China's Chance to Lead: Acquiring Global Influence via Infrastructure Development and Digitalization 9781009385886, 9781009385879, 9781009385855, 1009385887

How is China acquiring global influence? Rather than focusing exclusively on China's interests, this book considers

103 35 2MB

English Pages 385 Year 2024

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
vii
Contents
List of Figures page x
List of Tables xi
Preface and Acknowledgements xiii
List of Abbreviations xvii
1 Introduction 1
1.1 Why It Matters 3
1.2 Existing Explanations 5
1.3 Summary of the Argument 11
1.4 Organization of the Book 14
2 Market Failures and China’s Chance to Lead 20
2.1 Infrastructure Financing and Development 22
2.2 China’s Motivation and Capacity to Address Infrastructure
Financing Needs 32
2.3 Transaction Costs and Coordination Failures 38
2.4 Which Technologies, Industries, and Markets Would Benefit
from Standardization? 40
2.5 Why Is the Time Ripe for Standardization? 44
2.6 Why Is China Well Positioned to Promote Standardization? 47
2.7 Why Is the BRI the Best Way to Promote Chinese Standards? 54
2.8 Conclusions 56
3 Measuring Infrastructure Needs and Foreign
Infrastructure Investment 57
3.1 Measuring Countries’ Infrastructure Financing Needs 58
3.2 Infrastructure Spending and Market Failures 60
3.3 Chinese Foreign Spending 63
3.4 Conclusions 71
4 Theory: Why Countries Vary in Their Participation
in the Belt and Road Initiative 73
4.1 The Demand Side: Recipient Country Characteristics 73
4.2 The Supply Side: Characteristics of Chinese Foreign
Infrastructure Spending 86
4.3 Compatibility of Chinese Infrastructure Spending by
Political Regime 94
viii Contents
4.4 Extensions 98
4.5 Adopting Chinese Standards 100
4.6 Conclusions 106
5 Measuring Clientelism and the Corporate Sector
across Political Regimes 108
5.1 The Distribution of Political Regimes 110
5.2 Political Regimes and Clientelism 113
5.3 Political Regimes and the Public-Private Orientation of the
Corporate Sector 115
5.4 Conclusions 136
6 Political Regimes and BRI Country-Level Patterns 138
6.1 Background on China’s Foreign Spending Activity 140
6.2 Trends and Patterns of the Belt and Road Initiative 142
6.3 Chinese Foreign Investment and Construction
Activity, 2005–2019 153
6.4 Trade Effects 160
6.5 Conclusions 164
Appendix 165
7 Political Regimes and BRI Project Characteristics 174
7.1 An Illustrative Example: The Kenya Standard Gauge
Railway Project 175
7.2 Extending the Political Regimes Theory to BRI
Project Characteristics 177
7.3 Patterns and Trends 181
7.4 Conclusions 203
Appendix 206
8 Case Studies of Political Regimes and the BRI 232
8.1 Background on BRI Projects, Political Opposition, and Clientelism 234
8.2 Background on China’s International Port Investments via the BRI 237
8.3 Closed Autocracy: The United Arab Emirates 240
8.4 Strongly Durable Electoral Autocracy: Djibouti 246
8.5 Weakly Durable Electoral Autocracy: Malaysia 254
8.6 Electoral Democracy: Indonesia 264
8.7 Liberal Democracy: Greece 273
8.8 Conclusions 283
Appendix 285
9 Chinese Exports of Digital Technologies and Standards 288
9.1 The Limits of Formal Standardization 289
9.2 The Promise of De Facto Standardization 293
9.3 The Digital Silk Road and Smart Cities Technologies 295
9.4 Chinese Foreign Aid to the Information and Communications
Technology Sector 301
9.5 Case Studies 304
9.6 Huawei’s Cloud Business 308
9.7 Conclusions 310
Appendix 312
ixContents
10 Conclusions and Implications 315
10.1 A Brief Summary of the Theory and Findings 317
10.2 Increasing Alignment between China and Electoral Autocracies:
Evidence from UN General Assembly Votes 320
10.3 Theoretical Implications 326
10.4 Policy Implications 327
10.5 Business Implications 331
10.6 Conclusion 334
References 336
Index 357
Recommend Papers

China's Chance to Lead: Acquiring Global Influence via Infrastructure Development and Digitalization
 9781009385886, 9781009385879, 9781009385855, 1009385887

  • Commentary
  • Politics, Business and Public Policy, Global Influence
  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

China’s Chance to Lead

How is China acquiring global influence? Rather than focusing exclusively on China’s interests, this book considers a vital but overlooked feature – the interests of recipient countries. Richard W. Carney argues that countries in which political leaders rely more heavily on clientelism coupled with greater control over the corporate sector have a higher demand for Chinese infrastructure spending. Through a combination of statistical analyses and case studies, Carney shows that electoral autocracies (in contrast to closed autocracies, electoral democracies, and liberal democracies) display these features most prominently and are the most avid recipients. This in turn contributes to elevated levels of Chinese digital technologies imports which facilitate the spread of Chinese technical standards, enabling China to create the scale to assert its dominance over the emerging digital economy. Electoral autocracies are the most prevalent type of regime and are therefore essential partners to China’s global ambitions. Richard W. Carney has lived and worked in China since 2017. He has extensive experience teaching both Chinese and global executives, as well as policymakers from across the Asia Pacific. An advisor to  the World Bank for its flagship project Businesses of the State, he is the author of Authoritarian Capitalism (Cambridge, 2018) which won the Masayoshi Ohira Memorial Prize.

