authors from China’s earlier
engagement with Sri Lanka (Mourdoukoutas 2018). The real objective for
the investment in Hambantota, argues Choudhury, for example, was to induce
Sri Lanka to fail to repay Chinese loans. As a result, Sri Lanka would lose
sovereignty over massive tracts of land around the port area, which then would
fall under the control of Chinese security and become a no-go area even for
Sri Lankan security forces (2017: 195). There are some media suggestions that
China’s acquisition of the port has led to substantial investment to upgrade
the port and which has increased the number of vessels using the port and
diversified the services offered by it (Ranaraja 2020). We should be somewhat
suspicious that despite the more than 60 countries China has engaged with
during the BRI, the evidence for the generalised intentions, motivations
and likely outcomes of the entire BRI project keep coming back to a single
example. The government of Pakistan was clear in its rebuttal and argued
these views were ‘one-sided, distorted, critiques of its economic relationship
with China’ and that China was actively supporting Pakistan’s development
at a time when foreign investment from other sources had dried up and the
economy was crippled by energy shortages (Washington Times 2018).
As was discussed in section titled ‘“Complementary Advantage”: The
Importance of China’, the Chinese government sees the BRI generally and
the CPEC more specifically as a strategic partnership revolving around
win–win cooperation. Pakistan lacks the foreign exchange or credit rating
necessary to finance the infrastructure which has become a critical constraint
on growth. The CPEC offers China the opportunity to receive construction
contracts, import agricultural products, boost exports and the opportunity to
relocate labour-intensive production. China argues that it builds on China’s
own experience in opening up in the late 1970s, when it too was short of
foreign exchange and desperately needed foreign technology, infrastructure and
opportunities for export, and also learns from China’s successful engagement
209
with Africa (Brautigam 2009: 46). The Chinese government proclaims that
the BRI initiative will be embedded in a system of responsible finance:
We should improve the system of risk response and crisis management, build
a regional financial risk early-warning system, and create an exchange and
cooperation mechanism of addressing cross-border risks and crisis. We should
increase cross-border exchange and cooperation between credit investigation
regulators, credit investigation institutions and credit rating institutions.
We should give full play to the role of the Silk Road and that of sovereign
wealth funds of countries along the Belt and Road, and encourage commercial
equity funds and private funds to participate in the construction of key projects
of the Initiative. (Government of China 2015a: 4).
Others have argued that the CPEC is something of a loss-leading showcase

project for China. As one unverified source stated, ‘Chinese officials themselves
recognise that 90% of the money ploughed into Pakistan, half that invested
in Myanmar and a full third that is expended in Central Asia will probably
be lost’ (Frankopan 2018: 120). This view sees the CPEC as being about
demonstrating the magnanimity of China; that it can invest in a difficult
location, generate economic benefits for the host and dispel any negative
thoughts about the ‘real’ intentions of China. Authors like Bruno Macaes see
the real intentions of China as binding together China, Southeast Asia, Central
Asia, the Middle East and Europe into a new Eurasian supercontinent from
which the US is marginalised and over which China is dominant (Macaes
2018a). The CPEC in this view can be interpreted as a loss-leader, losing
money on the investment, benefiting Pakistan and providing a good advert
for China’s attempts to engage sceptical countries in Europe to eventually join
the BRI.
In practice, the financial side of the debate is difficult to resolve. China does
not report cross-border lending in a systematic or transparent manner beyond
headline investment numbers, announced by Chinese officials and dutifully
reported by the media (Hurley, Morris and Portelance 2018). The emerging
difference with other international donors and lending is striking. Looking
at two project reports plucked randomly from the website of the ADB and
World Bank reveal this clearly. The ADB project is for the Karachi Bus Rapid
Transit Project, which aims to develop a sustainable urban transport system
in Karachi during 2016–2018. There is a wealth of planning and evaluation
documentation available. The projected cost of $220 million is broken down
210
into a timeline of expenditure. The project loan, interest charge and repayment
period from the ADB as well as the contribution from the Government of
Pakistan are outlined in detail. The project is described as being embedded in
the published ADB–Pakistan country partnership strategy for 2015–19.
The recruitment plans for project staff is also described (Asian Development
Bank 2016). The World Bank project is the Karachi Water and Sewerage
Services Improvement Project, aiming to improve access to safe water and
increase the operational and financial performance of the state-owned
sewerage utility in Karachi. The financing ($100 million) of the project
is described and broken down into expenditure subheadings and World
Bank, Government of Pakistan and other sources of financing. The project
gives both end and intermediate project performance goals and indicators.
The project is embedded in a reform roadmap agreed between the Government
of Sindh and the World Bank up to 2030 (World Bank 2019c).
Hurley, Morris and Portelance (2018) use publicly reported sources to
construct a BRI project lending pipeline for a subset of 23 (from 68 involved in
the BRI) countries to determine the vulnerability to debt. As do the Chinese,
they consider any Chinese infrastructure project in an identified BRI country
since the announcement of the initiative in 2012 to be a BRI project. They
focus on those infrastructure projects they believe are likely to be financed
through sovereign or sovereign-guaranteed concessional and commercial
loans or export credits. This includes loans from Chinese policy banks to
state-owned enterprises in BRI countries even if authorities claim there is
no sovereign guarantee. These projects may not be added to the national
public-debt figures. There are numerous examples of loans that are contracted
without explicit sovereign guarantees to state-owned or quasi-official entities
that are implicitly guaranteed by the national government.
They characterise a BRI pipeline project as any project whose financing
may not be captured by a country’s public debt figures at the end of 2016.
Using these pipeline projects, they aggregate the debt component of each
identified project. They also include projects that have yet to move beyond
the initial announcement but which seem to be based on firm political
commitments. There are significant problems with these estimates as we ‘do
not know the financing terms for most of the pipeline projects nor the terms
for many existing project loans’ (Hurley, Morris and Portelance 2018: 25).
The indicators they consider are the country’s overall public debt-to-GDP
ratio as well as the concentration of that debt with China as a creditor.
211
There are 33 countries who, according to three major credit rating agencies
(Standard and Poor, Moody, Fitch Ratings), are currently rated as being below
investment grade (Hurley, Morris and Portelance 2018). Of these, 10–15 could
suffer from debt-distress and 8 are of particular concern. Table 7.11 shows
the significant and rising levels of actual and forecast government debt for
Djibouti, Kyrgyz Republic, Laos, Maldives, Mongolia, Montenegro, Pakistan
and Tajikistan. The expected levels of debt in the eight focus countries are
well above the average for their peers. For all of these countries, China is the
dominant creditor (Hurley, Morris and Portelance 2018).
The CPEC is currently estimated at $62 billion with at least $33 billion of
this expected to be invested in energy projects. China will reportedly finance
roughly 80 per cent of that amount. Despite this ambition, there have already
been cancelled projects, including three major road projects at the end of 2017
(Hurley, Morris and Portelance 2018). The energy projects are financed by
foreign investment under contracts that involve sovereign guarantees giving
investors a 17 per cent rate of return in dollars on their equity investment.
From a commitment of $50 billion, 70 per cent, or $35 billion, would be
coming to Pakistan in the form of FDI financed by loans taken out by Chinese
companies mainly from Chinese banks. The import of equipment and services
from China for the projects would be shown under the current account while
Table 7.11 Central government gross debt (% of GDP)
Actual Forecast
2015 2016 2017 2018
World 80.6 83.6 83.1 82.8
MICs 44.5 47.4 48.6 49.8
Mongolia 62.1 87.6 85.3 89.0
Montenegro 76.8 78.0 79.7 80.9
Pakistan 65.7 70.0 69.1 67.6
LICs 36.1 40.4 41.9 41.6
Maldives 73.1 83.1 96.5 109.0
Djibouti 72.1 86.6 88.1 87.5
Laos 65.8 67.8 69.0 70.3
Kyrgyz Republic 65.0 62.1 64.2 65.5
Tajikistan 33.4 44.8 51.8 56.8
Source: Hurley, Morris and Portelance (2018: 13).
212
the corresponding financing item would be FDI brought in by the Chinese

under the capital and finance account. These borrowings would not impose
any liabilities on the Pakistani government (Ahmad 2017; Husain 2017).
The infrastructure projects (roads, railways, port expansion) are mostly financed
through concessional loans (reportedly at 2 per cent) from the Chinese
government and the associated loan repayment flows should be moderate
in size. Overall, the current financing for the CPEC can be summarised as
FDI 64 per cent, concessional loans 24 per cent, commercial loans 6 per cent,
and grants 1 per cent (Ahmad 2017: 56). The secretary of the Economic
Affairs Division in Pakistan, Noor Ahmed, estimated that Chinese-CPEC
loans represented only 10 per cent of Pakistan’s total foreign debt (of about
$106 billion). The vast bulk was with traditional lenders such as the IMF and
western governments (Modern Diplomacy 2020).
Assuming the extreme case that the entire equity would be financed by
Chinese companies (though this is not true in the case of Hubco and Engro
projects, where equity and loans are being shared by both Pakistani and Chinese
partner companies), then the 17 per cent guaranteed return on these projects
would entail annual payments of $2.4 billion from the current account.
The second component of the CPEC is the infrastructure to be financed
through government-to-government loans, which amounts to about $15
billion. These loans would be concessional and repaid over a 20–25-year
period. These calculations suggest that the annual repayments would not
exceed $3.5 billion annually. This amounts to about 7 per cent of the 2016
total foreign exchange earnings, and these calculations do not take into
account the incremental gains from GDP growth that will occur because of
investment in energy and infrastructure (Husain 2017; IMF 2017). This may
also be offset by savings on imports as the input mix for the energy sector will
shift from furnace oil to coal, gas and renewables. The losses to the national
income due to energy shortages were recently estimated to be around
$6 billion annually and exports dropped from $25 to $21 billion partly
because of outages (Husain 2017). CPEC projects may also help Pakistan
boost its exports. The net CPEC flows could be absorbed without much stress
on the balance of payments if exports grew at 10 or 14 per cent per annum
(Husain 2017). This rate of export growth has not been achieved in Pakistan’s
recent history and there is no indication that the CPEC has boosted export
growth (see Chapter 6).
213
Even were Pakistan to suffer the consequences of debt problems with

China, it would not necessarily result in a loss of sovereignty. Other options
include debt relief from China and Pakistan renegotiating the terms and
conditions of its lending. Pakistan has already requested six debt bailouts
from western creditors under the Paris Club, for example. China has provided
debt relief to other countries on an ad hoc, case-by-case basis. Sri Lanka was
unwilling/unable to service an $8 billion loan, some of which was lent at
6 per cent interest, which was used to finance the construction of the
Hambantota port. It is often argued that China agreed in July 2017 to a
debt-for-equity swap accompanied by a 99-year lease for managing the port.
Moramudali (2020) argues on the contrary that no debt was cancelled; those
debts are still being repaid and that instead the port was 70 per cent leased to
China in return for $1.12 billion, which was used to strengthen Sri Lanka’s
foreign exchange reserves. China has also demonstrated a willingness to
provide additional credit so that a borrower can avoid default. For example,
China agreed in 2017 to extend an RMB15 billion swap line for Mongolia
for three years in support of an IMF extended fund facility. There are some
signs that Chinese officials are moving towards greater policy coherence and
discipline when it comes to avoiding unsustainable debt. In November 2017,
the China Banking Regulatory Commission issued its first-ever regulation
for China’s policy banks emphasising greater risk controls for the overseas
activities of the CDB, the China Eximbank and the Agricultural Development
Bank of China (ADBC) (Hurley, Morris and Portelance 2018).
China has generally refrained from participating in multilateral approaches
to debt relief though it does participate in debt-relief discussions at the
international financial institutions and engages informally with IMF staff on
individual country cases. This contrasts with other major official creditors, all
of whom participate actively in multilateral institutions dealing with sovereign
defaults, in particular the Paris Club, which was formed in 2005. China gave
serious consideration to Paris Club membership during its G20 presidency
in 2016 but did not make a commitment to pursuing the membership.
The club works through mutual accountability based on ownership,
alignment, transparency, harmonisation and results. Given the Paris Club
members’ commitment to sharing data on their claims on a reciprocal basis
(partly to prevent project overlap), a decision by the Chinese government to
fully participate in the Paris Club membership would be a significant signal
of the government’s willingness to change its history of non-transparent
214
credit activities. This European transparency is often a smokescreen. In 2000,

for example, Europe promised to provide $15 billion for infrastructure and
this transparent promise had still not been fulfilled. The emphasis on pushing
local ownership by the Paris Club is seen by China as irrelevant, as it already
does not impose conditionality and so argues that programmes are already
locally owned (Brautigam 2009).
A second possible option would be for Pakistan to renegotiate the terms
and conditions of its financial engagement with China. Here, though, there is a
real danger of Pakistan perceiving China to be its only option and so reducing
its own bargaining strength. There is a widespread perception that ‘Pakistan
has put all eggs into the basket of CPEC’ (Choudhury 2017: 144) and ‘in
Pakistan, where the minister of commerce, Khurram Dastgir Khan, suggested
that for Pakistan “China is the only game in town”’ (Frankopan 2018: 174).
There are signs that the new prime minister has become more enamoured of
the CPEC after assuming office, which may have reduced Pakistan’s ability to
exercise agency. In earlier years, Imran Khan did not seem to be a fan of the
CPEC; he had criticised Pakistan as acting under pressure from China during
the siege of Islamabad at the Lal Masjid in 2007. Abdul Razak Dawood, the
cabinet minister responsible for commerce, textiles, industry and investment,
said that the previous government had done a bad job of negotiating the CPEC
and gave away a lot. Even on 10 September 2018, the Financial Times reported
that the CPEC was being reviewed with the aim of renegotiating trade
agreements that unfairly benefits Chinese companies. It does seem clear that
Imran Khan came under pressure from much of the Pakistan establishment
to embrace the CPEC (Choudhury 2018). By the middle of 2020, as reports
filtered through that the majority of the CPEC projects had been partially or
adversely affected by coronavirus, Prime Minister Imran Khan was vowing to
complete the CPEC ‘at any cost’ (The Print 2020).
Other countries have demonstrated significant agency in relations
with China. The DRC was able to leverage access to its substantial natural
resources for Chinese FDI that represented a significant improvement over
existing mining agreements with more than 60 other firms. In 2007 and 2008,
DRC received $13.5 billion in loans in return for security in $14 billion of
copper and cobalt reserves (compared to a total annual state budget of only
$1.3 billion).The contracts placed a limit of 20 per cent on Chinese employment,
$3 billion of new investment into the mines through a joint Sino-Congolese-
owned company and for China to allocate $8.5 billion to a variety of projects,
215
including a high-voltage power distribution network, highway and railway

extensions, 31 hospitals, 145 health clinics, 5,000 houses and 2 universities
(Kaplinsky and Morris 2009). Aged 93, Mahathir was re-elected as the prime
minister of Malaysia in 2018, 15 years after having retired. He proclaimed
his mandate was to extract Malaysia from suffocating debt. Even on a visit to
China, he proclaimed in the Great Hall in Beijing that Malaysia did not want
a new version of colonialism. He cancelled the $20 billion East Coast Rail
Link project and two natural gas pipelines worth $2.3 billion. This was a clear
demonstration of national agency (Choudhury 2018: 46). A comparison of
African states shows that they also have significant agency in bargaining with
China, particularly those with significant strategic resources and democratic
governance. Ghana and Angola, for example, have used their oil reserves
to provide leverage in negotiations with China. In 2004 Angola received
$2 billion for public sector projects and in 2007 an additional $2 billion for
about 100 projects. Despite this, Angola tightened local content rules to
boost local participation, reduced the interest rate it was paying and insisted
on an increased repayment period. Angola also made efforts to diversify
the sources of its borrowing. This has included a $1 billion loan from the
Canada Export Development Bank for infrastructure development in 2007
and a $1.5 billion loan from the Brazilian Development Bank to purchase
construction equipment from Brazil in 2009. Between 2009 and 2013, Angola
also received $1 billion in lending from the World Bank to promote economic
diversification. In 2008 the Afghan government replaced the Chinese state-
owned enterprise Sinopec with the US firm Kellogg Brown and Root for the
construction of an $8 billion refinery. In Ghana, civil society has strengthened
local voice in relations with China. China agreed to provide $622 million to
finance the Bui Dam, via loans and a buyer’s credit secured by future cocoa
exports. Ghana used outside (French and British) consultancy reports to
provide the environmental and social norms by which Chinese construction
was asked to abide. At the same time, Ghana cracked down on environmentally
destructive and illegal Chinese gold miners and 4,600 such miners were asked
to leave the country. NGOs have scrutinised deals and spearheaded NGO
efforts to ensure greater transparency of deal-making. The $3 billion Western
Corridor Gas Infrastructure Development Project, combining oil, roads and
a port project, was renegotiated after falling oil prices made it less viable
for Ghana. The presence of Chinese aid was used as leverage to improve the
terms on which Japanese aid was given. Civil society, including the media,
216
plays the role of watchdog over the external engagements of the Ghanaian
state. The effective separation of power between the executive and the
legislative in Ghana allows for parliamentary oversight (Chipaike and Bischoff
2018). In 2020 Pakistan asked China for an extension of the repayment period
for $30 billion of the CPEC loans owing to the negative implications of the
coronavirus crisis. Commentators thought this concession was likely to be
granted (Asian Review 2020).
It should also be remembered that infrastructure can increase the ability to
engage with other countries and help reduce Pakistan’s dependence on China
(other than standards lock-in—for example, if only Chinese-built trains can
run on Chinese-built railway systems) (P. Khanna 2019). For example,
Myanmar, like Pakistan, has a long relationship with China. In the 2000s,
China began upgrading roads and ports in Myanmar to ease Chinese access
to the Andaman Sea through an alternative route to the Straits of Malacca.
India was spurred by this Chinese intervention to develop the Sittwe port as
its commercial bridgehead and Japan made Myanmar a strategic investment
priority and is expanding the Thilwa port near Yangon. China’s efforts ‘have
thus awakened India, Japan, and others to invest in cross-regional integration
and inspired the Myanmar government to renegotiate terms on certain Chinese
projects to bring down their exorbitant costs and reduce Myanmar’s debt burden’
(P. Khanna 2019: 118). Even the Sri Lankan example of a well-cited failure is
mitigated by the fact that the Hambantota port and airport are now divided
between China and India, which reduces the cost of any Sri Lankan debt
dependency on China.
IN SUMMARY
This chapter started with economic theory and demonstrated the wide range
of market failures that create a theoretical rationale for the use of an industrial
policy. History has demonstrated the successful use of an industrial policy in
nineteenth-century Germany, and to an extent in Russia and in contemporary
Ethiopia, to ensure that big infrastructure generated benefits for the domestic
economy. Pakistan has all the characteristic features of an economy suffering
market failures that could justify the use of an industrial policy, including low
productivity, slow structural change and slow technological upgrading. These
failures have not been lost on scholars of the CPEC, who call for industrial-
policy style interventions by the Government of Pakistan. Such calls fail to
217
recognise that a theoretical case is not sufficient to justify real policy changes.
As demonstrated by numerous internationally comparable indices, Pakistan
shows no indication of having the necessary state capacity to successfully utilise
an industrial policy. Policymaking in Pakistan is dominated by the short-term
and lending to the politically well-connected, and recent attempts at utilising
an industrial policy in the textiles sector have been well-documented failures.
Recall from the discussion in Chapter 5 that big infrastructure is not a win–
win policy but will create both winners and losers. How the state manages
the potential political conflict resulting from these distributional impacts is
important. A political response may be to forget efficiency considerations and
try to spread the benefits of infrastructure, building uneconomic roads and
SEZs in remote locations, for example. The state may also seek to preserve the
efficiency of infrastructure and pay compensation to the losers or else use a
measure of coercion to ensure big infrastructure maximises economic growth.
The weak state and weak political parties of Pakistan give us no reason to
think that Pakistan will be able to successfully manage the distributional
implications of the CPEC. The boom–bust cycle under General Musharraf in
the 2000s offers a salient example of how Pakistan failed to turn a boom into
sustainable and rapid economic growth. The financing of the CPEC remains
a puzzle and while the data that is available shows concerns for Pakistan,
those concerns are not enough to suggest that China is engaged in predatory
lending in an attempt to push Pakistan into debt-led dependency. Pakistan
could do much more to negotiate better terms and conditions in lending and
economic links (such as the China–Pakistan FTA) by using its undoubted
geopolitical importance to China, to Saudi Arabia, to Iran, to the US and
others as leverage. Recall the discussion in Chapter 5 on how Ethiopia used its
role as a force for regional stability, its developmental vision and its efforts to
reduce poverty as leverages across multiple donors to maximise developmental
assistance. Pakistan can learn from this experience and should not succumb to
the trap of thinking that China is the only game in town.
218
8
Conclusion
The Way of the Dragon or the Way of the Falcon?
What created the Asian Tiger or Dragon economies in the post-war era?
Scholars have emphasised a range of factors, including those running from
culture, to geography, to the successful policies implemented by a developmental
state driven by the politics of catching up with the west. Littered among
those explanations are external factors. Some have argued that the fear of
communism (the proximity of South Korea to North Korea, of Hong Kong
and Taiwan to China, of Japan to the USSR) drove a national mission to
promote rapid economic growth to avoid the threat of being swallowed by
aggressive neighbours. Others have suggested that the politics of the Cold
War meant that the US facilitated the rise of those countries, through foreign
aid, easy market access and not being overly concerned with violations of
patents and copyrights. The US allowed these countries to free-ride on its
own economic dominance in order to promote powerful and capitalist Cold
War allies. We can draw on scholarship that emphasises the importance of
external agency in the creation of the miracle economies. It is reasonable then
to start here with the question: can Chinese efforts to build the New Silk
Road generate a new miracle in Pakistan, a ‘Falcon Economy’?
The story of the CPEC is primarily a Chinese-dominated narrative, the
most important aspect being China’s transformation since the late 1970s—
China’s rise from a poor agrarian economy in desperate need of foreign exchange
and advanced technology, to the need for raw materials for an industrialising
economy and, finally, to the need to restructure and create national champions
for a mature industrial economy. The generosity and size of the CPEC can be
less explained by careful diplomacy by Pakistan and more by Pakistan’s long-
standing friendship with China and the luck of its geographical proximity
to Western China, the Middle East and Afghanistan. This book was framed
in terms of a debate. On the one side are the CPEC optimists who see the
219
magnitude and long-term commitment of the CPEC as heralding a new