Published online by Cambridge University Press

Business and Public Policy Series Editor ASEEM PRAKASH, University of Washington, Seattle Series Board Liliana Andonova, Graduate Institute of International and Development Studies, Geneva Sarah Brooks, Ohio State University David Coen, University College, London Christopher Knill, Ludwig-Maximilians-University Munich David Konisky, Indiana University, Bloomington David Levi-Faur, Hebrew University, Jerusalem Nathan Jensen, University of Texas, Austin Layna Mosley, University of North Carolina, Chapel Hill Abraham Newman, Georgetown University Leonard Seabrook, Copenhagen Business School Edward Walker, University of California, Los Angeles Michael Vandenberg, Vanderbilt University Henry Yeung, National University of Singapore This series aims to play a pioneering role in shaping the emerging field of business and public policy. Business and Public Policy focuses on two central questions. First, how does public policy influence business strategy, operations, organization, and governance, and with what consequences for both business and society? Second, how do businesses themselves influence policy institutions, policy processes, and other policy actors and with what outcomes? Other books in the series and david switzer The Profits of Distrust: Citizens, Drinking Water, and the Crisis of Confidence in American Government jane l . sumner The Cost of Doing Politics: How Partisanship and Public Opinion Shape Corporate Influence kathryn hochstetler Political Economies of Energy Transition:Wind and Solar Power in Brazil and South Africa paasha mahdavi Power Grab: Political Survival through Extractive Resource Nationalization stefan renckens Private Governance and Public Authority: Regulating Sustainability in a Global Economy manuel p . teodoro , samantha zuhlke

(continued after the index)

Published online by Cambridge University Press

China’s Chance to Lead Acquiring Global Influence via Infrastructure Development and Digitalization Richard W. Carney Chinese University of Hong Kong

Published online by Cambridge University Press

Shaftesbury Road, Cambridge CB2 8EA, United Kingdom One Liberty Plaza, 20th Floor, New York, NY 10006, USA 477 Williamstown Road, Port Melbourne, VIC 3207, Australia 314–321, 3rd Floor, Plot 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 Cambridge University Press & Assessment, a department of the University of Cambridge. We share the University’s mission to contribute to society through the pursuit of education, learning and research at the highest international levels of excellence. www.cambridge.org Information on this title: www.cambridge.org/9781009385886 DOI: 10.1017/9781009385879 © Richard W. Carney 2024 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 & Assessment. First published 2024 A catalogue record for this publication is available from the British Library Library of Congress Cataloging-in-Publication Data Names: Carney, Richard W., author. Title: China’s chance to lead : acquiring global influence via infrastructure development and digitalization / Richard W. Carney, Chinese University of Hong Kong. Description: New York, NY : Cambridge University Press, 2024. | Includes bibliographical references and index. Identifiers: LCCN 2023015200 | ISBN 9781009385886 (hardback) | ISBN 9781009385879 (ebook) Subjects: LCSH: China – Foreign economic relations. | Economic development – China. Classification: LCC HF1604 .C3543 2024 | DDC 337.51–dc23/eng/20230720 LC record available at https://lccn.loc.gov/2023015200 ISBN 978-1-009-38588-6 Hardback ISBN 978-1-009-38585-5 Paperback Cambridge University Press & Assessment 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.

Published online by Cambridge University Press

For Mirabelle

Published online by Cambridge University Press

Published online by Cambridge University Press

Contents

List of Figurespage x List of Tablesxi Preface and Acknowledgementsxiii List of Abbreviationsxvii 1 Introduction 1.1 1.2 1.3 1.4

Why It Matters Existing Explanations Summary of the Argument Organization of the Book

2 Market Failures and China’s Chance to Lead 2.1 Infrastructure Financing and Development 2.2 China’s Motivation and Capacity to Address Infrastructure Financing Needs 2.3 Transaction Costs and Coordination Failures 2.4 Which Technologies, Industries, and Markets Would Benefit from Standardization? 2.5 Why Is the Time Ripe for Standardization? 2.6 Why Is China Well Positioned to Promote Standardization? 2.7 Why Is the BRI the Best Way to Promote Chinese Standards? 2.8 Conclusions

3 Measuring Infrastructure Needs and Foreign Infrastructure Investment 3.1 3.2 3.3 3.4

Measuring Countries’ Infrastructure Financing Needs Infrastructure Spending and Market Failures Chinese Foreign Spending Conclusions

4 Theory: Why Countries Vary in Their Participation in the Belt and Road Initiative 4.1 The Demand Side: Recipient Country Characteristics 4.2 The Supply Side: Characteristics of Chinese Foreign Infrastructure Spending 4.3 Compatibility of Chinese Infrastructure Spending by Political Regime

1 3 5 11 14

20 22 32 38 40 44 47 54 56

57 58 60 63 71

73 73 86 94

vii

Published online by Cambridge University Press

viii

Contents 4.4 Extensions 4.5 Adopting Chinese Standards 4.6 Conclusions

5 Measuring Clientelism and the Corporate Sector across Political Regimes

98 100 106

108

5.1 The Distribution of Political Regimes 5.2 Political Regimes and Clientelism 5.3 Political Regimes and the Public-Private Orientation of the Corporate Sector 5.4 Conclusions

110 113 115 136

6 Political Regimes and BRI Country-Level Patterns

138

6.1 Background on China’s Foreign Spending Activity 6.2 Trends and Patterns of the Belt and Road Initiative 6.3 Chinese Foreign Investment and Construction Activity, 2005–2019 6.4 Trade Effects 6.5 Conclusions Appendix

140 142 153 160 164 165

7 Political Regimes and BRI Project Characteristics

174

7.1 An Illustrative Example: The Kenya Standard Gauge Railway Project 7.2 Extending the Political Regimes Theory to BRI Project Characteristics 7.3 Patterns and Trends 7.4 Conclusions Appendix