opportunity for Pakistan to realise its innate potential and perhaps become
that ‘Falcon Economy’. On the other side are the CPEC pessimists who see
China as engaged in predatory lending, anxious to tip Pakistan into a deeper
form of debt-dependency so as to leverage extra influence to gain control of
Pakistan’s resources, geography and policymaking.
This book, for the first time, utilised the large and rigorous economic
literature on the impact of big infrastructure projects on economic outcomes
to think about the likely impact of the CPEC. One positive for Pakistan
from this effort is the consistent and positive story of the economic impact
of big infrastructure on employment, investment, growth and productivity.
The size of these effects, though generally positive, is highly conditional
on whether new transport links represent a transformational change to the
existing transport system. The opportunity cost of investment is important
and very marginal gains to infrastructure financed by massive and expensive
borrowing are possible. The motivations for building infrastructure are
important and the threat of white elephants, or projects with no economic
rationale, or infrastructure enclaves that exist only to serve another country
are real possibilities. This book raised concerns about whether big CPEC
infrastructure will generate large and positive economic outcomes.
Scholars of the CPEC have made much of the dramatic reduction in
distance for firms in China to Europe to herald the fact that the CPEC will
make markets more efficient. But questions about efficiency have tended to
stop there. Scholars have rarely followed up this question with the question:
efficient to do what? This book has started to answer this question. There are
well-documented impacts from Pakistan’s experience of public investment in
infrastructure crowding-in private investment. Here, Chinese and Pakistani
state investment through the CPEC is just not enough to crowd-in enough
private investment to raise its share to the 30 per cent or more of GDP that
has been consistent with sustaining rapid economic growth in other countries.
There are also concerns about the quality of public investment in Pakistan;
too much public investment is wasted in poorly executed and politically
driven projects. This book also looked at production spillovers and found that
history offers us stories of diverse outcomes. In nineteenth-century Mexico,
India and Brazil, the construction of railways did create spillovers, but
these spillovers leaked out of the domestic economy to promote industriali-
sation overseas. In nineteenth-century Germany, those spillovers stimulated
220
CONCLUSION
domestic industrialisation. There are some grounds for optimism with regard
to domestic cement production in Pakistan and possibly also employment
where the CPEC does appear to be generating domestic spillovers. Concerns
remain, though, regarding iron and steel, glass and bitumen and the potential
for China to export these goods.
This book chose to evaluate the success of the CPEC in terms of what the
governments of China and Pakistan claim it will achieve. There are big claims
from both governments about the CPEC strengthening a complementary
relationship between China and Pakistan; that the CPEC will generate
win–win outcomes; that the CPEC will lead to economic transformation
(take flight the Falcon Economy). Both governments claim their two
economies are complementary in nature, and so expanded economic exchange
will offer significant benefits for each of them. The economic evolution of
Western China has been hitherto ignored by scholars of the CPEC. Existing
production of cotton and fruit and the rapid rise of textile production in
Xinjiang mean that Western China is becoming a more competitive threat
to Pakistan’s economy. Both governments ignore distributional implications
by claiming that the CPEC will represent a win–win outcome. This claim
ignores the voluminous historical and contemporary case study evidence
that shows big infrastructure always creates both winners and losers.
The safety valve of migration—people moving to take advantage of economic
opportunities created by the CPEC—may help in the case of Pakistan, which
is a country born of migration. If a geographic region gets tipped by the
CPEC into economic decline, it need not drag down the local inhabitants into
poverty; they can migrate to take advantage of opportunities being opened
up by the CPEC elsewhere. Women, though, are unlikely to take part in this
migration in any great number; there are severe cultural constraints on their
ability to work outside the household and especially to migrate long distances
in search of employment. Finally, there are doubts about the magnitude of
any CPEC impact on Pakistan’s economy. Pakistan’s roads and railways are
already in good enough condition, such that they are rarely cited as constraints
by business. The minimal price differentials that exist between cities indicate
that Pakistan already has fairly efficient markets. Evidence shows that there
are other, more significant constraints on economic growth, such as education
and skills, lack of long-term credit, enforcing contracts and political instability,
which will not be tackled by the CPEC.
221
SEZs have been widely used as a means to promote industrialisation,

technology transfer, employment growth and exports over the last 70 years.
Theory and empirical evidence show that they have a record that encompasses
both positive and negative consequences. The Government of Pakistan has
absorbed the positive side of this story without much critical reflection.
Nine SEZs are planned as part of the CPEC investment. There is much
optimism but a lack of critical thinking. Why will FDI flow to these SEZs
rather than the thousands that exist in countries with better governance?
Why will the CPEC-SEZs be a success when so many similar efforts have
failed in Pakistan’s own recent past? China’s own history with SEZs since
they were first established in 1979 does serve as an important marker, but it is
far-fetched to think that Gwadar or Pakistan can become a ‘new Shenzhen’.
Chinese constructed and invested SEZs in Africa have been a mixed success.
The most successful examples have been in Ethiopia, where they had been
driven by the proactive interventions of a developmental state. Pakistan does
not have an equivalent developmental state, nor does it have the clarity of
vision and strong leadership offered by Ethiopia. Recent foreign media reports
have suggested that the construction of SEZs will likely be delayed by the
Covid-19 outbreak (Economic Times 2020).
China has had a massive impact on world trade; an explosive growth of both
exports and imports from the 1980s onwards transformed the global trading
environment. Production and exports in both Africa and Latin America have
been pushed out of manufacturing and towards raw materials and primary
products. This contemporary process has been largely ignored by the CPEC
scholarship, which focuses instead on the opportunities offered by Chinese
FDI and markets. The FTA signed by China and Pakistan in 2006 has helped
China boost exports of manufactured goods to Pakistan. Pakistani exports
to China, which comprise mainly cotton yarn, have stagnated. This is hardly
surprising; it replicates the earlier experience of Africa and Latin America
and, as this book has argued, there are numerous constraints on industrial
growth in Pakistan that will not be tackled by the CPEC. Aspects of the 2006
FTA appear to have been poorly negotiated and there is a lack of ‘free’ in the
FTA that continues to deprive Pakistan of market access in various sectors in
which it had export potential. The margins of preference given to Pakistan
were significantly eroded by FTAs China signed with other countries in the
same period. China may be good friends with Pakistan but the warmth of its
friendship is hardly exclusive.
222
CONCLUSION
History has demonstrated that the successful use of an industrial policy in

nineteenth-century Germany, to a lesser extent in nineteenth-century Russia,
and also in contemporary Ethiopia is crucial to ensure that big infrastructure
generates benefits for the domestic economy. Failures related to productivity
growth, structural change and technological upgrading all contribute to a case
for an industrial policy in Pakistan. This case has been echoed by scholars who
see the CPEC as a means to leverage such policy changes in Pakistan.
Too often these calls jump from theory to policy and forget that Pakistan
currently lacks the necessary state capacity to successfully implement such
demanding practical policy changes. Policy is dominated by the politics of
short-term opportunism, not long-term dynamic efficiency, as revealed by
recent failures to implement just such a policy reform in textiles. Unlike
the official rhetoric, the CPEC will create both winners and losers. How
will Pakistan and its weak state and weak political parties cope with the
distributional implications of the CPEC, in particular the losers? The boom–
bust cycle under General Musharraf in the 2000s offers a salient example
of how Pakistan failed to turn an investment-led boom into sustainable and
rapid economic growth. There are grounds to dismiss other concerns.
The financing of the CPEC remains opaque but what data can be teased out
of the official documentation suggest that while we are right to have some
concerns, we have little reason to fear that China is engaged in predatory
lending in an attempt to push Pakistan into debt-led dependency.
So, to conclude, there is little reason to think that the CPEC will turn
Pakistan into a ‘Falcon Economy’. At the same time, there is not enough
evidence to justify a conclusion that the CPEC is a Chinese-dictated agenda
that aims to turn Pakistan into a debt-dependent and subservient ally.
The investment in energy capacity has already made a significant difference
in energy availability in Pakistan. Firms need to spend less time and expense
insulating themselves from the risk of electricity shortages with high-
cost private generators. Although the CPEC infrastructure is not likely to
transform Pakistan, there is a huge amount of global evidence to demonstrate
that big infrastructure does have positive impacts on investment, productivity
and growth over time. The long-term relationship with China, a country with
a long history of successful developmental interventions, is likely to reorient
Pakistan towards more long-term thinking in its economic policymaking.
Governance is unlikely to be transformed but is likely to be improved by this
close exposure to a teacher with demonstrated patience.
223
With some irony, it is often the supporters of the CPEC, those who label
the CPEC a game-changer, that are doing the greatest disservice to Pakistan.
One excitable conclusion was that the CPEC ‘is basically a collection of
infrastructure projects in Pakistan to develop Pakistan’s shattered economy’
(Chen, Joseph and Tariq 2018: 62). These views are a disservice because
of the vigorous enthusiasm with which they run down 70 years’ worth of
moderate economic success in Pakistan. Here we can return to and emphasise
the importance of the brief empirics in Chapter 1. The Tiger and Dragon
economies achieved their labels of miracle economies by accelerating structural
change from agriculture to industry in the 1960s and 1970s. Pakistan has
achieved this, but took 50 rather than 20 years. At the time of independence,
75 per cent of Pakistan’s GDP and 99 per cent of its exports were dependent
on agriculture and raw materials. By the mid-1990s, the share of industry had
overtaken the share of agriculture in GDP and four-fifths of Pakistan’s exports
were by then manufactured goods. In 1947 the vast bulk of Pakistan’s exports
went to India. By 2019 Pakistan exported to almost 200 countries, with the
largest shares going to China, the US, Afghanistan, Germany and the UK.
Pakistan does not have a shattered economy.
Recall the quote in Chapter 7 from the Pakistani minister who declared
that China is the only game in town. It is not. Pakistan should remember
that. Pakistan should not denigrate its own economic achievements since 1947
and convince itself that China as a saviour will provide a rescue act. Pakistan
should enter negotiations with China and everyone else with a degree of self-
belief and self-confidence. This will help Pakistan do much more to negotiate
better terms and conditions in lending and economic links (such as the China–
Pakistan FTA) by using not a feeling of vulnerable desperation but instead its
undoubted geopolitical importance to China, to Saudi Arabia, to Iran, to the
US and others as leverage. Recall the discussion in Chapter 7 how Ethiopia
used its role as a force for regional stability, its developmental vision and its
efforts to reduce poverty as leverages across multiple donors to maximise
developmental assistance. Pakistan should learn from this experience.
The efforts by China to promote economic development in Western China are
not just about investment, employment and exports; above all, they are about
the desire for political stability in Xinjiang, calming the civil and terrorist
disturbances that have wracked the region for the last couple of decades.
Pakistan is absolutely crucial to that effort. Pakistan may end up supplying
much of the raw materials and markets that will underpin the industrialisation
224
CONCLUSION
of Xinjiang. Pakistan should seek concessions elsewhere in recognition of that

effort. Beijing needs Pakistani help in reaching out into the world of militancy
to dissuade them from attacking the Chinese state. Pakistan’s links to the
Middle East, to Central Asia and to Afghanistan provide the pivot that China
needs to take the political heat out of its problems in Xinjiang (Small 2015).
Perhaps the easiest way to convince that a 15-year package of investments
will not generate an economic miracle is to read a description of a key corner
of the CPEC, the crossing from Pakistan into China:
Perched 4,693 metres above sea level in the Karakoram Mountains on

Pakistan’s northern border with the south-western region of Xinjiang, China,
Khunjerab Pass, with an international border gate, stands lonely against the
stunning snowy mountains around. On the Chinese side, besides an isolated
army border control station and a few adventurous tourists, the Pass is
accompanied by a few scattered yurts and rarely seen mountain yaks. (Chen,
Joseph and Tariq 2018: 61)
Recall for a final time that story of moderate economic success this book
documented in Chapter 2. After 70 years of moderate economic growth,
Pakistan is now too big and too industrial for $60 billion of investment
over 15 years to make a transformative difference. Pakistan has a GDP of
$320 billion and is a middle-income country. Laos by comparison is a low-
income, landlocked economy with a total GDP of some $20 billion. Here
Chinese infrastructure investment does have the ability to transform the
economy. Telling a more positive story about Pakistan should not, however,
distract policymakers. There are worries. Pakistan faces real challenges relating
to sustaining growth over the next decades, including the failure to diversify and
upgrade exports away from simple textile products into more technologically
demanding sectors and problems relating to education, governance and access
to credit. The game-changer optimists could unwittingly prove a distraction
in making policymakers think that the CPEC will provide a solution to all of
Pakistan’s woes.
225
Bibliography
Abbas, A., and F. Maaz (2017). ‘Joint Ventures under CPEC and Prospects for
Pakistan’s Industry: A Remarkable Yield of One Belt One Road Initiative’.
Centre of Excellence, CPEC, Working Paper No. 18, Islamabad.
Abbas, A., and S. Ali (2017a). ‘Nine Proposed Priority SEZs under CPEC and
SEZ Act: An Approach to Industrial Development’. Centre of Excellence,
CPEC, Working Paper No. 016, Islamabad.
——— (2017b). ‘China Regional Pattern and Prospects for Pakistan’. Centre of
Excellence, CPEC, Working Paper No. 015, Islamabad.
Afzal, A., M. McCartney and A. Asif (forthcoming). Pakistan, the Goldilocks
Economy: Economic Resilience 1960 to 2019.
Afzal, M., and Z. M. Nasir (1987). ‘If Female Labour Force Participation Really
Low and Declining in Pakistan? A Look at Alternative Data Sources’.
The Pakistan Development Review 26 (4): 699–709.
Ahmad, D. (2017). ‘The China Pakistan Economic Corridor: Review and
Analysis’. In Burki Institute of Public Policy, The State of the Economy: China
Pakistan Economic Corridor Review and Analysis, 50–63. Lahore: Shahid Javed
Burki Institute of Public Policy.
Ahmar, M. (1996). ‘Ethnicity and State Power in Pakistan: The Karachi Crisis’.
Asian Survey 36 (10): 1031–48.
Ahmed, E. (2010). ‘Why Is It So Difficult to Implement GST in Pakistan’. Lahore
Journal of Economics 15 (1): 139–69.
Ahmed, I., and A. Qayyam (2007). ‘Do Public Expenditure and Macroeconomic
Uncertainty Matter to Private Investment? Evidence from Pakistan’.
Akhtar, A. S. (2006). ‘The State as Landlord in Pakistani Punjab: Peasant Struggles
on the Okara Military Farms’. Journal of Peasant Studies 33 (3): 479–501.
Akhtaruzzaman, M., N. Berg and D. Lien (2017). ‘Confucius Institutes and FDI
Flows from China to Africa’. China Economic Review 44 (C): 241–52.
226
BIBLIOGRAPHY
Alahdad, Z. (2011). ‘Turning Energy Around’. In Pakistan: Beyond the Crisis State,
edited by M. Lodhi, 231–50. Karachi: Oxford University Press.
Alam, A., and A. A. Bhatti (2014). ‘Relative Commodity Price Convergence in
Pakistan’. School of Economics, International Islamic University, Islamabad.
Alder, S. (2014). ‘Chinese Roads in India: The Effect of Transport Infrastructure
on Economic Development’. University of Zurich, Switzerland.
Ali, L., J. Mi, M. Shah, S. J. Shah, S. Khan and K. Bibi (2017). ‘The Potential Socio-
Economic Impact of China Pakistan Economic Corridor’. Asian Development
Policy Review 5 (4): 191–8.
Ali, T. (2008). The Duel: Pakistan on the Flight Path of American Power. London:
Simon and Schuster.
Alkon, M. (2018). ‘Do Special Economic Zones Induce Developmental Spillovers?
Evidence from India’s States’. World Development 107 (C): 396–409.
Alves, A. (2013). ‘China’s “Win–Win” Cooperation: Unpacking the Impact
of Infrastructure-for-Resources Deals in Africa’. South African Journal of
International Affairs 20 (2): 207–26.
Aman-Rana, S. (2019). ‘Meritocracy in Bureaucracy? Evidence from Pakistan’.
London School of Economics, London.
Amendolagine, V., A. Boly, N. D. Coniglio, F. Prota and A. Seric (2013). ‘FDI
and Local Linkages in Developing Countries: Evidence from Sub-Saharan
Africa’. World Development 50 (C): 41–56.
Amjad, R. (1983). ‘Industrial Concentration and Economic Power’. In Pakistan:
The Roots of Dictatorship: The Political Economy of a Praetorian State, edited by
H. Gardezi and J. Rashid, 228–69. London: Zed Press.
Amjad, R., S. Chandrasiri, D. Nathan, S. Raihan, S. Verick and A. Yusuf (2015).
‘What Holds Back Manufacturing in South Asia’. The Economic and Political
Weekly (7 March): 36–45.
Amsden, A. H., and T. Hikino (2000). ‘The Bark Is Worse than the Bite: The
WTO Law and Late Industrialisation’. Annuals of the American Academy of
Political and Social Sciences 570 (1): 104–14.
Andrabi, T., and M. Kuehlwein (2010). ‘Railways and Price Convergence in
British India’. The Journal of Economic History 70 (2): 351–77.
Ansar, A., B. Flyvbjerg, A. Budzier and D. Lunn (2016). ‘Does Infrastructure
Investment Lead to Economic Growth or Economic Fragility? Evidence from
China’. Oxford Review of Economic Policy 32 (2): 360–90.
Appel, H. C. (2012). ‘Walls and White Elephants: Oil Extraction, Responsibility,
and Infrastructural Violence in Equatorial Guinea’. Ethnography 13 (4): 439–65.
227
BIBLIOGRAPHY
Aschauer, D. A. (1989a). ‘Is Public Expenditure Productive?’ Journal of Monetary