177 181 203 206

8 Case Studies of Political Regimes and the BRI

232

8.1 Background on BRI Projects, Political Opposition, and Clientelism 8.2 Background on China’s International Port Investments via the BRI 8.3 Closed Autocracy: The United Arab Emirates 8.4 Strongly Durable Electoral Autocracy: Djibouti 8.5 Weakly Durable Electoral Autocracy: Malaysia 8.6 Electoral Democracy: Indonesia 8.7 Liberal Democracy: Greece 8.8 Conclusions Appendix

9 Chinese Exports of Digital Technologies and Standards 9.1 9.2 9.3 9.4

The Limits of Formal Standardization The Promise of De Facto Standardization The Digital Silk Road and Smart Cities Technologies Chinese Foreign Aid to the Information and Communications Technology Sector 9.5 Case Studies 9.6 Huawei’s Cloud Business 9.7 Conclusions Appendix

Published online by Cambridge University Press

175

234 237 240 246 254 264 273 283 285

288 289 293 295 301 304 308 310 312

Contents

10 Conclusions and Implications 10.1 A Brief Summary of the Theory and Findings 10.2 Increasing Alignment between China and Electoral Autocracies: Evidence from UN General Assembly Votes 10.3 Theoretical Implications 10.4 Policy Implications 10.5 Business Implications 10.6 Conclusion

ix

315 317 320 326 327 331 334

References336 Index357

Published online by Cambridge University Press

Figures

5.1 Distribution of political regimes among low- and middle-income countries, 1980–2019 page 111 5.2 World map of political regimes, 2019 112 5.3 Total investment in PPPs by regime type, 1990–2019 131 6.1 BRI foreign investment and construction spending combined, 2013–2019 143 6.2 BRI foreign investment, 2013–2019 143 6.3 BRI foreign construction spending, 2013–2019 144 6.4 Chinese foreign investment and construction spending, 2005–2019 154 6.5 Chinese foreign investment, 2005–2019 154 6.6 Chinese foreign construction spending, 2005–2019 155 6.7 Chinese exports to BRI-participating economies versus OECD countries, 2000–2019 162 6.8 Chinese exports to BRI-participating economies by political regime, 2000–2019 163 6A.1 Two-way interaction for electoral autocracy and durability from Model (4) in Table 6.2 165 6A.2 Two-way interaction for electoral autocracy and durability from Model (5) in Table 6.3 165 6A.3 Triple interaction for electoral autocracy*durable*2014 and later from Model (7) in Table 6.4 166 6A.4 Two-way interaction for electoral autocracy and infrastructure spending from Model (3) in Table 6.5 166 8.1 Chinese foreign spending into the UAE, 2005–2019 242 8.2 Chinese foreign spending in Djibouti, 2005–2019 249 8.3 Chinese foreign spending into Malaysia, 2005–2019 257 8.4 Chinese foreign spending into Indonesia, 2005–2019 268 8.5 Chinese foreign spending in Greece, 2005–2019 275 9.1 Chinese ICT sector number of aid commitments 303 9.2 Chinese ICT sector total amount of aid commitments (in constant 2017 USD) 304 x

Published online by Cambridge University Press

Tables

3.1 Market failure indicators and infrastructure spending page 64 3.2 Potential explanatory variables for Chinese foreign spending 69 4.1 Political regimes, clientelism, and the public-private orientation of the corporate sector 75 4.2 The compatibility of Chinese infrastructure spending by political regime 96 5.1 Infrastructure spending by political regime in 2011 (% of GDP) 113 5.2 Clientelism proxies 114 5.3 Publicly listed state-owned enterprises by regime type 118 5.4 Listed SOEs and regime type regressions 119 5.5 Rise and decline of SOEs, the Philippines (1935–2000) 122 5.6 SOE industry concentration, 2010–2019 127 5.7 Protections for private capital by regime type 129 5.8 Factors attracting private investments in PPPs in emerging economies 133 5.9 PPP legal framework dimensions by regime type and income level 135 6.1 BRI foreign investment and construction spending by political regime 145 6.2 BRI construction spending 150 6.3 BRI investment 152 6.4 Chinese foreign construction spending, 2005–2019 156 6.5 Chinese foreign investment, 2005–2019 158 6.6 Logistics performance index: Infrastructure score 161 6.7 Logistics performance index: Tracking and tracing score 162 6A.1 Chinese foreign investment and construction spending by political regime, 2005–2019 167 6A.2 Chinese foreign investment and construction spending by country-year and political regime, 2005–2019 169 xi

Published online by Cambridge University Press

xii

List of Tables

6A.3 Top two sectors for Chinese foreign investment and construction spending, 2005–2019 170 6A.4 Troubled transactions and greenfield investments 172 7.1 BRI projects initiated and completed by political regime 183 7.2 BRI project characteristics by political regime 185 7.3 Differences in proportions for BRI project characteristics across political regimes 191 7.4 Regime durability and differences in BRI project characteristics for electoral autocracies 195 7.5 Differences in proportions for BRI project characteristics in electoral autocracies 201 7A.1 Closed autocracies 206 7A.2 Electoral autocracies 208 7A.3 Electoral democracies 210 7A.4 Liberal democracies 212 7A.5 Regime durability and differences in BRI project characteristics 214 7A.6 Pairwise correlations for electoral autocracies with durable above the median 220 7A.7 Pairwise correlations for electoral autocracies with durable below the median 222 7A.8 Pairwise correlations for electoral autocracies with durable in the top 25% 224 7A.9 Pairwise correlations for electoral autocracies with durable in the bottom 25% 226 7A.10 Pairwise correlations for electoral autocracies with durable in the top 10% 228 7A.11 Pairwise correlations for electoral autocracies with durable in the bottom 10% 230 8A.1 Chinese foreign spending in the UAE, 2005–2019 285 8A.2 Chinese foreign spending in Djibouti, 2005–2019 286 8A.3 Chinese foreign spending in Malaysia, 2005–2019 286 8A.4 Chinese foreign spending in Indonesia, 2005–2019 287 8A.5 Chinese foreign spending in Greece, 2005–2019 287 9.1 Summary of Chinese smart cities technology exports 297 9.2 Chinese smart cities technologies exports by political regime 300 9.3 Huawei Cloud deals with foreign governments and SOEs 310 9A.1 Smart cities technologies by country 312 10.1 Foreign policy similarity based on UNGA votes 325

Published online by Cambridge University Press

Preface and Acknowledgements

The origins of this book go back to work I conducted tracing the rise in foreign state ownership across the Asia Pacific in the wake of the Asian Financial Crisis (Authoritarian Capitalism, Cambridge University Press, 2018). Although there were numerous anecdotes about Chinese stateowned enterprises’ (SOEs) growing expenditures in foreign markets, these were not captured by their ownership stakes of foreign firms. What could account for this discrepancy? Foreign state ownership of companies (i.e., the acquisition of assets) is the domain of conventional foreign direct investment. However, Chinese SOEs’ foreign spending is different; it is much more about infrastructure projects. This is a critical distinction. As in Authoritarian Capitalism, my initial approach was to examine whether the semi-competitive elections of electoral autocracies (or dominant party authoritarian regimes) incentivized political leaders of those regimes to attract more Chinese spending than other regimes. But it soon became apparent that this approach called for greater specificity to understand why and how this occurred differently in electoral autocracies than in other regimes, such as closed autocracies, electoral democracies, or liberal democracies. I realized it is necessary to combine both the motivation for host country leaders to attract Chinese spending and their capacity for doing so. Ultimately, the argument of this book extends that of my previous book by arguing it is due to the combination of both the extent of clientelism political leaders rely on to remain in power and the public-private orientation of the corporate sector which affects leaders’ ability to control and distribute clientelist resources. These vary systematically across political regimes, in turn yielding varying degrees of compatibility with Chinese spending on infrastructure projects. These two characteristics are most prominent in electoral autocracies, less prominent in closed autocracies and electoral democracies, and least prominent in liberal democracies. A related and potentially more impactful realization was that these features would not only affect Chinese infrastructure spending, but also xiii

Published online by Cambridge University Press

xiv

Preface and Acknowledgements

exports of Chinese digital technologies and standards. Given the clear patterns that emerged, I further posited that similar findings might be found across a wider range of issues as demonstrated by United Nations General Assembly votes. It is still early days for Chinese foreign spending, but the findings suggest a seismic shift is underway. By exploiting a vacuum in the enormous unmet demand for infrastructure among developing countries, which are predominantly autocratic, China’s influence is rapidly spreading. The analysis in this book focuses on the Belt and Road Initiative, but the aim is to identify relationships that extend well beyond this specific context. The China Europe International Business School (CEIBS) has been an ideal location to study these issues. The main campus is located in Shanghai, with additional campuses in Beijing and Shenzhen, as well as in Accra, Ghana, and Zurich, Switzerland. Although CEIBS was only founded in 1994, it has risen rapidly in business school rankings. For the past several years, the Financial Times has ranked both the Global Executive MBA (GEMBA) and MBA programs among the top ten in the world. Due to the school’s global presence and prestige, it attracts applicants from across the globe, many of whom have substantial government, SOE, and private sector experience. Additionally, the Chinese EMBA program is one of the largest EMBA programs in the world and is regarded as a leader in Chinese executive education. Research relating to this book project has benefited from teaching courses in each of these programs (via translators for Chinese EMBA courses) during the past several years. These have provided unparalleled opportunities to speak directly with hundreds of managers and senior executives with direct experience in Chinese foreign spending projects and the development and export of digital technologies. Exchanges with students both inside and outside the classroom allowed me to rapidly broaden and deepen my knowledge about the vast reach of the BRI, and to understand the opportunities and risks of greatest concern to Chinese and foreign business leaders. These exchanges have also provided a laboratory to test and refine my arguments. While the book draws on an abundance of data (17 datasets with over 20 million firm-year observations spanning more than 150 countries), the findings and conclusions are firmly grounded in the real experiences of business executives who have worked in SOEs, private firms, and in government positions. CEIBS faculty also teach students from EMBA and MBA programs from around the world that seek to learn about Chinese business, and I have been a regular contributor. These have provided additional opportunities to learn, test, and refine my ideas with audiences who have