Economics 23 (2): 177–200.
——— (1989b). ‘Does Public Capital Crowd Out Private Capital?’ Journal of
Monetary Economics 24 (2): 171–88.
Asher, S., and P. Novosad (2016). ‘Market Access and Structural Transformation:
Evidence from Rural Roads in India’. Mimeo.
Asian Development Bank (2013). ‘Pakistan: Private Participation in Infrastructure
Program’. Validation Report PVR-300, Manila, Philippines.
——— (2016). ‘Islamic Republic of Pakistan: Karachi Bus Rapid Transit Project’.
Manila, Philippines.
Asian Productivity Organization (2013). APO Productivity Databook 2013. Tokyo:
Keio University Press.
Asian Review (2020). ‘Pakistan Request Opens Door for Belt and Road Project
Debt Relief ’. 11 May.
Asturias, J., M. Garcia-Santana and R. Ramos (2017). ‘Competition and the
Welfare Gains from Transportation Infrastructure: Evidence from the Golden
Quadrilateral of India’. Mimeo.
Atack, J., F. Bateman, M. Haines and R. A. Margo (2010). ‘Did Railroads Induce
or Follow Economic Growth? Urbanisation and Population Growth in the
American Midwest, 1850–1860’. Social Science History 34 (2): 171–97.
Atique, Z., M. H. Khan, U. Azhar and A. H. Khan (2004). ‘The Impact of FDI on
Economic Growth under Foreign Trade Regimes: A Case Study of Pakistan’.
Atkin, D., A. Chaudhury, S. Chaudry, A. Khandelwal, T. Raza and E. Verhoogen
(2015). ‘On the Origins and Development of Pakistan’s Soccer-Ball Cluster’.
Department of Economics, Lahore School of Economics, Lahore.
Auffray, C., and X. Fu (2015). ‘Chinese MNEs and Managerial Knowledge
Transfer in Africa: The Case of the Construction Sector in Ghana’. Journal of
Chinese Economic and Business Studies 13 (4): 285–310.
Baig, Z. (2018). ‘Cement Exports Increase by 18.41 Percent in February’. Business
Recorder, 3 March.
Balakrishnan, P. (2006). ‘Benign Neglect or Stategic Intent? Contested Lineage
of Indian Software Industry’. Economic and Political Weekly (9 September):
3865–72.
Banerjee, A., E. Duflo and N. Qian (2012). ‘On the Road: Access to Transportation
Infrastructure and Economic Growth in China’. National Bureau of Economic
Research Working Paper 17897, Cambridge, MA.
228
BIBLIOGRAPHY
Bardhan, P. (1984). The Political Economy of Development in India. New Delhi:

Oxford University Press.
Baum-Snow, N. (2007). ‘Did Highways Cause Suburbanisation’. The Quarterly
Baum-Snow, N., L. Brandt, J. V. Henderson, M. A. Turner and Q. Zhang (2012).
‘Roads, Railways, and Dcentralisation of Chinese Cities’. International
Growth Centre Working Paper, London School of Economics, London.
Becquelin, N. (2004). ‘Staged Development in Xinjiang’. The China Quarterly 178:
358–78.
Birdsall, N., and M. Kinder (2010). ‘The US Aid Surge to Pakistan: Repeating
a Failed Experiment: Lessons for US Policymakers from the World Bank’s
Social-Sector Lending in the 1990s’. Centre for Global Development
Working Paper No. 205, Washington, D.C.
Blanchard, I. (2000). ‘Russian Railway Construction and the Urals Charcoal Iron
and Steel Industry, 1851–1914’. Economic History Review 53 (1): 107–26.
Boni, F. (2016). ‘Civil–Military Relations in Pakistan: A Case Study of Sino-
Pakistani Relations and the Port of Gwadar’. Commonwealth and Comparative
Politics 54 (4): 546–65.
Bowen, R. H. (1950). ‘Rise of Modern Industry: The Role of Government and
Private Enterprise in German Industrial Growth, 1870–1914’. The Journal of
Economic History 10 (S1): 68–81.
Boyce, T. (2017). ‘The China–Pakistan Economic Corridor: Trade Security and
Regional Implications’. Sandia Report, California.
Brautigam, D. (2009). The Dragon’s Gift: The Real Story of China in Africa. London:
Brautigam, D., and T. Xiaoyang (2011). ‘African Shenzhen: China’s Special
Economic Zones in Africa’. Journal of Modern African Studies 49 (1): 27–54.
Brautigam, D., and X. Tang (2014). ‘Going Global in Groups: Structural
Transformation and China’s Special Economic Zones Overseas’. World
Development 63 (3): 78–91.
Brown, D., M. Fay, S. V. Lall, H. G. Wang and J. Felkner (2008). ‘Death of
Distance? Economic Implications of Infrastructure Improvement in Russia’.
European Investment Bank Papers 13 (2): 127–47.
Brugier, C. (2014). ‘China’s Way: The New Silk Road’. European Union Institute
for Security Studies, Brief Issue (May): 1–4.
Brun, L. (2016). Overcapacity in Steel: China’s Role Is a Global Problem. Washington,
D.C.: Duke Center on Globalization, Governance and Competitiveness.
229
BIBLIOGRAPHY
Bryceson, D. F., A. Bradbury and T. Bradbury (2008). ‘Roads to Poverty Reduction?

Exploring Rural Roads’ Impact on Mobility in Africa and Asia’. Development
Policy Review 26 (4): 459–82.
Burgess, R., R. Jedwab, E. Miguel, A. Morjana and G. P. Miquel (2015). ‘The
Value of Democracy: Evidence from Road Building in Kenya’. American
Economic Review 105 (6): 1817–51.
Burki, A. A., and D. Terrell (1998). ‘Measuring Production Efficiency of Small
Firms in Pakistan’. World Development 26 (1): 155–69.
Burki, S. J. (1999). Pakistan: Fifty Years of Nationhood. Boulder: Westview Press.
Business Recorder (2018). ‘Xinjiang’s Textile Park: A New Challenge’. 14 October.
Business Standard (2018). ‘Cheaper Pakistani Imports, High Tax at Home Hurt
Indian Cement Industry’. 23 September.
——— (2020). ‘CPEC Phase-II to Create Massive Employment Opportunities
in Pak: Asim Bajwa’. 5 February.
Busse, M., C. Erdogan and H. Muhlen (2016). ‘China’s Impact on Africa: The
Role of Trade, FDI and Aid’. Kyklos 69 (2): 228–62.
Cadot, O., L-H. Roller and A. Stephan (2006). ‘Contribution to Productivity or
Pork Barrel? The Two Faces of Infrastructure Investment’. Journal of Public
Economics 90 (6–7): 1133–53.
Cantos, P., M. Gumbau-Albert and J. Maudos (2005). ‘Transport Infrastructures,
Spillover Effects and Regional Growth: Evidence of the Spanish Case’.
Transport Review 25 (1): 25–50.
Cashin, P., and R. Sahay (1996). ‘Internal Migration, Center-State Grants, and
Economic Growth in the States of India’. IMF Staff Papers 43 (1): 123–71.
Castells, A., and A. Sole-Olle (2005). ‘The Regional Allocation of Infrastructure
Investment: The Role of Equity, Efficiency and Political Factors’. European
Economic Review 49 (5): 1165–1205.
Chakraborty, C., and C. Jayachandran (2001). ‘Software Sector: Trends and
Constraints’. Economic and Political Weekly (25 August): 3255–61.
Chandra, A., and E. Thompson (2000). ‘Does Public Infrastructure Affect
Economic Activity? Evidence from the Rural Interstate Highway’. Regional
Science and Urban Economics 30 (4): 457–90.
Chang, H-J. (1993). ‘The Political Economy of Industrial Policy in Korea’.
Cambridge Journal of Economics 17 (2): 131–57
———. (1999). ‘The Economic Theory of the Developmental State’. In The
Developmental State, edited by M. Woo-Cumings, 182–99. New York: Cornell
University Press.
———. (2002). Kicking Away the Ladder: Development Strategy in Historical
Perspective. London: Anthem Press.
230
BIBLIOGRAPHY
Chaudhry, A., and G. Andaman (2014). ‘The Need for a Coordinated Industrial
Strategy to Boost Pakistani Exports: Lessons from Asia’. The Lahore Journal of
Economics 19 (SE): 177–206.
Chaudhry, T., N. Jamil and A. Chaudhry (2017). ‘Pakistan’s Experience with the
Pakistan–China FTA: Lessons for CPEC’. The Lahore Journal of Economics 22
(SE): 1–24.
Chaziza, M. (2016). ‘China–Pakistan Relationship: A Game-changer for the
Middle East?’ Contemporary Review of the Middle East 3 (2): 147–61.
Cheema, A. A. (1978). ‘Productivity Trends in the Manufacturing Industries’. The
Pakistan Development Review 17 (1): 44–65.
Chen, D. (2014). ‘China’s “Marshall Plan” Is Much More’. The Diplomat, 10
November. http://thediplomat.com/2014/11/chinas-marshall-plan-is-much-
more/, accessed 12 April 2021.
Chen, S., S. K. Joseph and H. Tariq (2018). ‘Betting Big on CPEC’. European
Financial Review (March): 61–70.
Chen, W., D. Dollar and H. Tang (2018). ‘Why Is China Investing in Africa?
Evidence from the Firm Level’. The World Bank Economic Review 32 (3):
610–32.
Chibber, V. (2003). Locked in Place: State-Building and Late Industrialisation in
India. Princeton: Princeton University Press.
China Daily (2015). ‘Textile Industry Booms in Xinjiang’. 1 July.
——— (2017). ‘Investment in Xinjiang Textile Industry Booming’. 15 February.
——— (2020). ‘Textile Industry Weaving Way Out of Poverty in Xinjiang’.
22 January.
China Signpost (2010). ‘Overseas Trumps Overland: China’s Oil Supply Future is
Maritime’. 26 May. Concord, MA.
Chipaike, R., and R. H. Bischoff (2018). ‘A Challenge to Conventional Wisdom:
Locating Agency in Angola’s and Ghana’s Economic Engagements with
China’. Journal of Asian and African Studies 53 (7): 1–16.
Chohan, U. W. (2017). ‘What Is One Belt One Road? A Surplus Recycling
Mechanism Approach’. Discussion Paper Series: Notes on the 21st Century.
SSRN Electronic Journal https://ssrn.com/abstract=2997650.
Choudhri, E., A. Marasco and I. Nabi (2017). ‘Pakistan’s International Trade: The
potential for expansion towards East and West’. International Growth Centre,
F-37311-PAK-1, London.
Choudhury, S. (2017). Is CPEC Economic Corridor or a Strategic Game Plan?
Bloomington: AuthorHouse.
———. (2018). Economic Growth or a Debt Trap for Pakistan: CPEC Can Be a
Mega Disaster for Pakistan. Bloomington: AuthorHouse.
231
BIBLIOGRAPHY
Christoffersen, G. (1993). ‘Xinjiang and the Great Islamic Circle: The Impact of
Transnational Forces on Chinese Regional Economic Planning’. The China
Quarterly 133: 130–51.
Clapham, C. (2018). ‘The Ethiopian Developmental State’. Third World Quarterly
39 (6): 1151–65.
Clark, C. (2007). A Farewell to Alms: A Brief Economic History of the World.
Princeton: Princeton University Press.
Coatsworth, J. H. (1979). ‘Indispensable Railroads in a Backward Economy: The
Case of Mexico’. The Journal of Economic History 39 (4): 939–60.
Collier, P. (2007), The Bottom Billion: Why the Poorest Countries Are Failing and
What Can Be Done About It. Oxford: Oxford University Press.
Collins, G., and A. Erickson (2010). ‘Still A Pipedream: A Pakistan-to-China
Rail Corridor Is Not a Substitute for Maritime Transport’. China Signpost,
Concord, MA.
Collins, W. J. (1999). ‘Labor Mobility, Market Integration, and Wage Convergence
in Late 19th Century India’. Explorations in Economic History 36 (3): 246–77.
Connell, A. K. (1885). ‘Indian Railways and Indian Wheat’. Journal of the Statistical
Society of London 48 (2): 236–76.
Cotton, J. (1991). ‘The Limits to Liberalization in Industrializing Asia: Three
Views of the State’. Pacific Affairs 64 (3): 311–27.
Crain, W. M. and L. K. Oakley (1995). ‘The Politics of Infrastructure’. The Journal
of Law and Economics 38 (1): 1–17.
Crescenzi, R., and A. Rodriguez-Pose (2008). ‘Infrastructure Endowment
as Determinants of Regional Growth in the European Union’. European
Investment Bank Papers 13 (2): 63–100.
Daily News (2019). ‘Foreign Exchange Sources Remain Cornerstone of Egyptian
Economy in 2019’. 27 January.
Daily Times (2019). ‘Job Creation under CPEC: A Myth’. 27 February.
Dalgaard, C-J., N. Kaarsen, O. Olsson and P. Selaya (2018). ‘Roman Roads to
Prosperity: Persistence and Non-Persistence of Public Goods Provision’.
Centre for Economic Policy Research Discussion Paper CP12745, London.
Dannenberg, P., Y. Kim and D. Schiller (2013). ‘Chinese Special Economic Zones
in Africa: A New Species of Globalisation?’ African East Asian Affairs 2: 4–14.
Das, J., P. Pandev and T. Zajonc (2012). ‘Learning Levels and Gaps in Pakistan:
A Comparison with Uttar Pradesh and Madhya Pradesh’. Economic and
Political Weekly (30 June): 228–40.
Dasgupta, S., and A. Singh (2005). ‘Will Services Be the New Engine of Indian
Economic Growth?’ Development and Change 36 (6): 1035–57.
232
BIBLIOGRAPHY
———. (2006). ‘Manufacturing, Services and Premature Deindustrialisation in

Developing Countries’. UNU-WIDER Research Paper No. 2006/49.
Datta, S. (2012). ‘The Impact of Improved Highways on Indian Firms’. Journal of
Development Economics 99 (1): 46–57.
David, P. A. (1969). ‘Transport Innovation and Growth: Professor Fogel on and
off the Rails’. The Economic History Review 22 (3): 506–25.
Dawn (2015). ‘Islamabad’s Traders Refuse to Close Shop Early’. 13 April.
——— (2017a). ‘Exclusive: CPEC Master Plan Revealed’. 21 June.
——— (2017b). ‘CPEC SEZs: Myth and Reality’. 22 May.
——— (2018). ‘US Senators Seek to Block IMF Bailouts for China’s Allies’.
10 August.
——— (2019). ‘PTI Dissolves All Its Wings before Restructuring’. 19 August.
——— (2020). ‘Chinese Firms Begin Due Diligence of Pakistan Steel Mills’.
5 April.
Decressin, J., and A. Fatas (1995). ‘Regional Labour Market Dynamics in Europe’.
European Economic Review 39 (9): 1627–55.
Deininger, K., and S. Jin (2003). ‘Land Sales and Rental Markets in Transition:
Evidence from Rural Vietnam’. Oxford Bulletin of Economics and Statistics
70 (1): 67–101.
Demurger, S. (2001). ‘Infrastructure Development and Economic Growth:
An Explanation for Regional Disparities in China’. Journal of Comparative
Economics 29 (1): 95–117.
Deraniyagala, S., and B. Fine (1999). ‘New Trade Theory versus Old Trade Policy:
A Continuing Enigma’. Cambridge Journal of Economics 25 (6): 809–25.
Derbyshire, I. D. (1987). ‘Economic Change and the Railways in North India,
1860–1914’. Modern Asian Studies 21 (3): 521–45.
Di Caprio, A., and A. Amsden (2004). ‘Does the New International Trade Regime
Leave Room for Industrialisation Policies in the Middle-Income Countries’.
Policy Integration Department World Commission on the Social Dimension
of Globalisation, Working Paper No. 22, International Labour Office, Geneva.
Din, M., E. Ghani and T. Mahmood (2007). ‘Technical Efficiency of Pakistan’s
Manufacturing Sector: A Stochastic Frontier and Data Envelope Analysis’.
Dobado-Gonzalez, R. and G. Marrero (2005). ‘Corn Market Integration in
Porfirian Mexico’. The Journal of Economic History 65 (1): 103–28.
Donaldson, D. (2010). ‘Railroads of the Raj: Estimating the Impact of
Transportation Infrastructure’. LSE Asia Research Centre Working Paper
No. 41, London.
233
BIBLIOGRAPHY
Donaldson, D., and R. Hornbeck (2016). ‘Railroads and American Economic

Growth: A “Market Access” Approach’. The Quarterly Journal of Economics
131 (3): 799–858.
Doner, R. F., B. K. Ritchie and D. Slater (2005). ‘Systematic Vulnerability and
the Origins of Developmental States: Northeast and Southeast Asia in
Comparative Perspective’. International Organisation 59 (2): 327–61.
Du, J., and Y. Zhang (2018). ‘Does One Belt One Road Initiative Promote
Chinese Overseas Direct Investment?’ China Economic Review 47: 189–205.
Duflo, E., and R. Pande (2007). ‘Dams’. The Quarterly Journal of Economics
122 (2): 601–46.
Duranton, G., P. M. Morrow and M. A. Turner (2014). ‘Roads and Trade: Evidence
from the US’. Review of Economic Studies 81 (2): 681–724.
Easterly, W. (2001a). The Elusive Quest for Growth: Economists’ Adventures and
Misadventures in the Tropics. Cambridge, MA: MIT Press.
———. (2001b). ‘Can Institutions Resolve Ethnic Conflict?’ Economic Development
and Cultural Change 49 (4): 687–706.
Economic Times (2020). ‘Pakistan Admits Disruption in CPEC Due to Covid-19
Outbreak’. 27 April.
Eichengreen, B. (1995). ‘Financing Infrastructure in Developing Countries:
Lessons from the Railway Age’. The World Bank Research Observer 10 (1):
75–91.
Ejaz, M. (2007). ‘Determinants of Female Labour Force Participation in Pakistan:
An Empirical Analysis of PSLM (2004–05) Micro Data’. The Lahore Journal
of Economics 12 (SE): 203–35.
Ermisch, J. F., and W. G. Huff (1999). ‘Hypergrowth in an East Asian NIC:
Public Policy and Capital Accumulation in Singapore’. World Development
27 (1): 21–38.
Esteban, M. (2016). ‘The China–Pakistan Corridor: A Transit, Economic or
Development Corridor’. The Institute of Strategic Studies, Islamabad, 63–74.
European Union Chamber of Commerce (2016). ‘Overcapacity in China: An
Impediment to the Party’s Reform Agenda’. Beijing, China.
Express Tribune (2019a). ‘CPEC to Create 700,000 More Jobs in Pakistan’.
8 January.
——— (2019b). ‘Traders Refuse to Close Shop’. 23 April.
——— (2019c). ‘China–Pakistan Economic Corridor: Backdoor Meetings
Helped Forge Consensus’. 4 August.
——— (2020a). ‘PTI Government Excludes PSM from CPEC Framework’.
18 February.
234
BIBLIOGRAPHY
——— (2020b). ‘Pakistan’s Steel Sector Opposes Reduction in Tariffs’. 28 April.

Ezdi, R. (2009). ‘The Dynamics of Land-Use in Inner-City Lahore: The Case of
Mochi Gate’. Environment and Urbanization 21 (2): 477–501.
Faber, B. (2014). ‘Trade Integration, Market Size, and Industrialisation: Evidence
from China’s National Trunk Highway System’. Review of Economic Studies
81 (3): 1046–70.
Faini, R. (1983). ‘Cumulative Processes of De-industrialisation in an Open
Economy: The Case of Southern Italy, 1951–1973’. Journal of Development
Economics 12 (3): 277–301.
Farole, T., and L. Moberg (2014). ‘It Worked in China, So Why Not in Africa? The
Political Economy Challenge of Special Economic Zones’. WIDER Working
Paper No. 2014/152, The World Institute for Development Economics
Research (UNU-WIDER), Helsinki.
Fatima, A., and H. Sultana (2009). ‘Tracing Out the U-Shape Relationship
between Female Labor Force Participation Rate and Economic Development
for Pakistan’. International Journal of Social Economics 36 (1–2): 182–98.
Fei, D. (2018). ‘Work, Employment, and Training through Africa–China
Cooperation Zones: Evidence from the Eastern Industrial Zone in Ethiopia’.
China Africa Research Institute, Working Paper No.19, John Hopkins
University, Baltimore, Maryland.
Felipe, J. (2007). ‘A Note on Competitiveness and Structural Transformation’.
ERD Working Paper Series No 110, Asian Development Bank, Manila.
Fessehaie, J. (2012). ‘What Determines the Breadth and Depth of Zambia’s
Backward Linkages to Copper Mining? The Role of Public Policy and Value
Chain Dynamics’. Resources Policy 37 (4): 443–51.
Fessehaie, J., and M. Morris (2013). ‘Value Chain Dynamics of Chinese Copper
Mining in Zambia: Enclave or Linkage Development?’ European Journal of
Development Research 25 (4): 537–56.
Feyrer, J. (2009). ‘Distance, Trade and Income: The 1967 to 1975 Closing of
the Suez Canal as a Natural Experiment’. NBER Working Paper 15557,
Cambridge, MA.
Financial Express (2018). ‘Cement Producers Claim Cheaper Pakistani Exports,
High Tax at Home Hurting Indian Cement Industry’. 23 September.
Fine, B., and C. Stoneman (1996). ‘Introduction: State and Development’. Journal
of Southern African Studies 22 (1): 5–26.
Fletcher, M. E. (1958). ‘The Suez Canal and World Shipping, 1869–1914’. The
Journal of Economic History 18 (4): 556–73.
Fogel, R. W. (1960). The Union Pacific Railroad: A Case Study of Premature
Enterprise. Baltimore: John Hopkins University Press.
235
BIBLIOGRAPHY
———. (1966). ‘Railroads as an Analogy to the Space Effort: Some Economic

Aspects’. The Economic Journal 76 (301): 16–43.
Franck, R., and I. Rainer (2012). ‘Does the Leader’s Ethnicity Matter? Ethnic
Favoritism, Education and Health in Sub-Saharan Africa’. The American
Political Science Review 106 (2): 294–325.
Frankopan, P. (2018). The New Silk Roads: The Present and Future of the World.
London: Bloomsbury Publishing.
Fremdling, R. (1977). ‘Railroads and German Economic Growth: A Leading
Sector Analysis with a Comparison to the United States and Great Britain’.
The Journal of Economic History 37 (3): 583–604.
———. (1980). ‘Freight Rates and State Budget: The Role of the National
Prussian Railways 1880–1913’. The Journal of European Economic History
9 (1): 21–39.
Garcia-Mila, T., T. T. McGuire and R. H. Porter (1996). ‘The Effect of Public
Capital in State-Level Production Functions Reconsidered’. The Review of
Economics and Statistics, 78 (1): 177–80.
Garlick, J. (2018). ‘Deconstructing the China–Pakistan Economic Corridor: Pipe
Dreams versus Geopolitical Realities’. Journal of Contemporary China 27 (112):
519–33.
Ge, W. (1999). ‘Special Economic Zones and the Opening of the Chinese
Economy: Some Lessons for Economic Liberalisation’. World Development
27 (7): 1267–85.
Gebre-Egziabher, T. (2009). ‘The Developmental Impact of Asian Drivers on
Ethiopia with Emphasis on Small-scale Footwear Producers’. The World
Economy 32 (11): 1613–37.
Ghani, E., A. G. Goswani and W. R. Kerr (2014). ‘Highway to Success: The
Impact of the Golden Quadrilateral Project for the Location and Performance
of Indian Manufacturing’. Economic Journal 126 (591): 317–57.
Ghani, E., and M-Ud Din (2006). ‘The Impact of Public Investment on Economic
Growth in Pakistan’. The Pakistan Development Review 45 (1): 87–98.
Ghauri, S. P., A. Qayyum and M. Farooq (2013). ‘Price Level Convergence:
Evidence from Pakistan Cities’. Pakistan Economic and Social Review 51 (1):
1–12.
Giannecchi, P., and I. Taylor (2018). ‘The Eastern Industrial Zone in Ethiopia:
Catalyst for Development’. Geoform 88: 28–35.
Giovannetti, G., and M. Sanfilippo (2009). ‘Do Chinese Exports Crowd-out
African Goods? An Econometric Analysis by Country and Sector’. European
Journal of Development Research 21 (4): 506–30.
236
BIBLIOGRAPHY
Global Cement (2018a). ‘Pakistan: Flying Cement Orders Mill from Loesche’.
25 September.
——— (2018b). ‘Overcapacity’. 18 June.
——— (2019). ‘Pakistan’s Export Picture Mixed to February 2019’. 29 March.
Globaloue (2015). ‘Xinijiang Textile and Garment Industry Is Thriving’. https://
www.globaloue.com/en/page/news/408, accessed 13 April 2021.
Global Times (2015). ‘Executive Refute Notion That CPEC Offloads China’s
Overcapacity’. 6 July.
——— (2018a). ‘Gwadar Port on Route to Becoming a Replica of Shenzhen’.
19 October.
——— (2018b). ‘Chinese-Built Industrial Parks, Free Trade Zones Provide New
Industrialisation Momentum Across Africa’. 28 December.
——— (2019). ‘Trade War Takes Toll, But Xinjiang Exporters Seek to Diversify
Markets’. 20 November.
Global Village Space (2018). ‘Pakistan’s World of Cement: Opportunities and
Challenges’. 18 April.
Golden, M. A., and L. Picci (2008). ‘Pork-Barrel Politics in Postwar Italy,
1953–94’. American Journal of Political Science 52 (2): 268–89.
Gorg, H., and D. Greenaway (2004). ‘Much Ado About Nothing? Do Domestic
Firms Really Benefit from Foreign Direct Investment’. The World Bank
Research Observer 19 (2): 171–97.
Government of Australia (2018). China–Australia Free Trade Agreement. Canberra.
https://www.dfat.gov.au/, accessed 13 April 2021.
Government of China (2007). ‘Doing Business in Xinjiang Province of China’.
Ministry of Commerce, Beijing.
——— (2015a). ‘Vision and Actions on Jointly Building Silk Road Economic
Belt and 21st Century Maritime Silk Road’. Beijing.
——— (2015b). ‘Long-Term Plan on China-Pakistan Economic Corridor’.
Beijing.
Government of Pakistan (2017). ‘Long Term Plan for China-Pakistan Economic
Corridor 2017–2030’. Ministry of Planning, Development and Reform,
Islamabad.
——— (2018). ‘CPEC Projects Progress Update: CPEC Energy Projects’.
Islamabad.
——— (2019). Economic Survey 2018–2019. Islamabad: Finance Division.
Government of Punjab (2018). ‘Punjab Industrial Policy’. Punjab Industries,
Commerce & Investment Department, Lahore.
237
BIBLIOGRAPHY
Green, D. (2004). ‘Thailand’s Solar White Elephants: An Analysis of 15 Yr of

Solar Battery Charging Programmes in Northern Thailand’. Energy Policy
32 (6): 747–60.
Griffin, K. B. (1965). ‘Financing Development Plans in Pakistan’. The Pakistan
Development Review 5 (4): 601–30.
Griffiths, R. T. (2017). Revitalising the Silk Road: China’s Belt and Road Initiative.
Leiden: Hipe Publications.
Gu, J. (2009). ‘China’s Private Enterprises in Africa and the Implications for
African Development’. European Journal of Development Research 21 (4):
570–87.
Guardian (2016). ‘A New Shenzhen? Poor Pakistan Fishing Towns Horror at
Chinese Plans. 4 February.
Gulf News (2020). ‘End of an Era for Pakistan Steel Mills’. 4 June.
Gunasekara, K., W. Anderson and T. R. Lakshmanan (2008). ‘Highway-induced
Development: Evidence from Sri Lanka’. World Development 36 (11): 2371–89.
Gyllestal, M., and D. Ekstrom (2013). ‘FDI in Western China: A Study of the
Factors Behind Location Choices of Nordic Companies’. Department of
Economics, University of Gothenburg.
Habib, I. (2006). Indian Economy, 1858–1914. New Delhi: Tulika Books.
Haines, M. R., and R. A. Margo (2006). ‘Railroads and Local Economic
Development: The United States in the 1850s’. NBER Working Paper Series,
No. 12381, Cambridge, MA.
Hali, S. M., T. Shukui and S. Iqbal (2015). ‘One Belt and One Road: Impact on
China–Pakistan Economic Corridor’. Strategic Studies 34 (4): 147–64.
Hamid,N.,and M.Khan (2015).‘Pakistan: A Case of Premature Deindustrialisation’.
Lahore Journal of Economics 20 (SE) (September): 107–41.
Hamid, N., and S. Hayat (2012). ‘The Opportunities and Pitfalls of Pakistan’s
Trade with China and Other Neighbours’. The Lahore Journal of Economics 17
(SE): 271–92.
Hamid, S. (2010). ‘Rural to Urban Migration in Pakistan: The Gender Perspective’.
Pakistan Institute of Development Economics Working Papers 2010: 56,
Islamabad.
Hammes, J. J., and J-E. Nilsson (2016). ‘The Allocation of Transport Infrastructure
in Swedish Municipalities: Welfare Maximisation, Political Economy or
Both?’ Economics of Transportation 7: 53–64.
Hanusch, M. (2012). ‘African Perspectives on China–Africa: Modelling Popular
Perceptions and their Economic and Political Determinants’. Oxford
Development Studies 40 (4): 492–516.
238
BIBLIOGRAPHY
Haqqani, H. (2020). ‘Pakistan Discovers the High Cost of Chinese Investment’.

The Diplomat, 18 May.
Harriss, J. (2000). ‘How Much Difference Does Politics Make? Regime Differences
across Indian States and Rural Poverty Reduction’. LSE Development Studies
Institute, DESTIN, Working Paper No. 01, London.
Hausmann, R., L. Pritchett and D. Rodrik (2004). ‘Growth Accelerations’. JFK
School of Govt, Harvard University, April.
Heller, J. (1996). ‘Social Capital as a Product of Class Mobilisation and State
Intervention: Industrial Workers in Kerala, India’. World Development 24 (6):
1055–71.
Herranz-Loncan, A. (2006). ‘Railroad Impact in Backward Economies: Spain,
1850–1913’. The Journal of Economic History 66 (4): 853–81.
———. (2011). ‘The Role of Railways in Export-led Growth: The Case of
Uruguay, 1870–1913’. Economic History of Developing Regions 26 (2): 1–32.
Hillman, J. E. (2018). ‘China’s Belt and Road Is Full of Holes’. Centre for Strategic
and International Studies, CSIS Briefs, Washington D.C.
HKTDC (2018a). ‘Xinjiang: Market Profile’. Hong Kong Trade Development
Council, Hong Kong.
——— (2018b). ‘Pakistan: Market Profile’. Hong Kong Trade Development
Council, Hong Kong.
——— (2020). ‘Xinjiang: Market Profile’. Hong Kong Trade Development
Council, Hong Kong.
Hodler, R., and P. A. Raschky (2014). ‘Regional Favoritism’. The Quarterly Journal
of Economics 129 (2): 995–1033.
Hoekman, B. R., K. E. Maskus and K. Saggi (2005). ‘Transfer of Technology
to Developing Countries: Unilateral and Multilateral Policy Options’. World
Development 33 (10): 1587–602.
Holl, A. (2006). ‘A Review of the Firm-Level Role of Transport Infrastructure
with Implications for Transport Project Evaluation’. Journal of Planning
Literature 21 (1): 3–14.
——— (2016). ‘Highways and Productivity in Manufacturing Firms’. Journal of
Urban Economics 93 (C): 131–51.
Hornung, E. (2014). ‘Railroads and Growth in Prussia’. Journal of the European
Economic Association 13 (4): 699–736.
Hortidaily (2018). ‘China: Exports of Xinjiang Vegetables to Pakistan’.
15 February.
Hirschman, A. O. (1958). The Strategy of Economic Development. New Haven: Yale
University Press.
Huff, G. (1995). ‘What Is the Singapore Model of Development?’ Cambridge
Journal of Economics 19 (6): 735–59. 239
BIBLIOGRAPHY
Hurd, J. (1975). ‘Railways and the Expansion of Markets in India, 1861–1921’.

Explorations in Economic History 12 (3): 263–88.
Hurley, J., S. Morris and G. Portelance (2018). ‘Examining the Debt Implications
of the Belt and Road Initiative from a Policy Perspective’. Center for Global
Development Policy Paper No.121, Washington D.C.
Husain, I. (2003). Economic Management in Pakistan, 1999–2002. Karachi: Oxford
University Press.
———. (2017). ‘CPEC and Pakistani Economy: An Appraisal’. Centre of
Excellence for CPEC, Islamabad.
Hussain, E. (2019). ‘Will Change in Government Affect China–Pakistan
Economic Corridor: The BRI, CPEC, and the Khan Government: An
Analysis’. Chinese Journal of International Review 1 (2): 1–19.
Hussain, Z. (2010). Frontline Pakistan: The Path to Catastrophe and the Killing of
Benazir Bhutto. New York: I. N. Tauris & Co.
Hyder, K. (2001). ‘Crowding Out Hypothesis in a Vector Error Correction
Framework: A Case Study of Pakistan’. The Pakistan Development Review
40 (4): 633–50.
International Labour Organization (2012). Global Wage Report 2012/13. Geneva:
International Labour Office
——— (2018). ‘Global Wage Report 2018/19. Geneva: International Labour Office.
IMF (2017). ‘Pakistan Selected Issues’. IMF Country Report No. 17/213,
Washington, D.C.
Irfan, M. (1986). ‘Migration and Development in Pakistan: Some Selected Issues’.
Iqbal, E. (2017). ‘Recent Economic Developments and Prospects’. In Burki
Institute of Public Policy, The State of the Economy: China Pakistan Economic
Corridor Review and Analysis, 8–13. Lahore: Shahid Javed Burki Institute of
Public Policy.
Isaksson, A., and A. Kotsadam (2018). ‘Chinese Aid and Local Corruption’.
Journal of Public Economics 159 (C): 146–59.
Isserman, A., and T. Rephann (1995). ‘The Economic Effects of the Appalachian
Regional Commission: An Empirical Assessment of 26 Years of Regional
Development Planning’. Journal of the American Planning Association 61 (3):
345–64.
Jackson, T. (2014). ‘Employment in Chinese MNEs: Appraising the Dragon’s
Gift to Sub-Saharan Africa’. Human Resource Management 53 (6): 897–919.
Jalal, A. (1990). The State of Martial Rule: The Origins of Pakistan’s Political Economy
of Defence. Cambridge: Cambridge University Press.
240
BIBLIOGRAPHY
Jamil, N., and R. Arif (2018). ‘Boosting Export Opportunities for Pakistan:
Leveraging Pakistan’s Trade Policy with respect to China’. Paper presented at
the 14th International Conference on Management of the Pakistan Economy
Accelerating Economic Growth in Pakistan, Lahore.
Jan, F. and S. Granger (2016), ‘Pakistan–China Symbiotic Relations’. In Pakistan at
the Crossroads: Domestic Dynamics and External Pressures, edited by C. Jaffrelot,
279–300. Gurgaon: Penguin.
Javed, U. (2019). ‘Profit, Protest, and Power: Bazaar Traders in Urban Punjab’.
In New Perspectives on Pakistan’s Political Economy: State, Class and Social
Change, edited by M. McCartney and A. Zaidi, 199–215. New Delhi:
Cambridge University Press.
Jenkins, R. (2008). ‘Measuring the Competitive Threat from China for other
Southern Exporters’. The World Economy 31 (10): 1351–66.
———. (2010). ‘China’s Global Expansion and Latin America’. Journal of Latin
American Studies 42 (4): 809–37.
———. (2012). ‘Latin America and China: A New Dependency?’. Third World
Quarterly 33 (7): 1337–58.
Jenkins, R., and L. Edwards (2015). ‘Is China “Crowding Out” South African
Exports of Manufactures’. European Journal of Development Research 27 (5):
903–20.
Jenkins, R., E. D. Peters and M. M. Moreira (2008). ‘The Impact of China on
Latin America and the Caribbean’. World Development 36 (2): 235–53.
Johansson, H., and L. Nilsson (1997). ‘Export Processing Zones as Catalysts’.
World Development 25 (1): 2115–28.
Johnson, C. (1982). MITI and the Japanese Miracle: The Growth of Industrial Policy,
1925–1975. Stanford: Stanford University Press.
Jones, B. F. and B. A. Olken (2004). ‘The Anatomy of Stop–Start Growth’. The
Review of Economics and Statistics 90 (3): 582–7.
Kaldor, N. (1967). Strategic Factors in Economic Development. Ithaca: Cornell
University Press.
Kamal, J., and M. H. Malik (2017). ‘Dynamics of Pakistan’s Trade Balance with
China’. SBP Staff Notes 04/017, Islamabad.
Kanwal, H., T. A. Naveed and M. A. Khan (2015). ‘Socio-Economic Determinants
of Rural-Urban Migration in Pakistan’. Journal of Asian Development Studies
4 (3): 72–85.
Kaplinsky, R., D. McCormick and M. Morris (2007). ‘The Impact of China on
Sub-Saharan Africa’. IDS Working Paper, No. 291, University of Sussex,
Brighton.
241
BIBLIOGRAPHY
Kaplinsky, R., and M. Morris (2009). ‘Chinese FDI in Sub-Saharan Africa:

Engaging with Large Dragons’. The European Journal of Development
Research 21 (4): 551–69.
Kaplinsky, R., A. Terheggen and J. Tijaja (2010). ‘What Happens When the
Market Shifts to China? The Gabon Timber and Thai Cassava Value Chains’.
World Bank Policy Research Working Paper No. 5206, Washington.
Kapur, D. (2007). ‘The Causes and Consequences of India’s IT Boom’.
In Globalization and Politics in India, edited by B. R.Nayar, 387–407.
New Delhi: Oxford University Press.
Ke, S. (2010). ‘Determinants of Economic Growth and Spread-backwash
Effects in Western and Eastern China’. Asian Economic Journal 24 (2):
179–202.
Keck, O. (1988). ‘A Theory of White Elephants: Asymmetric Information in
Government Support for Technology’. Research Policy 17 (4): 187–01.
Keller, W., and C.H. Shiue (2008). ‘Tariffs, Trains and Trade: The Role of
Institutions versus the Technology in the Expansion of Markets’. Centre for
Economic Policy Research Working Paper No. 6759, London.
Kemal, A. R. (1999). ‘Patterns and Growth of Pakistan’s Industrial Sector’.
In Fifty Years of Pakistan’s Economy: Traditional Topics and Contemporary
Concerns, edited by A. R. Khan, 150–74. Karachi: Oxford University Press.
Kemmerling, A., and A. Stephan (2002). ‘The Contribution of Local Public
Infrastructure to Private Productivity and Its Political Economy: Evidence
From a Panel of Large German Cities’. Public Choice 113 (3–4): 403–424
Kenaway, E. (2016). ‘The Economic Impacts of the New Suez Canal’. IEMed
Mediterrean Yearbook 2016, Barcelona.
Khan, A. (2002).‘Pakistan’s Sindhi Ethnic Nationalism: Migration, Marginalisation
and the Threat of Indianisation’. Asian Survey 42 (2): 213–29.
Khan, A. H. (1988). ‘Macroeconomic Policy and Private Investment in Pakistan’.
——— (2000). ‘Determinants of Internal Migration in Pakistan: Evidence from
the Labour Force Survey, 1996–97’. The Pakistan Development Review 39 (4):
695–712.
Khan, A. H. and L. Shehnaz (2000). ‘Determinants of Internal Migration in
Pakistan: Evidence from the Labour Force Survey, 1996–97’. The Pakistan
Khan, B. A. (1999). ‘Financial Markets and Economic Development in Pakistan:
1947–1995’. In 50 Years of Pakistan’s Economy: Traditional Topics and
Contemporary Concerns, edited by S. R. Khan. Karachi: Oxford University
Press.
242
BIBLIOGRAPHY
Khan, M. H. (2008). ‘Technological Upgrading in Bangladeshi Manufacturing:

Governance Constraints and Policy Responses in the Readymade Garments
Industry’. SOAS, London.
———. (2011). ‘The Economic Development of Bangladesh: Identifying
Pragmatic Policy Responses’. In Bangladesh Tomorrow, edited by M. Ullah,
11–68. Dhaka: CFSD.
———. (2013a). ‘Political Settlements and the Design of Technology Policy’.
In The Industrial Policy Revolution II. Africa in the Twenty-First Century, edited
by J. Stiglitz, J.Y. Lin and E. Patel, 243–80. London: Palgrave.
———. (2013b). ‘Technology Policies and Learning with Imperfect Governance’.
In The Industrial Policy Revolution I: The Role of Government Beyond Ideology’,
edited by J. Stiglitz and J.Y. Lin, 79–115. London: Palgrave.
Khan, M. H., and S. Blankenburg (2009). ‘The Political Economy of industrial
Policy in Asia and Latin America’. In (2009). Industrial Policy and Development,
edited by G. Dossi and M. Cimoli, 336–77. Oxford: Oxford University Press.
Khan, S. H. (2009). ‘Making People Employable: Reforming Secondary Education
in Pakistan’. The Pakistan Development Review 48 (4): 603–17.
Khanna, G. (2016). ‘Road Oft Taken: The Route to Spatial Development’.
Department of Economics, University of Michigan.
Khanna, P. (2019). The Future Is Asian: Global Order in the Twenty-first Century.
London: W&N.
Khattry, B. (2003). ‘Trade Liberalisation and the Fiscal Squeeze: Implications for
Public Investment’. Development and Change 34 (3): 401–24.
Khattry, B., and J. M. Rao (2002). ‘Fiscal Faux Pas? An Analysis of the Revenue
Implications of Trade Liberalisation’. World Development 30 (8): 1431–44.
Khwaja, A. I., and A. Mian (2005). ‘Do Lenders Favor Politically Connected
Firms? Rent Provision in an Emerging Financial Market’. The Quarterly
Kim, Y. (2013). ‘Chinese-led SEZs in Africa: Are They a Driving Force of
China’s Soft Power’. Centre for Chinese Studies Discussion Paper, 1/2013,
Stellenbosch University.
Kite, G. (2013). ‘India’s Software and IT Services Revolution: A Teacher to
Treasure’. Economic and Political Weekly (27 July): 164–72.
Kite, G., and M. McCartney (2017). ‘Pro-business and Pro-market Reforms In
Pakistan: Economic Growth and Stagnation 1950–51 to 2011–12’. Journal of
Asia Pacific Economy 22 (3): 1–17.
Kohli, A. (2004). State-Directed Development: Political Power and Industrialisation
in the Global Periphery. Cambridge: Cambridge University Press.
243
BIBLIOGRAPHY
Kozel, V., and H. Alderman (1990). ‘Factors Determining Work Participation

and Labour Supply Decisions in Pakistans Urban Areas’. The Pakistan
Krieckhaus, J. (2002). ‘Reconceptualising the Developmental State: Public Savings
and Economic Growth’. World Development 30 (10): 1697–712.
Kumar, A. (2013). ‘Development Focus and Electoral Success at State Level:
Nitish Kumar as Bihar’s Leader’. South Asia Research 33 (2): 101–21.
Kumar, S. (2007). ‘The China–Pakistan Strategic Relationship: Trade, Investment,
Energy and Infrastructure’. Strategic Analysis 31 (5): 757–90.
Lai, H. H. (2002). ‘China’s Western Development Programme: Its Rationale,
Implementation, and Prospects’. Modern China 28 (4): 432–66.
Lall, S. (1992). ‘Technological Capabilities and Industrialisation’. World
Development 20 (2): 165–86.
———. (1994). ‘The East Asian Miracle: Does the Bell Toll for Industrial
Strategy’. World Development 22 (4): 645–54.
———. (2000). ‘The Technological Structure and Performance of Developing
Country Manufactured Exports, 1985–98’. Oxford Development Studies 28 (3):
337–69.
Lall, S., J. Weiss and H. Oikawa (2005). ‘China’s Competitive Threat to Latin
America: An Analysis for 1990–2002’. Oxford Development Studies 33 (2):
163–94.
Leah, M. C., and A. H. Munnell (1990). ‘How Does Public Infrastructure
Affect Regional Economic Performance?’ New England Economic Review
(September): 11–32.
Lee, W. R. (1988). ‘Economic Development and the State in Nineteenth-Century
Germany’. The Economic History Review 41 (3): 346–67.
Leftwich, A. (1995). ‘Bringing Politics Back In: Towards a Model of the
Developmental State’. Journal of Development Studies 31 (3): 400–27.
———. (2000). States of Development: On the Primacy of Politics in Development.
Cambridge: Polity Press.
Levi-Faur, D. (1998). ‘The Developmental State: Israel, South Korea and Taiwan
Compared’. Studies in Comparative International Development 33 (1): 65–93.
Levine, R., and D. Renelt (1992). ‘A Sensitivity of Cross-Country Growth
Regressions’. American Economic Review 82 (4): 942–63.
Lewis, B. D. (1998). ‘The Impact of Public Infrastructure on Municipal Economic
Development: Empirical Results from Kenya’. Review of Urban and Regional
Development Studies 10 (2): 142–56.
Li, H., and Z. Li (2013). ‘Road Investments and Inventory Reduction: Firm Level
Evidence from China’. Journal of Urban Economics 76: 43–52.
244
BIBLIOGRAPHY
Li, Z., and Y. Chen (2013). ‘Estimating the Social Return to Transport Infra-
structure: A Price-Difference Approach Applied to a Quasi-Experiment’.
Journal of Comparative Economics 41 (3): 669–83.
Liang, Z. (1999). ‘Foreign Investment, Economic Growth, and Temporary
Migration: The Case of Shenzhen Special Economic Zone, China’.
Development and Society 28 (1): 115–37.
Lieven, A. (2011). Pakistan: A Hard Country. London: Allen Lane.
Lin, J. Y. (2010). ‘Six Steps for Strategic Government Intervention’. Global Policy
1 (3): 330–1.
Lin, Y. (2017). ‘Travel Costs and Urban Specialisation Patterns: Evidence from
China’s High Speed Railway System’. Journal of Urban Economics 98 (C):
98–123.
Litwack, J. M., and Y. Qian (1998). ‘Balanced or Unbalanced Development:
Special Economic Zones as Catalysts for Transition’. Journal of Comparative
Economics 26 (1): 117–41.
Livemint (2019). ‘CPEC to Inflict Heavy Debt Burden on Pakistan: US’. 24
November.
Loayza, N., and T. Wada (2012). ‘Public Infrastructure Trends and Gaps in
Pakistan’. World Bank Policy Paper Series on Pakistan PK 10/12, Washington.
Looney, R. E. (1997). ‘Infrastructure and Private Sector Investment in Pakistan’.
Journal of Asian Economies 8 (3): 393–420.
Loriaux, M. (1999). ‘The French Developmental State as Myth and Moral
Ambition’. In The Developmental State, edited by M. Woo-Cumings, 235–75.
New York: Cornell University Press.
Lovell, J. (2018). Maoism: A Global History. London: Bodley Head.
Lu, J., and Z. Tao (2009). ‘Trends and Determinants of China’s Industrial
Agglomeration’. Journal of Urban Economics 65 (2): 167–80.
Lu, Z., and X. Deng (2011). ‘China’s Western Development Strategy: Policies,
Effects and Prospects’. MPRA Paper No. 35201, Berlin.
Macaes, B. (2018a). The Dawn of Eurasia: On the Trail of the New World Order.
London: Allen Lane.
———. (2018b). Belt and Road: A Chinese World Order. London: C. Hurst and Co.
Mangan, J. A. (2008). ‘Prologue: Guarantee of Global Goodwill: Post-Olympic
Legacies—Too Many Limping White Elephants?’ The International Journal of
the History of Sport 25 (14): 1869–83.
Mardon, R. (1990). ‘The State and the Effective Control of Foreign Capital: The
Case of South Korea’. World Politics 43 (1): 111–38.
Marshall, A. (1920). Principles of Economics, 8th ed. London: MacMillan and Co.
245
BIBLIOGRAPHY
Marshall, S. (2015). The Egyptian Armed Forces and the Remaking of an Economic
Empire. Washington, DC: Carnegie Middle East Centre.
Maurer, N., and C. Yu (2008). ‘What T.R. Took: The Economic Impact of the
Panama Canal, 1903–1937’. The Journal of Economic History 68 (3): 686–721.
McCartney, M. (2009). India: The Political Economy of Growth, Stagnation and the
State, 1951–2007. London : Routledge.
——— (2011a). ‘Pakistan, Growth, Dependency and Crisis, 1951–2009’. The
Lahore Journal of Economics 16 (SE): 71–94.
——— (2011b). Pakistan: The Political Economy of Growth, Stagnation and the
State, 1951–2008’. London: Routledge.
——— (2012). ‘Competitiveness and Pakistan: A Dangerous, Distorting and
Dead-End Obsession?’ Lahore Journal of Economics 17 (SE): 213–41.
——— (2015a). ‘The Missing Economic Magic: The Failure of Trade
Liberalisation and Exchange Rate Devaluation in Pakistan, 1980–2015’. The
——— (2015b). Economic Growth and Development: A Comparative Introduction.
London: Palgrave MacMillan.
——— (2015c). ‘From Boom to Bust: Economic Growth and Security in Pakistan
2003–2013’. In Democratic Transition and Security in Pakistan, edited by
S. Gregory, 76–101. London: Routledge.
Mearsheimer, J. J. (2010). ‘The Gathering Storm: China’s Challenge to US Power
in Asia’. The Chinese Journal of International Politics 3 (4): 381–96.
Mehmood, Z. (2017). ‘A Win–Win Proposal to Establish China-Pakistan Special
Economic Zones’. Hilal English. https://www.hilal.gov.pk/eng-article/a-win-
win-proposal-to-establish-china-pakistan-special-economic-zones/NDYy.
html, accessed 15 April 2021.
Menon, V. (2003). From Movement to Government: The Congress in the United
Provinces, 1937–42. New Delhi: Sage.
Metzer, J. (1974). ‘Railroad Development and Market Integration: The Case of
Tsarist Russia’. The Journal of Economic History 34 (3): 529–50.
Miao, M., O. Lang, D. G. Borojo, J. Yushi and X. Zhang (2020). ‘The Impacts
of Chinese FDI and China–Africa Trade on Economic Growth of African
Countries: The Role of Institutional Quality’. Economies 8 (3): 1–20.
Michaels, G. (2008). ‘The Effect of Trade on the Demand for Skill: Evidence
from the Interstate Highway System’. The Review of Economics and Statistics
90 (4): 683–701.
246
BIBLIOGRAPHY
Moberg, L., and V. Tarko (2014). ‘Why No Chinese Miracle in Africa?

Special Economic Zones and Liberalization Avalanches’. https://ssrn.com/
abstract=2352520. Modern Diplomacy (2020). ‘CPEC: Whether a Debt Trap
for Pakistan or Not?’ 22 May.
Mohanty, M. (2007). ‘Political Economy of Agrarian Transformation: Another
View of Singur’. Economic and Political Weekly (3 March): 737–41.
Mohsin, H. M., and S. Gilbert (2010). ‘Relative City Price Convergence in
Pakistan: Empirical Evidence from Spatial GLS’. MPRA Paper No. 27901,
Munich.
Moneyhon, M. D. (2003). ‘China’s Great Western Development Project in
Xinjiang: Economic Palliative or Political Trojan Horse’. Denver Journal of
International Law and Policy 31 (3): 491–519.
Moramudali (2020). ‘The Hambantota Port Deal: Myths and Realities’ The
Diplomat, 1 January.
Mostafa, M. M. (2004). ‘Forecasting the Suez Canal traffic: A Neural Network
Analysis’. Maritime Policy and Management 31 (2): 139–56.
Mourdoukoutas, P. (2018). ‘What Is China Doing to Pakistan? The Same Thing
It Did to Sri Lanka’. Forbes. 15 April.
Morris, M., and G. Einhorn (2008). ‘Globalisation, Welfare and Competitiveness:
The Impacts of Chinese Imports on the South African Clothing and Textile
Industry’. Competition and Change 12 (4): 355–76.
Mukherjee, M. (1980). ‘Railways and Their Impact on Bengal’s Economy,
1870–1920’. The Indian Economic and Social History Review 17 (2): 191–209.
Munshi, K., and M. R. Rosenzweig (2005). ‘Why Is Mobility in India so Low?
Social Insurance, Inequality and Growth’. CID Working Paper No. 121,
Harvard University.
Naqvi, N. H. (2002). ‘Crowding Out or Crowding in? Modelling the Relationship
between Public and Private Fixed Capital Formation Using Co-integration
Analysis: The Case of Pakistan’. The Pakistan Development Review 41 (3):
255–75.
Nathan Associates (2009). ‘Cost Competitiveness of Pakistan’s Textiles and
Apparel Industry’. USAID, Washington, D.C.
NDTV (2018). ‘US Warns Against IMF Bailout for Pak That Could Help Pay
Off China Loans’. 31 July.
New York Times (2018). ‘How China Got Sri Lanka to Cough Up a Port’. 25 June.
Nicolas, D. P. (2015). ‘Chinese Infrastructure in South Asia: A Realist and Liberal
Perspective’. Naval Postgraduate School, Monterey, CA.
Nicolas, F. (2017). ‘Chinese Investors in Ethiopia: The Perfect Match?’ OCP
Policy Centre and IFRI, Paris.
247
BIBLIOGRAPHY
Nolan, P. (2001). China and the Global Economy: National Champions, Industrial
Policy and the Big Business Revolution. Basingstoke: Palgrave.
Noman, A. (2015). ‘The Return of Industrial Policy and Revival of Pakistan’s
Economy: Possibilities of Learning, Industrial and Technology Policies’. The
Ó Riain, S. (2000). ‘The Flexible Developmental State: Globalization, Information
Technology and the “Celtic Tiger”’. Politics and Society 28 (2): 157–93.
Ottaviano, G. I. P. (2008). ‘Infrastructure and Economic Geography: An Overview
of Theory and Evidence’. European Investment Bank Papers 13 (2): 9–35.
Ovadia, J. S. (2013). ‘Accumulation with or without Dispossession? A “Both/And”
Approach to China in Africa with Reference to Angola’. Review of African
Political Economy 40 (136): 233–50.
Oyeranti, O. A., A. M. Babatunde, E. O. Ogunkola and A. S. Bankole (2010). ‘The
Impact of China–Africa Investment Relations: The Case of Nigeria’. Policy
Brief No. 8. https://www.africaportal.org/publications/the-impact-of-china-
africa-investment-relations-the-case-of-nigeria/, accessed 14 April 2021.
Pagano, A., G. Wang, O. Sanchez, R. Ungo and E. Tapiero (2012). ‘Impact of the
Panama Canal Expansion on the Panamanian economy’. Maritime Policy and
Management 39 (7): 705–22.
———. (2016). ‘The Impact of the Panama Canal Expansion of Panama’s
Maritime Cluster’. Maritime Policy and Management 43 (2): 164–78.
Pakistan Business Council (2018). Pakistan’s Readymade Garments Sector: Challenges
and Opportunities. Karachi: Pakistan Business Council.
———. (2019). 5th Review of the China-Pakistan Free Trade Agreement with
Recommendations for Phase II Negotiations. Karachi: Pakistan Business Council.
Palpacuer, F., P. Gibbon and L. Thomsen (2005). ‘New Challenges for Developing
Country Suppliers in Global Clothing Chains: A Comparative European
Perspective’. World Development 33 (3): 409–30.
Parente, R., K. Rong, J-M. G. Geleilate and E. Misati (2018). ‘Adapting and
Sustaining Operations in Weak Institutional Environments: A Business
Ecosystem Assessment of a Chinese MNE in Central Africa’. Journal of
International Business Studies 50 (2): 275–91.
Perveen, A. (1993). ‘Inter-Provincial Migration in Pakistan 1971–1981’. The
Pakistan Development Review 32 (4): 725–35.
Piracha, M., and M. Moore (2015). ‘Understanding Low-Level State Capacity:
Property Tax Collection in Pakistan’. International Centre for Tax and
Development Working Paper No. 33, IDS, Sussex.
Presbitero, A. F. (2016). ‘Too Much and Too Fast? Public Investment Scaling Up
and Absorptive Capacity’. Journal of Development Economics 120: 17–31.
248
BIBLIOGRAPHY
Press TV (2020). ‘Pressure Sought on Pakistan to Complete Iran Gas Pipeline’.

7 January.
Puga, D. (2008). ‘Agglomeration and Cross-border Infrastructure’. European
Investment Bank Papers 13 (2): 103–24.
Puhani, P. A. (2001). ‘Labour Mobility: An Adjustment Mechanism in Euroland?
Empirical Evidence for Western Germany, France and Italy’. German Economic
Review 2 (2): 127–40.
Qin, Y. (2014). ‘“No Country Left Behind?” The Distributional Impact of High-
Speed Rail Upgrade in China’. IRS Working Paper 2014-024, National
University of Singapore, Singapore.
Raheman, A., T. Afza, A. Qayyum and M. A. Bodla (2008). ‘Estimating Total
Factor Productivity and Its Components: Evidence from Major Manufacturing
Industries of Pakistan’. The Pakistan Development Review 47 (4): 677–94.
Ranaraja, Y. (2020). ‘Is Hambantota International Port Better Off with China?’
Sea Time Maritime News. https://www.seatrade-maritime.com/opinions-
analysis/hambantota-international-port-better-china, accessed 30 July 2020.
Renard, M-F. (2011). ‘China’s Trade and FDI in Africa’. African Development
Bank Group Working Paper, No. 126, Tunisia.
Rephamm, T., and A. Isserman (1994). ‘New Highways as Economic Development
Tools: An Evaluation Using Quasi-Experimental Matching Methods’.
Regional Science and Urban Economics 24 (6): 723–51.
Robinson, J. A., and R. Torvik (2005). ‘White Elephants’. Journal of Public
Economics 89 (2–3): 197–210.
Rodrik, D. (2006). Industrial Development: Stylised Facts and Policies. Cambridge:
Harvard University.
———. (2008). ‘Industrial Policy: Don’t Ask Why, Ask How’. Middle East
Development Journal, Demo Issue: 1–29.
Rosenstein-Rodan, P. N. (1943). ‘Problems of Industrialisation of Eastern and
South-Eastern Europe’. Economic Journal 53 (210/211): 202–11.
Rostow, W. W. (1956). ‘The Take-Off into Self-Sustained Growth’. Economic
Journal 66 (261): 25–48.
———. (1960). The Stages of Economic Growth: A Non-Communist Manifesto.
Cambridge: Cambridge University Press.
———. (1975). How It All Began: Origins of the Modern Economy. London:
Methuen.
Rothermund, D. (1993). An Economic History of India: From Pre-Colonial Times to
1991, 2nd ed. London: Routledge.
Roy, T. (2002). ‘Economic History and Modern India: Redefining the Link’.
Journal of Economic Perspectives 16 (3): 109–30.
249
BIBLIOGRAPHY
Rud, J. P. (2012). ‘Electricity Provision and Industrial Development: Evidence

from India’. Journal of Development Economics 97 (2): 352–67.
Samaa TV (2017). ‘Government Announces Attractive Terms in CPEC-related
Projects in SEZs’. 18 November.
———. (2019). ‘Pakistan’s Markets Closed as Traders Protest Increase in Taxes’.
13 July.
Sandrey, R., and H. Edinger (2011). ‘China’s Manufacturing and Industrialisation
in Africa’. African Development Bank Group Working Paper No. 128,
Washington.
Sarkar, A. (2007). ‘Development and Displacement: Land Acquisition in West
Bengal’. Economic and Political Weekly (21 April): 1435–42.
Schrank, A. (2001). ‘Export Processing Zones: Free Market Islands or Bridges to
Structural Transformation’. Development Policy Review 19 (2): 223–42.
Scitovsky, T. (1954). ‘Two Concepts of External Economies’. Journal of Political
Economy 62 (143): 143–51.
Seyoum, M., R. Wu and L. Yang (2015). ‘Technology Spillovers from Chinese
Outward Direct Investment: The Case of Ethiopia’. China Economic Review
33 (C): 35–49.
Shabir, S., and R. Kazmi (2007). ‘Economic Effects of the Recently Signed
Pak–China Free Trade Agreement’. The Lahore Journal of Economics 12 (SE,
September): 173–202.
Shafqat, S., and S. Shahid (2018). China Pakistan Economic Corridor: Demands,
Dividends and Directions. Lahore: Centre for Public Policy and Governance,
Forman Christian College.
Sherdil (2017). ‘CPEC Scope, Statue and Potential Impact’. In Burki Institute
of Public Policy, The State of the Economy: China Pakistan Economic Corridor
Review and Analysis, 65–84. Lahore: Shahid Javed Burki Institute of Public
Policy.
Shi, Y., S. Guo and P. Sun (2017). ‘The Role of Infrastructure in China’s Regional
Economic Growth’. Journal of Asian Economics 49 (C): 26–41.
Sial, S. (2014). ‘The China–Pakistan Economic Corridor: An Assessment of
Potential Threats and Constraints’. Conflict and Peace Studies 6 (2): 1–20.
Siddiqui, R., H. H. Jalil, M. Nasir, W. S. Malik and M. Khalid (2011), ‘The Cost
of Unserved Energy: Evidence from Selected Industrial Cities of Pakistan’.
PIDE Working Papers 2011:75, Islamabad.
Sinha, A. (2005). The Regional Roots of Developmental Politics in India. Indianapolis:
Indiana University Press.
250
BIBLIOGRAPHY
Sklair, L. (1991). ‘Problems of Socialist Development: The Significance of