Published online by Cambridge University Press

Preface and Acknowledgements

xv

substantial knowledge of their local business environment and the entry of Chinese firms. In addition to teaching, CEIBS faculty frequently take part in indepth company visits that typically include a tour of the facilities and an overview of the company’s past, present, and future projects by the company’s executives. Visits to Alibaba, SenseTime, and Baozun were especially informative and enlightening with regard to the historical trajectory and future aspirations of China’s digital initiatives. CEIBS has also provided numerous opportunities to host talks and forums that have provided further opportunities to deepen my knowledge and hone my arguments. For example, I co-hosted a China–Africa Forum in which Chinese entrepreneurs shared their experiences with starting up and running businesses in places such as Chinese built industrial parks in Ethiopia. A presentation of this project to CEIBS’s Faculty Research Workshop offered another venue to receive valuable feedback from Chinese business scholars from a range of disciplinary perspectives. For offering helpful comments or varying kinds of support that contributed to the successful completion of this book project, I would especially like to thank John Ravenhill, Miles Kahler, Natasha HamiltonHart, Steph Haggard, Travers Child, David Erkens, Xu Bin, Li Mingjun, Shameen Prashantham, Terence Tsai, Bala Ramaswamy, Nikos Tsikriktsis, Aldo Musacchio, Tae-Yeol Kim, Peter Moran, Sam Park, Jane Lu, Omrane Guedhami, Sadok El-Ghoul, Helen Wang, Isa Luo, and Jasmine Liu. Although China’s travel restrictions created a variety of difficulties for expats, and especially for foreign students, they granted me a sustained period of focused attention on the book project. Indeed, the lockdowns in Shanghai at the start of 2022 were particularly challenging but arrived at an especially fortuitous time with regard to completing the research and writing up the findings. Despite the solitude and deprivation of lockdown, I found joy from uncovering simplicity amid a chaos of data. Of course, this project would not have been possible without the help of others. CEIBS provided ample research funding. I am especially grateful to a team of excellent RAs who, with great dedication and attention to detail, compiled, analyzed, and made numerous follow-up adjustments to the datasets used in the book. I am especially thankful to Zhi Shao and Qing Qing Niu. I also want to thank Bo Lu who helped with the early stages of the project. At Cambridge University Press, I have had the good fortune to work with John Haslam. I am also grateful to work again with the editor of the Business and Public Policy series, Aseem Prakash. They both provided enthusiastic and sustained support for this project, and I am enormously

Published online by Cambridge University Press

xvi

Preface and Acknowledgements

grateful for their sage editorial judgment and advice. I also wish to thank the reviewers for their extremely insightful and helpful comments. For her incredible dedication and amazing skills at helping a ‘laˇowài’ navigate Chinese society, I owe an enormous debt of gratitude to my wife, Michelle Zheng. A notable example of her remarkable abilities occurred during the Shanghai lockdown when people had to ‘fight’ for food via various online apps. Michelle was so determined to place her order before the food was sold out that one of the apps mistook her rapid finger movements for a bot and blocked her from the site. I also owe special thanks to my mother-in-law. For two academics simultaneously raising a preschooler, her enormous dedication to helping us care for our daughter from the day she was born up through the Shanghai lockdown granted us invaluable time necessary for our research and teaching. Despite all this help, I ultimately take responsibility for what follows. My hope is that this book will provide a useful step in moving the literature on foreign investment beyond the focus on private firms and democracies, and in outlining an agenda for continuing research. My previous book was dedicated to my wife, Michelle. Since then, our daughter Mirabelle was born. I dedicate this book to her in the hope that she will grow up in a world less divided by misunderstandings about foreign nations than the current one.

Published online by Cambridge University Press

Abbreviations

1MDB 4G

1Malaysia Development Berhad Fourth Generation (technology standard for broadband telecommunications networks) 5G Fifth Generation (technology standard for broadband telecommunications networks) AFC Asian Financial Crisis AI Artificial Intelligence ASEAN Association of Southeast Asian Nations BGH Beibu Gulf Holding BN Barisan Nasional BOT Build-Own-Transfer BRI Belt and Road Initiative CCECC China Civil Engineering Construction Corporation CCP Chinese Communist Party CGBGIP China Guangxi Beibu Gulf International Port Group CGIT China Global Investment Tracker CRBC China Road and Bridges Corporation CRG China Railway Group CSO Civil Society Organization DAC Development Assistance Committee DDID FTZ Djibouti Damerjog Industrial Development Free Trade Zone DFTZ Digital Free Trade Zone DIFTZ Djibouti International Free Trade Zone DP Dubai Ports DPFZA Djibouti Ports and Free Zones Authority DSO Development Standards Organization DSR Digital Silk Road ECRL East Coast Rail Line EPC Engineer, Procurer, and Contractor EU European Union e-WTP Electronic World Trade Platform xvii

Published online by Cambridge University Press

xviii

List of Abbreviations

FDI FDII FOCAC GDP GFC GFCF GFDD GLC GLIC GVC HSR HTI ICRG ICT IMF IMIP IoT IPO JICA JOCIC KIZAD KPC KPH KRC M&A MCKIP MDB MIC MNC MOFA MOFCOM MOHURD MOU MPP MSRI NDRC OBOR ODA OECD OLS

Foreign Direct Investment Foreign Direct Investment in Infrastructure Forum on China–Africa Cooperation Gross Domestic Product Global Financial Crisis Gross Fixed Capital Formation Global Financial Development Database Government-Linked Corporations Government-Linked Investment Companies Global Value Chain High Speed Rail Hizbut Tahrir Indonesia International Country Risk Guide Information and Communications Technologies International Monetary Fund Indonesia Morowali Industrial Park Internet of Things Initial Public Offering Japan International Cooperation Agency Jiangsu Provincial Overseas Cooperation and Investment Company Limited Khalifa Industrial Zone Kuantan Port Consortium Kuantan Pahang Holding Kenya Railway Corporation Mergers and Acquisitions Malaysia–China Kuantan Industrial Park Multilateral Development Bank Made in China Multinational Corporation Ministry of Foreign Affairs Ministry of Commerce Ministry of Housing and Urban–Rural Development Memorandum of Understanding Multipurpose Port Maritime Silk Road Initiative National Development and Reform Commission One Belt One Road Overseas Development Assistance Organization for Economic Cooperation and Development Ordinary Least Squares

Published online by Cambridge University Press

List of Abbreviations

xix

OOF Other Official Flows PEACE Pakistan–East Africa Connecting Europe PPA Piraeus Port Authority PPC Port-Park City PPI Private Participation in Infrastructure PPP Public-Private Partnership PRC People’s Republic of China PRS Political Risk Service QJIC Qinzhou Jinqu Investment Company Ltd R&D Research and Development RMB Renminbi SASAC State-Owned Assets Supervision and Administration Commission of the State Council SBY Susilo Bambang Yudhoyono SDR Special Drawing Rights SEP Standards-Essential Patent SGR Standard Gauge Railway SME Small- and Medium-Sized Enterprises SOE State-Owned Enterprise TFP Total Factor Productivity UAE United Arab Emirates UMNO United Malay Nasional Organisation UN United Nations UNGA United Nations General Assembly UNICEF United Nations Children’s Fund US United States USD United States Dollars V-Dem Varieties of Democracy VOF Vague Official Finance WHO World Health Organization WTO World Trade Organization

Published online by Cambridge University Press

Published online by Cambridge University Press

1 Introduction

In March 2013, Xi Jinping became president of the People’s Republic of China. During the next two years, President Xi would unveil his grand vision for a new era of global connectivity, with China at its center. The first official expression of this vision occurred on September 7, 2013, during a visit to Kazakhstan, where he spoke about his ambition of connecting China to Europe with new infrastructure investments spanning the great Eurasian landmass. The next month, during a trip to Indonesia, President Xi introduced the Maritime Silk Road Initiative with the aim of enhancing China’s seaborne connectivity via new investments in foreign countries’ ports and surrounding economies, like a string of pearls. The aim of promoting global connectivity over land and sea was extended to information, later dubbed the Digital Silk Road (DSR), in a document published in March 2015, entitled “Visions and Actions on Jointly Building the Silk Road Economic Belt and 21st Century Maritime Silk Road.”1 This “Visions and Actions” statement provided the first coherent policy framework for the seemingly disparate and disconnected projects that collectively comprise the Belt and Road Initiative (BRI). As signaled by President Xi’s speech in October 2013, Indonesia appeared to be an ideal candidate for Chinese infrastructure spending. Yet, Malaysia emerged as a far more avid participant for reasons that are both surprising and puzzling. First, Indonesia had (and still has) the largest economy in the strategically important region of Southeast Asia, nearly three times larger than Malaysia’s. Second, Indonesia’s population of 250 million is nearly nine times larger than Malaysia’s, making it the most populous country in Southeast Asia. Third, Indonesia has an abundance of natural resources and was already a major natural resources exporter to China whereas Malaysian natural resources exports 1

The DSR was introduced by a white paper co-authored by the National Development and Reform Commission (NDRC) together with the Ministry of Foreign Affairs (MOFA) and the Ministry of Commerce (MOFCOM). See www.mfa.gov.cn/ce/ ceuk//eng/zywl/t1251719.htm#:~:text=The%20Chinese%20government%20has%20 drafted,African%20countries%20more%20closely%20and (accessed May 10, 2022).

1

https://doi.org/10.1017/9781009385879.001 Published online by Cambridge University Press

2

China’s Chance to Lead

to China were negligible.2 Fourth, Malaysia’s GDP per capita was more than three times larger than Indonesia’s, suggesting a far greater need for infrastructure development in the latter.3 Finally, Indonesia possessed institutional arrangements that are widely regarded as more conducive to attracting inward FDI; its democratic institutions include more veto points which help ensure greater policy stability and thereby reduce political risk in comparison to Malaysia’s government which was dominated by a single ruling party. Despite Indonesia’s numerous advantages, Malaysia attracted a far larger volume of Chinese infrastructure spending, including major investments into numerous new megaprojects. The value of newly announced infrastructure projects into Malaysia surged from USD3.5 billion in 2012 to over USD8.6 billion in 2016.4 For Indonesia during the same time period, the value of newly announced projects rose modestly from USD3.75 billion to USD3.77 billion, and there was a notable absence of similarly ambitious megaprojects.5 Malaysia also rapidly and enthusiastically participated in the DSR Initiative. In 2016, the Najib administration engaged Jack Ma of Alibaba as an advisor to develop e-commerce in Malaysia. This led to the creation in 2017 of a Digital Free Trade Zone (DFTZ), an international logistics hub, next to the Kuala Lumpur International Airport. It is dedicated to the delivery of e-commerce goods and includes an online service platform called the e-World Trade Platform (e-WTP). The cloud-based capabilities of the e-WTP and affiliated data center provided the digital foundation necessary to introduce Alibaba’s City Brain smart city solution to Kuala Lumpur in January 2018, the first city outside China to adopt it.6 DSR projects into Indonesia have been far fewer, slower, and less ambitious. What can account for these contrasting responses to China’s new initiative? I argue we must consider the public–private orientation of the corporate sector and the extent of clientelism prevalent across different types of political regimes. Clientelism refers to the delivery of goods and services in exchange for political support; the public (or state) control of the corporate sector grants political rulers greater control over the 2

For data on Malaysia’s natural resources exports to China in relation to other export categories, see Hong, Sun, Beg, and Zhou (2020). In 2013, Indonesian GDP per capita was USD3600, whereas Malaysia’s was USD11000 according to data from the World Bank. 4 See Chapter 8 for a list of Chinese-funded megaprojects in Malaysia. 5 A partial exception to Indonesia’s absence of megaprojects is the Jakarta–Bandung High Speed Rail project. 6 The smart city platform is an integrated AI-enabled system that conducts real-time data collection and integration of traffic and emergency response data from hundreds of traffic cameras and other sources. 3

https://doi.org/10.1017/9781009385879.001 Published online by Cambridge University Press