Shenzhen Special Economic Zone for China’s Open Door Development
Strategy’. International Journal of Urban and Regional Research 15 (2): 197–215.
Small, A. (2015). The China–Pakistan Axis: Asia’s New Geopolitics. Gurgaon:
Penguin Random House.
Sonobe, T., J. E. Akoten and K. Otsuka (2009). ‘An Exploration into the
Successful Development of the Leather-Shoe Industry in Ethiopia’. Review of
Development Economics 13 (4): 719–73.
Sourcing Journal (2014). ‘China: Xinjiang Plans $3.2 billion Fund to Improve
Textile Industry’. 21 July.
Speakman, J., K. Afzal, Y. Yuge and J. Hanna (2012). ‘Toward an Innovation
Policy for Pakistan’. World Bank Policy Paper Series on Pakistan, PK06/12,
Washington, DC.
Spectre Energy (2020). ‘Our Projects’. http://spectrenergy.com/projects.php,
accessed 14 July 2020.
Stanev, K., E. J. Alvarez-Palau and J. Marti-Henneberg (2017). ‘Railway Develop-
ment and the Economic and Political Integration of the Balkans, c.
1850–2000’. Europe–Asia Studies 69 (10): 1601–25.
Strafor (2013). ‘China’s Ambitions in Xinjiang and Central Asia’. 30 September,
Austin, Texas.
Summers, T. (2016). ‘China’s New Silk Roads: Sub-National Regions and
Networks of Global Political Economy’. Third World Quarterly 37 (9):
1628–43.
Summerhill, W. R. (2005). ‘Big Social Savings in a Small Laggard Economy:
Railroad-led Growth in Brazil’. The Journal of Economic History 65 (1): 72–102.
Swaminathan, M. (1990). ‘Village Level Implementation of IRDP: Comparison
of West Bengal and Tamil Nadu’. Economic and Political Weekly (31 March):
17–27.
Tang, D., and K. N. Gyasi (2012). ‘China–Africa Foreign Trade Policies: The
Impact of China’s Foreign Direct Investment (FDI) Flow on Employment of
Ghana’. Energy Procedia 16: 553–7.
Tang, J. P. (2014). ‘Railroad Expansion and Industrialisation: Evidence from Meiji
Japan’. The Journal of Economic History 74 (3): 863–86.
Tang, K. (2019). ‘Lessons from East Asia: Comparing Ethiopia and Vietnam’s
Early Stage Special Economic Zone Development’. China Africa Research
Initiative, Working Paper No. 36, John Hopkins University.
Tehsin, M., A. A. Khan and T. Sargana (2017). ‘CPEC and Sustainable Economic
Growth for Pakistan’. Pakistan Vision 18 (2): 102–18.
251
BIBLIOGRAPHY
Tewari, M. (2005). ‘Post-MFA Adjustments in India’s Textile and Apparel

Industry: Emerging Issues and Trends’. Indian Council for Research on
International Economic Relations, Working Paper No. 167, New Delhi.
Textile Focus (2017). ‘The Future of The New Textile Industry: A Dialogue
between Xinjiang, China and Europe’. 31 May.
Tharoor, S. (2016). An Era of Darkness: The British Empire in India. New Delhi:
Aleph Book Co.
The Hindu (2018). ‘Behind India’s Leap in Ease of Doing Business’. 11 November.
The Nation (2018). ‘SEZs May Change the Fate of Pakistan’. 18 May.
The News (2020). ‘M-5 Reflective of Rapid Progress on CPEC’. 14 July.
The Print (2020). ‘Pakistan PM Imran Khan Vows to Complete CPEC at Any
Cost’. 4 July.
Tilly, R. (1967). ‘The Political Economy of Public Finance and the Industrialisation
of Prussia, 1815–1866’. Journal of Economic History 27 (3): 484–96.
Trading Economics (2019). Pakistan Export Data. https://tradingeconomics.
com/, accessed 29 May 2020.
Transparency International (2019). ‘Corruption Perceptions Index’. Berlin.
https://www.transparency.org/.
Tribune India (2018). ‘Cheaper Cement from Pakistan Shakes Domestic Industry’.
12 September.
United Steel Workers (2017). ‘Chinese Steel Overcapacity: A Legacy of Broken
Promises’. Pittsburgh.
US–China Business Council (2010). ‘Economic Development Policies for Central
and Western China’. Washington, D.C.
Van Leemput, E. (2016). ‘A Passage to India: Quantifying Internal and External
Barriers to Trade’. International Finance Discussion Papers No. 1185, Board
of Governors of the Federal Reserve System, Washington D.C.
Wade, R. H. (2003). ‘What Strategies Are Viable for Developing Countries
Today? The World Trade Organisation and the Shrinking of Development
Space’. Review of International Political Economy 10 (4): 621–44.
———. (1990). Governing the Market: Economic Theory and the Role of Government
in East Asian Industrialisation. Princeton: Princeton University Press.
———. (2012). ‘Return of Industrial Policy?’ International Review of Applied
Economics 26 (2): 223–39.
Wang, J. (2013). ‘The Economic Impact of Special Economic Zones: Evidence
from Chinese Municipalities’. Journal of Development Economics 101 (C):
133–47.
Washington Times (2018). ‘US Balks at IMF Bailout Loans, Fears Debt Trap for
Pakistan’. 8 August.
252
BIBLIOGRAPHY
Wei, X. (2000). ‘Acquisition of Technological Capability through Special Economic

Zones (SEZs): The Case of Shenzhen SEZ’. Industry and Innovation 7 (2):
199–221.
Weiss, L. (2005). ‘Global Governance, National Strategies: How Industrialised
States Make Room to Move under the WTO’. Review of International Political
Economy 12 (5): 723–49.
Weisskoff, R., and E. Wolff (1977). ‘Linkages and Leakages: Industrial Tracking
in an Enclave Economy’. Economic Development and Cultural Change 25 (4):
607–28.
Weyland, K. (1998). ‘From Leviathan to Gulliver? The Decline of the
Developmental State in Brazil’. Governance 11 (1): 51–75.
White, C. M. (1976). ‘The Concept of Social Saving in Theory and Practice’. The
Economic History Review 29 (1): 82–100.
Wizarat, S. (2002). The Rise and Fall of Industrial Productivity in Pakistan. Karachi:
Wolcott, S., and G. Clark (1999). ‘Why Nations Fail: Managerial Decisions and
Performance in Indian Cotton Textiles, 1890–1938’. The Journal of Economic
History 59 (2): 397–423.
Wolf, S. O. (2019). The China–Pakistan Economic Corridor of the Belt and Road
Initiative. Cham: Springer.
Wood, A., and J. Mayer (2011). ‘Has China De-industrialised Other Developing
Countries?’ Review of World Economics 147 (2): 325–50.
Wong, K-Y. (1987). ‘China’s Special Economic Zone Experiment: An Appraisal’.
Human Geography 69 (1): 27–40.
World Bank (2007a). Pakistan—Enterprise survey 2007. Washington, DC.
——— (2007b). Pakistan: Infrastructure Implementation Capacity Assessment
(PIICA). Washington, D.C.
——— (2009a). World Development Report 2009: Reshaping Economic Geography.
Washington, DC.
——— (2009b). Doing Business Report 2009. Washington, D.C.
——— (2014). Doing Business Report 2014. Washington, D.C.
——— (2019a). Doing Business Report 2019. Washington, D.C.
——— (2019b). World Development Indicators. Washington, D.C.
——— (2019c). Karachi Water and Sewerage Services Implementation Project
(KWSSIP): Combined Project Information Documents/ Integrated Safeguards
Datasheet. Washington, DC.
World Economic Forum (2006). The Global Competitiveness Report, 2006–2007.
Geneva: World Economic Forum.
253
BIBLIOGRAPHY
——— (2009). The Global Competitiveness Report, 2006–2007. Geneva: World

Economic Forum.
Economic Forum.
—— (2015). The Global Competitiveness Report, 2015-2016. Geneva: World
Economic Forum.
Economic Forum.
World Justice Project (2019). ‘Rule of Law Index’. Washington, D.C.
Xiaoyang, T. (2015). ‘Assessing the Impact of Chinese Investment on Southeast
Africa’s Cotton: Moving up the Value Chain’. SAIS-CARRI Policy Brief No.
06/2015, Johns Hopkins University.
Xinhua (2018). ‘World’s Largest Textile Mill of Colored Yarns Opens in Xinjiang’.
25 August.
——— (2019). ‘China Focus: Industrial Upgrade Moves Fast in Xinjiang’.
2 March.
Xu, H. (2016). ‘Domestic Railroad Infrastructure and Exports: Evidence from the
Silk Route’. China Economic Review 41 (C): 129–47.
Xu, H., and K. Nakajima (2013). ‘Highways and Development in the Peripheral
Regions of China’. PRIMCED Discussion Paper No. 33, Hitotsubashi
University, Tokyo.
Yafei, H. (2014). ‘China’s Overcapacity Crisis Can Spur Growth through Overseas
Expansion’. South China Morning Post, 7 January.
Yamasaki, J. (2017). ‘Railroads, Technology Adoption and Modern Economic
Development: Evidence from Japan’. Institution of Social and Economic
Research Discussion Paper No. 1000, Osaka University.
Yarns and Fibres (2017). ‘Xinjiang to Introduce 10 year Textile Plan’. 11 September.
Yuan, Y., H. Guo, W. Li, S. Luo, H. Lin, and Y. Yuan (2010). ‘China’s First Special
Economic Zone: The Case of Shenzhen’. In Building Engines of Growth and
Competitiveness in China: Experience with Special Economic Zones and Industrial
Clusters’, edited by D.Z. Zeng. Washington: World Bank.
Yueh, L. (2013). China’s Economic Growth: The Making of an Economic Superpower.
Oxford: Oxford University Press.
Yeung, Y., J. Lee and G. Kee (2009). ‘China’s Special Economic Zones at 30’.
Eurasian Geography and Economics 50 (2): 222–40.
Zafar, A. (2007). ‘The Growing Relationship between China and Sub-Saharan
Africa: Macroeconomic, Trade, Investment, and Aid Links’. The World Bank
Research Observer 22 (1): 103–30.
254
BIBLIOGRAPHY
Zaidi, M. A. (1992). ‘Relative Poverty in Pakistan An Estimation from the

Household Income and Expenditure Survey (1984–85)’. The Pakistan
Zaidi, S. A. (2005). Issues in Pakistan’s Economy, Second Edition Revised and
Expanded. Karachi: Oxford University Press.
Zhai, F. (2018). ‘China’s Belt and Road Initiative: A Preliminary Quantitative
Assessment’. Journal of Asian Economics 55 (C): 84–92.
Zhang, L. (1994). ‘Location-specific Advantages and Manufacturing Direct
Foreign Investment in South China’. World Development 22 (1): 45–53.
Zhang, X. S., K. A. Rashid and A. Ahmed (2014). ‘Escalation of Real Wages
in Bangladesh: Is It the Beginning of Structural Transformation’. World
Development 64 (C): 273–85.
Zhu, J. (1994). ‘Changing Land Policy and its Impact on Local Growth: The
Experience of the Shenzhen Special Economic Zone, China, in the 1980s’.
Urban Studies 31 (10): 1611–23.
Zia, M. M., B. A. Malik and S. Waqar (2017). ‘Special Economic Zones (SEZs):
A Comparative Analysis for CPEC SEZs in Pakistan’. Centre of Excellence
for CPEC Working Paper No. 012, Islamabad.
Zia, M. M., and S. Waqar (2018). ‘Employment Generation and Labour
Composition in CPEC and Related Road Infrastructure Projects’. Centre of
Excellence for CPEC, Islamabad.
255
Index
access to credit, 12, 15, 26, 188, 225 Appalachian Regional Development
Afghanistan, xv, 10, 12, 38, 53, 71, 92, Act (1965), US, 99
152, 158, 207, 219, 224–5 Argentina, 36, 60, 147–8
Africa countries, xvii, 53 Armenia, 112
Chinese SEZs in, 115, 120, 130–43 arms exports, 9
engagement between China and, xvii Asian Development Bank (ADB), 190
Africa Growth and Opportunity Act Karachi Bus Rapid Transit Project,
(AGOA) 2000 210–11
rules of origin, 16 –Pakistan country partnership
US launched, 16 strategy for 2015–19, 211
Afrobarometer data, 138 Asian Infrastructure Investment Bank
agglomeration externalities, 93–4 (AIIB), 76
Agreement on Trade-related Asian Tiger (or Dragon economies),
Investment Measures (TRIMs), 4, 219
172–3 asphalt/bitumen, 68, 74–5, 77, 221
Agricultural Development Bank of Association of South-east Asian
China (ADBC), 214 Nations (ASEAN) countries,
air routes, 106 157–8
Alatau, 82 Atlantic, 19
Albania, 8 ATT/Lucent Technologies, 172
Algeria, 8, 133 Australia, 132, 157–8
cement market, 69 Austria, 65
Allahabad, 103 Austro-Hungarian, 53
All Pakistan Cement Manufacturers automobile industry, 73
Association (APCMA), 72 average hourly wage in China, 122
aluminium, 81, 148 Awami National Party (ANP), 193
Amdahl, 172 Azad Kashmir, 3
Andaman Sea, 217 Azerbaijan, 112
Angola, 130–1, 148, 216
256
INDEX
Babbin, Jed, 209 Bosnia, 54

Bajwa, Asim, 67 Brazil, 77, 132, 147–8, 175–6, 198,
Balkans, 53 216, 220
Balochistan, 6, 52, 56, 92, 120 debt-to-GDP ratio, 34
insurgencies since 1947, 82 railway infrastructure investment, 64
Baltistan University, 3 social savings, 37
Bangladesh, 66, 11–13, 31–2, 66–7, 69, unit savings on freight services, 37
116, 120–4, 155, 170–1, 175, 183 Brazilian Development Bank
Bank of China, 73 (BNDES), 176, 216
Beijing, 8–9, 13, 52–3, 72, 80–2, 85, Britain, xvi, 19–20, 59, 63, 104, 106,
133, 208, 216, 225 132, 166–8, 208
Beijing Summit (2000), 14 building materials, CPEC created
Belarus, 112 robust demand for, 68–77
Belt and Road Initiative (BRI), 1–2, 9, bureaucracy, 108, 179–80, 191, 194
16, 76, 82, 87, 91–3, 209, 211. See Burma, 108
also China–Pakistan Economic business climate, 110
Corridor (CPEC)
brand, 18 California, 21
China’s emergence as net exporter of Cambodia, 84, 120, 123–4
capital, 18 Cameroon, 148
economic corridors, 17 Canada, 59–60, 132, 206
follows market operation, 75 Canada Export Development Bank,
impact of, 112–13 216
map of, 17 Cape of Good Hope, 19
pipeline project capital goods, 38, 83, 175
characteristics of, 211 Caribbean, 21
no sovereign guarantee, 211 cash crops, 51, 63, 95
sovereign guarantee, 211 causality tests for causality (Granger’s),
surplus recycling mechanism, 17 45
system of responsible finance, 210 Cement Manufacturers’ Association,
Bengal, 63, 105 India, 70
Berlin, 94 cement production and capacity in
Bestway Cement, 71 China, 69–70
Bestway Group UK, 71 Central Asia/Asian, 9, 16, 40, 50, 53,
Bethlehem Steel, 132 81–2, 88, 91, 158, 210, 225
bicycles, 99, 122, 154 Central Asian Republics (CARs),
big push, 54 92, 158
Bihar, 100 Central China, 49–50, 53, 106
biotechnology, 85 central government gross debt, 212
Bombay, 19, 104 Chad, 130
257
INDEX
Chambishi MFEZ (Zambia– reduce travel distance from

China Economic and Trade Central China to Middle East,
Cooperation Zone), 134, 136–7, 49–50
139 enthusiasts, 4
Chaudhry, Iftikhar, 203 Eurasia-wide Chinese vision, 1
chemical by-products, 75 falcon economy (see Falcon economy)
Chile, 148, 175 financing of infrastructure through
China Development Bank (CDB), 8, government-to-government
80, 208, 214 loans, 213
China National Oil Corporation focus areas, 5
(CNOC), 15 generate employment and local
China Nonferrous Mining Co. demand in construction,
(CNMC), 134, 139 49, 66–7
China-Pakistan Economic Corridor India and US opposition to, 6
(CPEC), xv–xvi, 23–6, 36, investment in energy projects, 212
39, 43–5, 48, 59, 77, 152, 155, logistics performance index (LPI), 31
160–61, 223–5. See also Belt and major projects of, 1–2
Road Initiative (BRI) net flows, 213
7th Coordination Committee ( JCC) new transport infrastructure to
meeting (2017), 158 reduce travel distance, 31
access to markets and raw materials opponents, 6
from Afghanistan, 38 optimists, 57, 108, 219–20
agriculture projects, 3 and Pakistan Government 83–92,
CBA for, 34 113, 217
characterised by Pakistan building infrastructure in rich in
Government, 2 natural resources, 81
and China’s development since 1976, claim of mutual benefit or win-
13–18 win relationship, 92–103
and crowding in private investment, consensus among stakeholders,
54–9 193–7
culmination of friendship between domestic savings, 78
China and Pakistan, 7–10 generate employment and local
current financing for, 213 demand in construction, 49,
debate by Liberal and Realist 66–7
theorists, 8–9 impact on GDP growth, 78
defined, 1 increase in growth rate of Pakistan,
distributional implications of, 99 78
eastern route of, 47 on jobs creation, 22
economic development in Pakistan management of finances, 207–17
since 1947, 10–13 and need for industrial policy,
and efficient markets 175–93
impact of, 50–1 pessimists, 220
258
INDEX
projected cost, 2 cobalt, 131, 215

promises of, xvii Cold War, 207, 219
reformers, 6 Colon Free Zone, 21
Safe City Project, 4 colonisation, xvi, 20, 209
special economic zones (SEZs), 114, competitive economies, 83–92
117–21 complementary economies, 83–92
spillovers in production, 59–62, 72, Computed General Equilibrium
75–7, 221 (CGE) model, 112
Sukkur–Multan highway project, 68 Confucius Institutes, 133
total investment, 78 Constanza (Black Sea), 19
total value of, 5 consumer goods, 79, 83, 128, 155, 200
transformative impact of, 103–13 coordination failure, 162
transit or economic corridor, 52–4 coordination problems, 56, 162
Western Region Development copper, 106, 130–32, 134, 138–9, 148,
Strategy, 82 155, 157, 215
China–Pakistan Joint Research Centre Corporaction Andina de Formento,
in Earth Sciences, 3 176
China Petroleum Pipelines Bureau corruption, 108, 123, 129, 132, 135,
(CPP), 52 177, 179, 181, 183–6, 191–3,
China Railway First Group Co. Ltd, 75 197, 206
China State Construction Engineering corruption perceptions index (CPI)
Corporation (CSCEC), 75 of Transparency International,
China Tourism Planning Institute, 8, 183–4
208 Costa Rica, 116, 147
Chinese Communist Party, 13 cost-insurance-and-freight (c.i.f.)
Chinese General Administration of basis, 153
Quality Supervision, Inspection cotton yarn, 25, 63, 87, 155–6, 158, 160,
and Quarantine (AQSIQ), 159 171
Chinese Overseas Port Holdings Ltd., Country Policy and Institutional
52 Assessment (CPIA), 184–5, 193
Chittagong (Bangladesh), 9 Covid-19 outbreak, 222
Chongqing, 79, 107 credit subsidies, 166
Chongqing–Huaihua Railway, 80 Croatia, 175
circular debt, 5, 111 Cuba, 148
civil society groups/organisations, 8, 20,
179–80, 195, 202, 205, 208, 216 Dawn newspaper, 8
coal bases Dawood, Abdul Razak, 204, 215
Aiver Valley, 84 debt bailouts, 214
Hami, 84 debt dependency/dependent, 6, 25,
Urumqi, 84 217, 220, 223
coal-fired power stations, 3, 62 debt-distress, 212
coastal shipping, 37 debt-for-equity swap, 214
259
INDEX
debt-forgiveness, 131, 207 108, 111, 113, 115, 122, 131–2,