Introduction

3

allocation of clientelist benefits. These two features are most prominent in electoral autocracies, less prominent in closed autocracies and electoral democracies, and least prominent in liberal democracies. 1.1

Why It Matters

Why should we care about China’s global connectivity ambitions in the context of the BRI? There are three main reasons. First, infrastructure spending promotes economic development and alleviates poverty, especially in low-income countries (Timilsina, Hochman, and Song 2020). A lack of infrastructure comes at an enormous economic and social cost. Billions of people around the world continue to suffer from poor access to water, sanitation, and hygiene (WHO/UNICEF 2022).7 In 2020, around one in four people lacked safely managed drinking water in their homes and nearly half the world’s population lacked safely managed sanitation. At the onset of the COVID-19 pandemic, three in ten people worldwide could not wash their hands with soap and water within their homes. As of 2021, nearly 800 million people – 10 percent of the world’s population – lacked access to electricity (IEA, IRENA, UNSD, World Bank, WHO 2022).8 For developing countries in general, there exists a huge demand for infrastructure relative to supply. The Asian Development Bank reported a financing gap of USD26 trillion between 2016 and 2030 in order to support expected rates of growth among its forty-five developing country members (Ra and Li 2018). After accounting for developing countries in Africa, the Middle East, Latin America, and Eastern Europe, this number rises to over USD35 trillion (African Development Bank 2018). Between 2005 and 2019, China’s total foreign spending was over USD2 trillion; far short of addressing this gap, but very substantial when compared to foreign investment from other countries (only the United States and Japan were higher in 2019). China’s substantial infrastructure spending in the context of the BRI could have a significant positive impact on economic development and poverty alleviation globally (Chen and Lin 2018). Second, China’s ambition to link infrastructure projects to new digital technologies will shape the contours of the global business environment in the coming decades. Many view the emergence of new digital technologies as contributing to the rise of a fourth industrial revolution 7

See www.unicef.org/press-releases/billions-people-will-lack-access-safe-water-sanitationand-hygiene-2030-unless (accessed March 12 2022). 8 See www.worldbank.org/en/news/press-release/2021/06/07/report-universal-access-tosustainable-energy-will-remain-elusive-without-addressing-inequalities (accessed March 14 2022).

https://doi.org/10.1017/9781009385879.001 Published online by Cambridge University Press

4

China’s Chance to Lead

termed Industry  4.0. It refers to adaptable manufacturing systems in which production processes automatically adjust for multiple types of products and changing conditions. This allows quality, productivity, and flexibility to increase while also enabling the production of customized products at a large scale and in a sustainable way with better resource utilization. In essence, it involves the development of new smart manufacturing capabilities that are integrated into smart supply chains, a labor force trained to use new emerging technologies, and smart products that are all integrated which enables companies to rapidly adjust their operational and market needs. By leading the development and deployment of these new technologies, China aims to set the technical standards that will shape the trajectory of future technologies and manufacturing processes in the coming decades. But the implications of new digital technologies are not confined to business; they will also contribute to new smart infrastructure with digital sensors that provide real-time data and analysis to improve the use of those assets (railways, electricity grids, waste and water management, etc.). These new capabilities will in turn contribute to the emergence of smart cities which offer the most comprehensive application of these new digital technologies. Smart cities include the gathering of real-time data from physical and virtual sensors, the interconnectivity of services and technologies within a city, and the analysis of the data to optimize sustainable economic growth and quality of life for residents. By developing the new technical standards to be utilized in these new emerging digital technologies, China aims to lock in Chinese digital products and services and lock out non-Chinese competitors wherever its standards are adopted. This is part of a grand ambition on the part of the Chinese leadership from Xi Jinping on down that calls for the “integrated development of the real economy and digital economy.”9 This requires constructing physical and virtual information infrastructure at home and abroad while simultaneously controlling the core technologies and technical standards that will shape the rules of the emerging network architecture. By incorporating digital technologies into hard infrastructure projects, China binds together new technologies in “bundles” that are enabled and linked together via key baseline technologies such as 5G telecommunications, artificial intelligence, cloud computing, big data and analytics, and the Internet of Things.10 The BRI presents the 9 Quoted in de la Bruyère (2021). 10 In the context of network infrastructure, “bundles” might include broadband networks, fiber to the home products, optical transmission networks, undersea cables, super fusion solid state drives, cache technology, and striping technology. See Table 9.1 for more examples.

https://doi.org/10.1017/9781009385879.001 Published online by Cambridge University Press

Introduction

5

opportunity to integrate and promote the adoption of these technologies and standards globally, thereby creating the scale to ensure China’s dominance over the emerging digital economy. Third, the global scale of the BRI presents China with the potential to reshape the world order by coordinating among and influencing recipient countries. Through the BRI, improving intergovernmental communication between China and recipients of Chinese spending may promote the alignment of high-level government policies like economic development strategies and plans for regional cooperation. Strengthening the coordination of hard infrastructure networks like transportation systems and power grids will help reduce transport times and costs. Encouraging the development of soft infrastructure such as the signing of trade deals, aligning of regulatory and technical standards, and improving financial integration will allow for a broader range of goods and services to be exchanged with fewer regulatory hurdles. China’s strengthening capacity to promote and coordinate trade, investment, and connectivity with other developing economies may render these countries more dependent on the Chinese economy, increasing China’s economic leverage over them. At the least, rising interdependence suggests increasing alignment of interests. For these reasons, many Western observers are concerned that the BRI may empower China to more readily reshape the rules and norms that govern global affairs. These concerns are amplified by the scale of the BRI. As of August 2022, there were 149 countries formally affiliated with and endorsing the project, including 115 low- and middle-income countries as well as 34 high-income countries.11 China’s influence among developing countries is particularly noteworthy given the importance of infrastructure spending to their development needs. Since 2000, growth among low- and middleincome countries has been more than twice that of advanced economies. By 2030, the IMF projects developing countries will account for nearly two-thirds of global GDP. This is significant to understanding the potential for China to shape global affairs because two-thirds of low- and middle-income countries are autocracies as of 2019, with electoral autocracies accounting for one-half of all countries in the developing world. 1.2