debt, government, 34, 194, 206, 212 134, 150–1, 163–4, 168–9,
debt (total) in China, rise in, 34 178–9, 187, 193, 197, 200, 202–
debt market, 188 3, 218–21, 223, 225
debt relief, 14, 131, 214 economic theory, 60, 93, 99, 161, 217
debt-to-GDP ratio (total) of China vs Ecuador, 148
US and Brazil, 34 efficient markets, impact of, 49–51
debt-trap-diplomacy, 209 Egypt, xvi, 19, 21, 53, 116, 133,
Deccan, 50 137–8
Democratic Republic of the Congo exports of commodities, 20
(DRC), 130–1, 142, 148, 215–16 industrialisation of economy, 20
democratisation, 20 Egypt Suez Economic and Trade
denim, 156–7 Cooperation Zone, 136
Denmark, 51 electrical machinery and equipment,
development finance institutions 155, 169–70
(DFIs), 176 electrolytic industry of China, 75
difference-in-difference approach, 40 employment, 23, 34, 40–1, 48–9, 54,
Digital, 172 66–7, 72, 77, 82–3, 85, 87, 95, 98,
Directorate General of Foreign Trade, 100–2, 111–13, 115, 118, 122,
India, 70 128, 130, 135, 137–8, 141, 144,
distributional effects of infrastructure, 146, 149–51, 167–8, 170, 177,
96 215, 220–2, 224
Djibouti, 134, 212 endogeneity of infrastructure, 45–7
Doab, 95 engineering, 20, 59–60, 63–5, 86
domestic capital markets, 55 enterprises in China, new role for, 89
Dominican Republic, 115, 148 Equatorial Guinea, 51
Donets Steel Company, 167 Ericsson, 172
doubling roads in China after 1990, 30 Ethiopia, 133–41, 144, 160, 177, 217–
dyeing, 88 18, 222–4
dynamic economies of scale, 163 developmental state, 168
Development Bank of Ethiopia, 176
East Asian countries, 115–16, 166, 178 domestic footwear firms in, 149
East Coast Rail Link project, 216 emerged as net exporter of shoes,
eastern coastal states, 79 149–50
East Turkestan Islamic Movement, 9 exports of animal feed by, 83
economic agglomeration, 93 FDI projects in, 168
economic corridor, 52–4 flooded by Chinese shoe imports,
economic growth and development, xv– 149
xvi, 5, 10–11, 19, 23, 40–2, 44–7, Growth and Transformation Plan
49–50, 54, 56–7, 60, 62–4, 76–7, (2010–15), 168
79, 82–6, 93, 95, 99–100, 104–5, growth of floriculture, 168
260
INDEX
Plan for Accelerated and Sustained fiscal decentralisation, 85

Development to End Poverty fiscal deficit, 189, 203
(2005–10), 168 fiscal resources, 180
rural road construction in, 98–9 fiscal stimulus, 72
Ethiopian Eastern Industrial Zone fiscal transfers, 96
(EIZ), 134–7, 139–40 Five-Year Plan of China
Eurasian supercontinent, xv, 91, 210 11th (2006), 132
European Union (EU), 41, 91, 98, 101, 12th (2011–15), 74, 79
146, 149, 151, 210 flat glass sector, China’s capacity in, 74
exchange rate, 110 FLSmidth, 71
depreciation, 175 Fluor, 132
exclusive infrastructure–commodities Flying Cement Company, 71
relationship, 132 Flying Geese model of economic
Export–Import Bank of China (China development, 84
Eximbank), 15, 214 food crops, 50–1
export-led growth, xv, 17 food industries, 20
export processing zones (EPZs), 119, foreign aid, 45, 192, 207, 219
121, 129, 145. See also China– foreign direct investment (FDI),
Pakistan Economic Corridor xv–xvii, 24, 61, 114–15, 120, 132,
(CPEC); special economic 134, 145, 147, 165, 172– 3, 189,
zones (SEZs) 207. See also China–Pakistan
access to good infrastructure, 116 Economic Corridor (CPEC)
benefits and costs of, 114–17 China/Chinese, 25, 79, 84, 127, 129–
defined, 114 30, 137–8, 141–4, 149–50, 157,
global, 116 160, 168, 213, 215, 222
export sophistication, 163 inward, 18, 121–2
Express Tribune, 193 outward, 18
external benefits, 56 lead to technology transfer to local
firms, 171
face-to-face meetings, 45 in Pakistan, 5, 91, 111, 120, 174–5,
Faisalabad, 67, 120 212
falcon economy, xv, xvii, 4–7, 25–6, 113, declining trend and Chinese
219–21, 223 investment through CPEC,
famines, 50–1, 104 xv, 92
Federally Administered Tribal Areas in textile sector, 87
(FATA), 118 Forum on China–Africa Cooperation
fertiliser, 3, 20, 75, 84, 89, 154–5, 162 (FOCAC), 14
imports from China to Pakistan, 85 France, 59, 167, 179, 197
fibre optics network, 1, 3, 61 Concorde supersonic airplane
Financial Times, 215 in, 35
First East Turkestan Republic, 80 free-on-board basis, 153
261
INDEX
free-trade agreement (FTA), 16, 75, Green programme in northern

114, 166, 172, 182 Thailand, 35
China and Australia in 2015, 157 Green Revolution, 89
China and Pakistan in 2006, 7, 25, gross capital formation of Pakistan vs
111–12, 145, 152, 155–7, 159– India, 58
60, 218, 222, 224 Guizhou province, 79
Gujarat, 100
Gabon, 148–9 Gujranwala, 120
Gansu province, 79 Gulf region, xv, 207
GDP growth of Pakistan vs world, 11– Gwadar Port, xv, 1–3, 7, 9, 38, 47, 49,
12, 26, 57–8, 66 50–4, 56, 62, 77, 92, 99, 107,
GDP growth stability analysis, 13 121–2, 144, 191, 207–8, 222
General Musharraf, 201–3, 218, 223
Georgia, 112 Hambantota (Sri Lanka) port, 9, 35,
German Thyssen Company, 132 194, 208–9, 214, 217
Germany, 12, 17–18, 53, 59, 71, 77, Han Chinese, 81
87, 144, 160, 152, 177, 194, 197, hardware sector, 62, 130
217, 220–1, 223–4 heavy transport equipment, 155
beginning of railroad construction, 65 Hebei Iron and Steel Group, 76
engineering and iron industries, 65 Hebei province, 91
import substitution process, 65 Hebei Tianzhu Iron and Steel Group,
industrialisation, 94 China, 73–4
iron production, 65 Henan Xinye Textile Co., 87
pig iron and bar iron production, 65 higher education, 102, 166, 175
promotion of infant industries, 166–7 high-speed railways (HSR), 33, 45,
Zollverein (customs union) followed 97–8, 104, 193
free trade, 166 in China and rise of GDP, 40
Ghana, 130, 132, 134, 143, 148, 208 substitute for air travel, 40
Chinese FDI, 137, 141 high-technology development zones
crackdown on Chinese gold miners, (HTDZs), 121
216–17 high-technology manufacturing, 100
Gilgit-Baltistan, 3, 6–7, 53, 89, 118–19, high-yield variety seeds, 89, 165
159 Hong Kong, 4, 10–11, 84, 116, 122,
glass, 63, 74, 77, 85, 91, 170, 221 127, 146, 219
global financial crisis of 2008, 72, 131, Honshu, 95
139 Houston, 51
global positioning system (GPS), 61 Huaneng Shandong Ruyi (Pak) Energy
Going Out Strategy for Chinese (HSR), 193
enterprises, 76, 143 Huawei, 15, 117
good governance, 142, 172 Hubco and Engro projects, 213
Great Power, 22 Hungary, 54
Greek Olympic games (2008), 35 Hyderabad, 101
262
INDEX
IBM, 172 Indonesia, 10–11, 66, 73, 84, 116, 123,

income tax, 85, 109, 134, 194 171, 175, 183
India, 6, 10–13, 19, 28, 39, 56–8, 60, cement production overcapacity in,
65, 68–9, 92, 112, 155, 163, 165, 69
168, 175, 206, 220 industrial capitalist-entrepreneurs
advantage by opening of Suez Canal, class, 176
106 Industrial Development Bank of
agricultural production marketed Pakistan (IDBP), 176
over long or medium distances, industrial FDI of China, 84
10–13, 19, 28, 39, 56–8, 60, 65, industrial growth, 39, 54, 118, 160, 163,
68–9, 92 222
average hourly wage in, 122 industrialisation, xvii, 13–14, 20, 25,
cotton textile industry in, 63 38–9, 47, 63, 73, 76–7, 81, 85,
cultivation shifted to high yield areas, 88, 92, 94–6, 113, 115, 118, 144,
95 161–62, 166–67, 170, 172–73,
cultural prohibitions on marriage, 176, 194, 220–22, 224–25. See
100 also China–Pakistan Economic
dam construction in, 97 Corridor (CPEC); special
immobility of labour, 100 economic zones (SEZs)
migration in, 101 industrial policy, 25, 140, 143–4, 160,
population of poor landlocked states, 170–1, 223
100 and development of infrastructure,
Pradhan Mantri Gram Sadak Yojana 166–9
(PMGSY) (or Village Road and historical experience, 166–9
Programme), 97 importance of, 161
price levels, variations in, 104 and market failures, 161–5
railways, important role of, 104–5 industrial policy in Pakistan, 217–8
road transport in and CPEC, 175–93, 200
congestions, 28 governance and future of, 190–3
freight movement, 28 need for, 169–75
Golden Quadrilateral (GQ) infrastructure development projects. See
project, 29, 106 also China–Pakistan Economic
highway system, 28 Corridor (CPEC)
SEZ Act in 2005, 116 cost–benefit analysis (CBA)
software-IT sector in, 61–2 improvement in safety standards,
total textile industry operator costs, 27
123 operating costs, 27
Indian National Congress, first meeting time savings, 27
in Bombay in 1885, 63 travel costs, 27
Indian Ocean, 9, 19, 52 weakness with, 33
indigo, 63, 95 expansion of telephone landline, 42
indium tin oxide (ITO), 74 impact of over-investment
263
INDEX
crowding out effect non- Iran, xv, 71, 92, 108, 152, 158, 218, 224
infrastructure, 42 –Pakistan gas pipeline, 52
diminishing returns, 42 Iraq, US invasion in 2003, 202
international airport and Ireland, 115, 172–3
international cricket stadium iron ore, 36, 64, 131, 148, 155, 157
by China, 35 irrigation networks, 55
Kenyan municipalities, impact on, 42 Islamabad, 3, 39, 51, 68, 71, 93, 118,
legacy of macroeconomic risks 201, 215
corporate debt, 34 Islamabad Chamber of Commerce and
non-performing loans, 34 Industry (ICCI), 201
long-term macroeconomic studies of Islamabad Red Mosque, 9
agriculture, 39 Islamic Party, 81
urbanisation, 39 Italy, 60, 93, 196–7
railroad to connect Urumqi and efforts to promote infrastructure
Lanzhou (China), 29 activities, 96
railway construction, impact of, 42
rehabilitation of roads in Colombo Jamiat-e-Ulema Islam-Fazal ( JUI-F),
and Kandy, benefits of 193
capital-intensive methods of Japan, 10, 17–18, 41, 59–60, 84, 95,
production, 28 132, 152, 163, 171–2, 179,
firm-level surveys, 28 216–17, 219
labour-intensive firms, 28 Japan International Cooperation
Spain, impact of highway Agency ( JICA), 134
construction in, 29 Jeddah (Saudi Arabia), 19
wages of skilled vs unskilled labour, Jinping, Xi, 7, 17
29 joint ventures, 13, 15, 52, 91, 117, 121,
infrastructure, global history of, 19–22 127–8, 131, 136, 142–3, 171–2,
Infrastructure Project Development 177, 189
Facility (IPDF), Pakistan, 191–2 Jordan, 116
intellectual property rights, 128, 164 jute, 106
international airport, 21, 35, 135, 194 handloom weaving, 63
International Labour Organization
(ILO), 123 Karachi, 3, 56, 67, 92, 101, 125, 175,
International Monetary Fund (IMF), 201
xv, 5, 8, 14, 16, 20, 45, 110, 131, Karakoram Highway, 7, 19, 38, 107,
172, 189–90, 208–9, 213–14 159
Pakistan’s per capita annual water Karakoram Mountains, 225
availability, 111 Karakoram range, 107
International Olympics Committee Kashgar, 1–2, 7, 82, 88, 107, 127, 159
(IOC), 34 Kashmir, 63, 118, 201
investment security, xvii Kazakhstan, 17, 81, 90, 108, 158
Iqbal, Ahsan, 78 Kenya, 42, 130, 146, 195
264
INDEX
Keqiang, Li, 7 defined, 30

Khan, Imran, 204–5, 215 features of, 30
kharif harvests, 103 use of IT-technology for, 31
Khorgas, 82 long-term lending for industry, 55
Khunjerab Pass, 7, 107, 225
Khyber Pakhtunkhwa (KPK), 3, 89, 92, Macaes, Bruno, xv, 87, 91, 210
120, 205 Macao, 127
Rashakai Economic Zone in, 119 macroeconomic stability, 161
Kolkata (earlier Calcutta), 28–9, 103 macroeconomic studies, 39–43
Korean War (1952), 7, 152 impressiveness of, 43–5
Korla Economic and Technological problem of endogeneity and its
Development Zone, 87 solutions, 45–7
Kunming, 82, 104 Madagascar, 130, 146
Kyrgyz Republic, 212 Madhya Pradesh, 100
Kyrgyzstan, 90, 158 Malacca route, 49, 52, 217
Malaysia, 10, 31–2, 66, 73, 84, 112, 116,
labour, 16, 29, 45, 86, 94–5, 98, 100–1, 123, 152, 163, 165, 171, 216
108, 118, 122–3, 128–30, 132, Maldives, 212
134, 138, 145, 153, 162–3, 169– managers, 16, 64, 67, 117, 141, 143,
70, 172, 175–7, 182, 200, 202, 162, 165, 170, 172, 201
209 Mao, 1, 13
employment opportunities through Maoist, 79
CPEC, 67 Maple Leaf Cement Factory, 71–2
-intensive industries/firms, 15, 28, marginal infrastructure, 76
84–5, 116–17, 143, 146–7 maritime cargoes, 106
-intensive technology, 164 market-determined supply, of skilled
requirement of, 66 workers, 162
skilled, 29, 61, 66, 79, 93, 110, 174, market failures, 25, 56, 59, 161–5, 205,
189 217
unskilled, 64, 66–7, 141 in public investment, 55
Lahore, 3, 39, 67, 71, 92–3, 101, 201 market imperfections, 140, 162, 165
landlords (private), 201 Marshall, Alfred, 93
Laos, 84, 212, 225 Mauritius, 15, 116, 130, 133, 138, 146
Latin America, 22, 88, 149, 155, 160, Mediterranean, 19
166, 222 Medium-Term Development
growth of Chinese trade, 147–8 Framework (MTDF), 34, 59,
Lawyers’ Movement, 183, 203 68, 187
leading sector, concept of, 23, 49, 59–60 mergers and acquisitions (M&As),
Lesotho, 130, 146 74, 76
Lin, J. Y., 165 Mexico, 16, 77, 94–5, 100, 105, 116,
Liverpool, 19 147–8, 175, 220
logistics, 20, 32, 36, 89 average hourly wage in, 122
265
INDEX
Central Railway, 64 Muttahida Majlis-e-Amal (MMA),

depends on overland transportation, 202
37 Myanmar, 210, 217
social savings, 37
tax revenue from taxes on exports, 64 Nasser, 20
Middle East, xv, 9, 15–16, 19, 38, 47, National Development and Reform
49–50, 52–3, 91, 106–7, 210, Commission (NDRC), China,
219, 225 7, 15, 191, 208
migration, 15, 46, 67, 81, 93, 96–7, 113, National Energy Administration,
129, 221 China, 8, 208
forced, 77 National Highway Authority (NHA),
impact in EU, 101 191
international, 62 National Highways Development
inward, 94 Project (NHDP), 28
long-term, 100 National Trunk Highway System
in Pakistan, 101–3 (NTHS), China, 97
permanent, 100 neoliberalism economics, 161
military elite, 9, 55 Netherlands, 168
Millennium Development Goals new Chinese west and American west,
(MDGs), 14 comparison between, 80
minerals, 3, 53–4, 64, 85, 115, 131, 133, New Silk Road, 1, 219
146, 148, 159, 170 New Zealand, 157–8
mining, 3, 134, 137, 139–40, 149, 215 nickel, 80, 130–1, 148
Ministry of Foreign Affairs, China, Nigeria, 131, 133, 139, 148
76, 131 Guangdong economic trade
Ministry of Railways, China, 39–40 cooperation zone, 135
Ministry of Transport, China, 208 Ogun Zone, 136–7
mobile phone penetration, 154 Nigeria Ogun–Guangdong Free Trade
modern (just-in-time) Zone, 135–6
manufacturing, 30 Ningxia, 79
Moldova, 112 non-abiding firms, 166
Mongolia, 212, 214 non-tariff barriers, 159
monopoly, 61, 129, 161, 164, 193, 198 North Korea, 8, 219
Montenegro, 212 North-West Frontier Province
Motorola, 117 (NWFP), 50, 202
Mubarak, 20 famine in, 50
Multan portion, 67
multinationals (MNCs), 14–15, 17, 61, oil exports, 131, 148
115, 117, 137, 141, 173–4 oil seeds, 106
Munchkin, Steven, 208 Okara, 202
Muslim/Turkik rebellion in southwest developmental repression in, 201
China, 80 Olympic Press Centre, 35
266
INDEX
opium, 63, 106 government intervention in financing

Orange Line urban train project, infrastructure, necessity of,
Lahore, 203 205–7
organic chemicals, 85 household final consumption
Orissa, crops failure in 1865–6, 50 expenditure as share of GDP,
Ottoman Empires, 53 203–4
Oudh, 95 hyper-politicised patronage society,
204
Pacific Northwest, 21 industry policy in, need for
Pakhtunkhwa Milli Awami Party and CPEC, 175–90
(PKMAP), 193 exports of readymade garment and
Pakistan, 165, 168–9. See also Belt and cotton yarn, 171
Road Initiative (BRI); China– joint venture with Japanese firm,
Pakistan Economic Corridor 171–2
(CPEC) labour productivity, 169–70
18th Amendment, 187 manufacturing sector, to increase
Automotive Development Policy share of, 171
(2016–21) of Ministry for need to impart vocational
Industries and Production, 190 education, 175
average hourly wage in, 122 opportunities to leverage FDI, 174
bazar traders of, 200 slow rate of structural change, 170
budget deficit in, 110 technological upgrading, 169
cement industry of, 71–2 textile exports, 171
Central Bank, 191 total factor productivity (TFP),
conflict in, 197–205 169–70
construction sector in, 66 infrastructure in, xv
Country Policy and Institutional largest salt reserves, 155
Assessment (CPIA) on macroeconomic management in, 109
present policy and institutional manufacturing employment, 150
framework, 184–5 Ministry of Inter-Provincial
coup in 1999, 202 Coordination, 192
domestic/foreign banks, 191 mobilisation of tax revenue in, 197–
economy, characteristics of, 25 205
education system failure in, 66 motorcycle industry, 154
energy shortages in, 5 nuclear and ballistic missile
engineers, shortage of, 66–7 programme of, 9
export base, 151 Planning Commission, 191
exports by, 12 policymaking in, 218
Federal Interior Ministry, 192 power sector in, 59, 110–11
garments exports by, 151 Programme Co-ordination Unit
government consumption and tax (PCU) of Ministry of Finance,
revenue in, 199 191
267
INDEX
rise in fiscal deficit, 189 Pakistan Industrial Development

savings and investment in, 199 Corporation (PIDC), 70
state and society, relationship Pakistan International Airlines (PIA),
between, 180 110
stock market, 191 Pakistan Labour Force Survey, 86, 88
sustainable social development in, 22 1996–7, 102
tariff comparison of exports to China, 2004–5, 102
157–8 2010–11, 102
tax revenue as a share of GDP 2014, 200
(1970–2017), 110 Pakistan Muslim League (Nawaz) (or
Textiles Policy (2009–14 and 2014– PML(N)), 4, 200, 203
19) of Ministry of Textiles, Pakistan Muslim League (Quaid e
189–90 Azam Group) (PML[Q]), 202
total textile industry operator costs Pakistan Peoples Party, 4
in, 123 Pakistan Planning Commission,
trade liberalisation in, 109, 111, projection on urban citizens
150–1 population, 68
upsurge in economic growth, 202 Pakistan’s Ministry of National Food
visa problem for transporters, 159 and Security (MNFSR), 159
vs world Pakistan’s Statutory Regulatory Order
corruption perceptions index Regime, 154
(CPI) of Transparency Pakistan Steel Mills, 72–3
International, 183–4 Pakistan Stock Exchange (PSE), 71
real wage growth, 123 Pakistan Tehreek-e-Insaf (PTI), 4, 193,
water shortages in, 111 200, 204–5
World Bank Doing Business ranking, Panama Canal, 19, 21–2, 53
181–2 Panasonic, 15
World Justice Project rule of law Paraguay, 147–8
index, 186–7 Paris Club, 214–15
zero tariffs on denim and surgical Pearl River deltas, 81
goods, 156–7 Perot, Ross, 16
Pakistan Administrative Services Persian Gulf, 9
(PAS), 179–80 Peru, 147–8
Pakistan Association of Large Steel Peshawar, 3–4, 67, 89, 92, 119, 201
Producers (PALSP), 73 pesticides, 89
Pakistan Business Council, 159, 176 petrochemical industry, 3, 84–5, 162
Pakistan–China Joint Cooperation pharmaceutical industry, 86, 112, 135,
Committee, 78 170
Pakistan Industrial Credit and pharmaceutical products, 154
Investment Corporation Philippines, 116, 120, 175
(PICIC), 176 pick winners, 165
268
INDEX
political instability, 110, 113, 221 107–8, 112–13, 119, 144, 154,
political school, 178–9 167, 194, 196, 205–6, 208, 213,
politics of infrastructure, in developed 216–17, 220–1
countries, 196 construction, 42, 105
Pompeo, Mike, 208 contributions of, 60
population density, 41, 97 equipment, 106, 167
Port Qasim, 3 investment in China, increase in,
Port Qasim Electric Power Company 97–8
Limited (PQEPCL), 193 rainfall, 43, 98, 105
Port Said, 19 Rajapaksa International Cricket
Port Tawfiq, 19 Stadium, 194
postal system, 63 Rajapaksa, Mahinda, 35, 194
poverty, 13–14, 81, 93, 98, 132, 138, Rajasthan, 100
168, 184, 204, 218, 221, 224 Rashakai Economic Zone, 89–90
PPP, 202–3 Rawalpindi, 201
Private Power and Infrastructure Board raw materials, 12, 14, 16, 30, 38, 45, 53,
(PPIB), Pakistan, 191–2 60, 79, 81, 83, 88, 111, 116, 120,
product diversification, 163, 189 133, 142, 149, 154–6, 160, 166,
productivity, 23, 25, 29–30, 34, 42, 48, 172, 206, 219, 222, 224
112, 141, 143, 153, 163–4, 169– readymade garments, 20, 157, 171
70, 175–7, 182, 189, 217, 220, Red Sea, 19
223 regional equality, 79
property rights, 109, 161, 181 remittance income, 20, 53, 111
Provincial Investment Promotion renewable energy, 85, 213
Authority (PIPA), Pakistan, 117 research and development (R&D)
Prussia, 46, 194 laboratories, 15
public subsidies, 165 Rostowian political revolution, 55
Puerto Rico, 115 Rostow, Walt, 54–5, 59–60, 197
Punjab, 3, 7, 47, 56, 89, 93, 119–20, royalties (or profits), 60, 172
179–80, 200–3, 205 Russia, xv, 29, 38, 52, 59, 87, 104–6,
108, 158, 177, 217, 223
Qaumi Watan Party-Sherpao, 193 demand for track materials, 167
Qing dynasty, collapse in 1911, 80 ferrous metals, 167
Qinghai province, 79 locomotives, 167
Qinghai–Tibet Railway, 80
Quetta, 3, 92 Sadat, 20
sales tax, 154, 189–90
rabi harvests, 103 Samsung, 15
race to the bottom, 173 Sao Paulo, 64
railways, xv, 1, 3, 29–30, 33, 36–8, 40–1, Saudi Arabia, 19, 152, 168, 218, 224
43, 46–7, 50–1, 53–4, 56, 59, Saxony, 94
63–5, 77, 81, 91, 93–5, 100, 104, scramble for Africa, 14, 130
269
INDEX
self-sustained economic growth, 54 Uruguay, 36–7

Serbia, 54 US, 36
SEZ Act (2012), Pakistan, 117–18, 128 software firms, 62
Shaanxi province, 79 Sokhana Port, 136
Shanghai, 13, 49–50, 81, 104, 107, 125 solar energy, 85
Shannon Airport, Ireland, 115 Somalia, 168
Shantou SEZ, 121 South Africa, 73, 76, 130, 140, 146, 175
Sharif, Shahbaz, 200 South America, 10, 21
shattered economy, 25–6, 224 South Asia/Asian, 8–9, 11–12, 105
shawl manufacture, in Srinagar, 63 South China University of Technology,
Shenzhen SEZ, 24, 101–2, 117, 121–3, 135
127–9 Southeast Asia, 87, 143, 210
Shihezi Economic and Technological South Korea, 4, 10–11, 84, 116–17, 163,
Development Zone, 85 166, 168–70, 172, 178, 198, 219
Shiv Sena party, Mumbai, 101 Spain, 29, 41–2, 196–7
Sialkot, 93, 120 special economic zones (SEZs), xvii,
Sichuan province, 79 1, 3, 54, 82, 85, 114, 144, 218,
Siemens Nixdorf, 172 222. See also China–Pakistan
Silicon Valley, 62 Economic Corridor (CPEC);
silk of Bengal, 63 export processing zones (EPZs)
silk processing, 85 African, 15
Silk Road Fund, 76, 210, 219 benefits and costs of, 114–17
Sindh, 3, 56, 89, 102, 119–20, 203, 211 China, 117
Singapore, 84, 116–17, 163, 198 in African countries, 130–43
Economic Planning Board (EPB), and reforms after 1979, 121–9
172, 191 and CPEC in Pakistan, 117–21
Jurong Town Corporation, 173 defined, 114
Sino-Pakistan Hybrid Rice in East Asia, 116
Research Centre, Karachi global, 115
University, 89 key themes of, 16
skills in demand, in China, 40 optimistic view of, 115
slowdown in industrial and GDP origin of, 115
growth after 1973, 163 purpose of, 114
Social Action Programme (SAP), 22 role of, 115
social saving, 27 from Shenzhen to Africa, 24
defined, 36 single-window clearance, 77
of railways and road transport special trade zones, 82
Argentina, 36 spillovers (production), 3, 49, 51–2, 56,
Brazil, 37 59–62, 77, 116–17, 161, 220
India, 38 from infrastructure, historical and
Mexico, 37 contemporary evidence of,
Russia, 38 62–5
270
INDEX
Sri Lanka, 11–13, 28, 31–2, 35–6, 116, sustainable economic development, xvi,
121, 123, 183, 208, 214, 217 22, 44
bailout package used to repay loans sustainable goodwill, 131
from China, 209 Swaziland, 146
political favouritism in Chinese Sweden, 59, 196
lending, 194 Swedish pulp industry, 59
state capacity, 25, 109, 181, 197, 218,
223 Taiwan, 4, 10, 14, 116, 162–3, 166, 168,
State Council, China, 15, 76, 91 219
State Council Circular, 2010, 74 Tajikistan, 90, 158, 212
State Council Guidance on Excess take-off growth, 54–5, 59
Capacity (2013), 74 Taklamakan Desert oil exploration
state-owned enterprises, 13, 34, 131, project, 82
141, 211 Tamil Nadu, 39, 100
steel, 3, 60, 64, 68, 76–7, 83, 86, 90–1, Tanzania, 148
106, 119, 135, 137–9, 146, 148–9, Tanzania–Zambia railway, 16
154, 157, 162, 167, 170, 221 tariff(s), 24, 107, 110, 132–3, 145, 150,
consumption and production 152–3, 155–6, 158, 166–7, 175
in China, 73–4 concessions, 157, 159
in Pakistan, 72–3 consumer, 6
Steel Industry Revitalisation Plan, effective, 154
2009, China, 74 exemptions on imported industrial
Straits of Malacca, 52, 217 equipment, 80
strategic sectors, 15 Tarim Basin, 82
String of Pearls, 9 Tashkurgan, 159
structural change, 25, 41, 161, 164–5, tax incentives, 166, 172
170, 175, 188, 190, 217, 223–4 tax mobilisation process, 109–10
sub-Saharan Africa (SSA), 10, 130, tax rates, 108
141, 148 Technical and Vocational Education
decline in clothing exports to US, and Training (TVET) Centre,
146 Addis Ababa, 140
ethnic favouritism in, 195 technological transfer, 164
infant mortality in, 195 technological upgrading, 25, 169, 173,
subsidy(ies), 35, 71, 86, 118, 165–7, 175–6, 190, 217, 223
173, 178, 188, 194, 203, 205–6 telecommunications, 3, 42, 61, 81, 84,
subsistence agriculture, 63, 94 117, 172–3, 177, 198
Sudan, 148, 168 telegraph network, 104
Suez Canal, xvi, 19, 20–2, 53, 104, 106 textile industry/sector, 3, 12, 15–16, 19,
Economic Area, 136 21, 26, 28, 55, 59–60, 63, 85, 87–8,
opened in 1867, 63 90, 95, 106, 112–13, 118, 120,
Sukkur, 3, 67 136, 138, 143, 146, 151, 155–7,
271
INDEX
159, 163, 177, 189–90, 206, 215, United Provinces, 95, 104
218, 221, 223, 225 United States (US), xv, 6, 12, 18, 59–60,
of Bangladesh, 67 152, 190, 219, 224
operator costs, 122–3 city-centre population of urban areas,
of Pakistan, 86, 169–71, 175 declining of, 96
Thailand, 10–11, 35, 66, 73, 84, 112, construction of urban transportation
123, 149, 156, 169–71, 183 infrastructure, 96
thread, of synthetic fibre, 154 debt-to-GDP ratio, 34
Tianjin Economic-Technological Federal Aid Highway Act of
Development Area (TEDA) 1944, 96
zone, 117 highway construction, 96–7
Toshiba, 15 internal migration in, 101
tourism, 35, 21, 128, 132, 152, 165 international sanctions against
receipts, 20 Iran, 52
trade policy between Pakistan and national highway system, 96
China, 24–5, 143 oil and gas companies, investment
trade unions, 16, 137
by, 51
trade war between US and China,
roads construction in Appalachian
159–60
region, 99
traditional manufacturing, 29, 85
social saving method, 36
transit corridor, 23, 52–4, 63, 77, 81
Union Pacific Railways, 205
T-shirt making costs in Pakistan vs
war on terror in Afghanistan, 207
world, 123–6
Uruguay Round of the General
Turkey, 155, 158, 168
Turkmenistan, 67, 158 Agreement on Tariffs and Trade
TV sets, 154 (GATT)-WTO trade, 172–3
Urumqi Economic and Technological
UAE, 66, 152 Development Zone, 85
Uganda, 130 Urumqi Hi-Tech Industrial
cement plants, 69 Development Zone, 85
Uighur militants, 9 Urumqi, riots in, 82
Ukraine, 52, 112 US Council on Foreign Relations, 131
under-industrialisation, 162 Uttar Pradesh, 100
unemployment, 96, 100, 196 Uzbekistan, 67, 158
Union of Soviet Socialist Republics
(USSR), 8, 81, 198, 219 value-added tax (VAT), 85, 126, 134
United Kingdom (UK), 12, 18, 35, 53, Venezuela, 148
63, 152, 174, 224 Vietnam, 4, 8, 10, 84, 110, 112, 121,
iron and shipbuilding industry, 20 123, 156, 165, 183
United Nations (UN) Millennial EPZs in, 119
Summit (New York), 13 rural road construction in, 98–9
272
INDEX
wages, 37, 73, 84, 86, 94, 98, 115, 122, logistics performance index (LPI),
129, 138, 153, 203 Pakistan vs world, 31–2
differentials, 100 pessimism and threats of cheap
flexible, 101 imports, 70
inequality, 100 quality of trade- and transport-
in Pakistan vs world, 123 related infrastructure, 32
of skilled labour, 29 on steel, 72
of unskilled labour, 29 World Economic Forum
Wakhan corridor, 158 Global Competitiveness Report of
weather shocks, 10 2017–18, survey in Pakistan,
welfare payments, 96, 99 108–9, 182–3
West Bengal, 39 world trade, China’s impact on
West–East Pipeline, 81 China and Pakistan trade
Western China, xvi, xvii, 6–7, 10, 24, negotiations, 151–60
38, 45–7, 49–50, 52–3, 74, 94, China’s demand for imports, 148–50
107, 113, 145, 219, 221, 224 comparative advantage, 145
complementary advantage and growth of trade with Latin America,
importance of, 79–92 147
Western Corridor Gas Infrastructure increase in exports of manufactures,
Development Project, 216 145–6
Western Development Programme World Trade Organization (WTO),
(WDP) (or Open Up the West 115, 143, 147, 172
campaign), 79–82, 84 World War II, 81
western provinces, 79
West Pakistan, 70 Xiamen, 121
white elephants infrastructure projects, Xi’an–Nanjing Railway, 80
34–6, 48, 60, 220 Xiaoping, Deng, 13, 79, 115
wide tax system, 85 Xinguang International Group, 135
wind energy, 85 Xinjiang Hoxud Economic Zone, 85
win-win situation, xvi, 76, 84, 92, 94, Xinjiang Huocheng Economic
113, 119, 131–2, 209, 218, 221 Development Zone, 85
Woody Islands (Paracel Islands), 9 Xinjiang–Kazakhstan border, 81
World Bank, 55, 59, 74, 76, 129, 142, Xinjiang Kuytun Economic
172, 181, 190, 192–3, 198, Development Zone, 85
210, 216 Xinjiang Miquan Industrial Park, 85
Enterprise Surveys for India, 30 Xinjiang Production and Construction
estimation on Pakistan domestic Corps, 89
cement capacity, 68–9 Xinjiang province, xvi, 3, 7, 9–10, 40,
infrastructure investment, 72 52–3, 79, 83–8, 93–4, 113, 127,
Karachi Water and Sewerage Services 221, 224–5
Improvement Project, 211 agriculture in, 90
273
INDEX
biggest producer of tomatoes, 90 Yafei, He, 76

incorporated into Qing Empire, 80 Yangtze delta, 81
major cotton crop producer, 90 yarn, 25, 63, 87, 106, 154–6, 158–60,
and Pakistan, complementary or 171, 222
competitive economies, 83–92 Yucatan Peninsula, 94
political concerns in, 81
Yunnan province, 79
population of, 80
poverty among Muslim Uyghurs, Zambia, 15–16, 130, 136–8, 140
81–2
copper mining by China in, 139
radical Islam spillover into, 10
Multi-Facility Economic Zone
resource extraction from, 82
rich in natural resources, 80 (MFEZ) policy, 133–4
total exports from, 90–1 rural road construction in, 98–9
Uighur ethnic riots in, 10 Zemin, Jiang, 13, 79
WDP presence in, 81 Zenawi, Meles, 168
Xizang [Tibet], 79 Zhuhai SEZ, 121
274

The Dragon From The Mountains - The CPEC From Kashgar To Gwadar

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Dragon From The Mountains - The CPEC From Kashgar To Gwadar

Uploaded by

Copyright:

Available Formats

Matthew McCartney is Associate Professor in the Political Economy and Human

Cambridge University Press is part of the University of Cambridge.

ISBN 978-1-108-83415-5 Hardback

List of Maps and Figures ix

2. Big Infrastructure: Big Problems or Big Benefits? 27

3. CPEC Spillovers Rippling Outwards 49

4. Through the Eyes of Who? Evaluating the Success of the CPEC 78

5. The Dragon Uncoils: Special Economic Zones (SEZs) from

6. The Dragon’s Embrace: Pakistan–China Trade Policy 145

7. The Will of the Dragon: The Importance of an Industrial Policy 161

1.1 GDP growth, 1990–2011 11

There is excitement in Pakistan, the academic and political equivalent of winning

and contemporary literature looking at big infrastructure demonstrates that

The introduction first introduces and explains the China–Pakistan Economic

WHAT IS THE CPEC?

Map 1.1 The Major Projects of the CPEC

region of Xinjiang and in Pakistan Islamabad, parts of Punjab, Sindh, KPK,

research collaboration between Chinese and Pakistani medical institutions.

CPEC CONTEXT I: ECONOMIC OPTIMISM AND

opening in Western China and are readying themselves to export high-quality

CPEC CONTEXT II: THE CULMINATION OF

China, asked the China Development Bank (CDB) to compile a detailed

benefit through better transport links, leading to more trade, investment

CPEC CONTEXT III: ECONOMIC DEVELOPMENT

Table 1.1 GDP growth, 1990–2011

Country GDP growth, 1990–2011 (%)

Pakistan’s performance becomes even more impressive when we compare

Figure 1.1 GDP growth trends, 1960–2016

Table 1.2 GDP growth stability analysis

CPEC CONTEXT IV: THE CHINA STORY SINCE 1976

human development. These promises and aspirations were encapsulated in

Map 1.2 The Chinese Belt and Road Initiative (BRI)

announced in 2013 during a visit by President Xi Jinping to Kazakhstan.

CONTEXT PART V: A LARGER GLOBAL HISTORY

that, despite these profound international influences, it had little enduring

THE MAIN FINDINGS OF THIS BOOK

Big Infrastructure: Big Problems or Big Benefits?

CPEC Spillovers Rippling Outwards

Through the Eyes of Who? Evaluating the Success of the CPEC

The Dragon Uncoils: Special Economic Zones (SEZs) from

The Dragon’s Embrace: Pakistan–China Trade Policy

The Will of the Dragon: The Importance of an Industrial Policy

Conclusion: The Way of the Dragon or the Way of the Falcon?

MODEST STUDIES OF SINGLE

the addition of bridges, bypasses and underpasses. The so-called Golden

‘Logistics’ refers to the movement, handling and storage of goods, from

Table 2.1 World Bank overall logistics performance index

Country 2007 2010 2012 2014 2016 2018

incorporated into studies of CPEC-related infrastructure. Some of this CPEC

Table 2.2 World Bank logistics performance index: Pakistan

2007 2010 2012 2014 2016 2018

Table 2.3 shows a measure of the LPI of particular relevance to the

Table 2.3 World Bank quality of trade- and transport-related infrastructure

Country 2007 2010 2012 2014 2016 2018

AGGREGATING MODEST STUDIES TO

relatively new highway from Lahore to Islamabad will readily appreciate

The provision of infrastructure can help explain differing regional patterns

THE MACROECONOMIC STUDIES

at approximately every 1 square kilometre, which are then aggregated to

reports related to some comments made by the IMF representative in Pakistan

THE ENDOGENEITY PROBLEM WITH

Chandra and Thompson (2000) used data on interstate highway construction

List of Maps and Figures ix

2. Big Infrastructure: Big Problems or Big Benefits? 27

3. CPEC Spillovers Rippling Outwards 49

4. Through the Eyes of Who? Evaluating the Success of the CPEC 78

5. The Dragon Uncoils: Special Economic Zones (SEZs) from

6. The Dragon’s Embrace: Pakistan–China Trade Policy 145

7. The Will of the Dragon: The Importance of an Industrial Policy 161