Existing Explanations

How is Chinese foreign spending in the context of both infrastructure and digital technologies similar to and different from foreign investment 11

See https://greenfdc.org/countries-of-the-belt-and-road-initiative-bri/ (accessed March 15 2022). For the most part, OECD countries remain absent from this list, as well as India and numerous countries in Latin America including Brazil and Mexico.

https://doi.org/10.1017/9781009385879.001 Published online by Cambridge University Press

6

China’s Chance to Lead

by Western MNCs? I organize this overview of existing approaches to Chinese foreign spending into four parts. First, I review the prevailing explanations for FDI that is initiated by private investors, which I term “Twentieth-Century FDI Explanations.” Second, I discuss the prevailing models that seek to account for Chinese state-led foreign spending, which I call “Explanations for FDI in the Twenty-First Century.” Third, I narrow the discussion to explanations that focus on foreign spending in the context of the BRI. Finally, I call attention to gaps in our current understanding of Chinese foreign spending. 1.2.1

Twentieth-Century FDI Explanations

In the decades following World War II, the volume of foreign direct investment grew enormously and has been dominated by MNCs from advanced economies. Research has therefore focused on issues relating to supply and demand factors salient to private investors from advanced democracies. I call this the twentieth-century approach to foreign investment. Supply factors commonly focus on the importance of interest rates (the supply of capital), imperfections in the credit market (e.g., a relatively lower exchange rate reduces the price of domestic assets), and technological changes that encourage investors to set up facilities in numerous locations (Fernandez-Arias 1996). For example, firms can integrate production facilities located around the world with real-time communications thanks to improvements in telecommunications contributing in turn to cost savings via just-in-time production networks. Demand-side factors include both economic determinants (Dunning 1977; Markusen 1995; Caves 1996) and policy and institutional variables. Because FDI is characterized by cross-border jurisdiction and ex post liquidity issues, a central concern for foreign direct investors into developing countries is political risk, which relates especially to the policy and institutional domains (Vernon 1971; Frieden 1991). One of the earliest strands of research focusing on the institutional arrangements of recipient countries considers whether democratic or autocratic rule impacts foreign investment differently. Some have argued that autocracies are more attractive because democracies possess greater policy instability, interest groups can influence government policy, and democracies often possess a redistributive bias in favor of the poor which can produce higher tax rates adverse to the business sector (Huntington 1968; O’Donnel 1978; Oneal 1994; Tuman and Emmert 2004). Others contend democracies should receive more FDI because they lower political risk due to greater policy stability and credibility, foreign firms can

https://doi.org/10.1017/9781009385879.001 Published online by Cambridge University Press

Introduction

7

influence policy outcomes, democracies possess openness and transparency in the policymaking process, and political leaders incur high reputational costs if they expropriate foreign assets (North and Weingast 1989; Olson 1993, 2000; Tsebelis 1995, 2002; Li and Resnick 2003; Jensen 2003, 2008; Garland and Biglaiser 2009). A second strand of institutional explanations of FDI focuses on transaction costs (Williamson 1979, 1983, 1985). More veto players in the policymaking process enhances the credibility of host country commitments by making policies more stable which investors are posited to favor (Henisz and Williamson 1999; Henisz 2000, 2002). However, political regimes with numerous veto points have often changed FDI policies, leading researchers to consider the underlying preferences of political actors. One branch of this literature examines the role of class conflict and the potential for host governments to collude with domestic and foreign capital to exploit the popular sectors (Evans and Gereffi 1982). Others point to the economic insecurity that arises from internationally mobile capital which can be especially harmful to workers (Scheve and Slaughter 2001, 2004). Another branch examines variation in subnational political institutions on FDI inflows, such as the impact of federalism (Jensen and McGillivray 2005) and the beneficial effect of candidate-centered electoral systems (Garland and Biglaiser 2009). In contrast to conventional FDI which focuses on greenfield investments and mergers and acquisitions, foreign direct investment in infrastructure (FDII) is characterized by high capital intensity, high vulnerability to political and regulatory interference, large economies of scale, long investment periods, and high transaction costs. Taken together, these characteristics of FDII elevate the salience of host country political risk (Ramamurti and Doh 2004; Kumari and Sharma 2017). Developing countries possess the greatest demand for infrastructure, but their weak legal, financial, and regulatory frameworks and weak institutions in general collectively contribute to higher levels of political risk for infrastructure projects as compared with developed countries. Consequently, research on infrastructure investments has found that host country political characteristics have an even greater influence on infrastructure spending as compared to conventional FDI (Henisz 2002; Doh and Ramamurti 2003; Jiang et al. 2019). 1.2.2

Explanations for FDI in the Twenty-First Century

The Chinese government announced its Go Global strategy in 2000, coinciding with its admission to the World Trade Organization, which contributed to a marked increase in outward FDI starting in the mid-2000s.

https://doi.org/10.1017/9781009385879.001 Published online by Cambridge University Press

8

China’s Chance to Lead

This prompted scholars to develop alternative approaches to account for the Chinese state’s potential to influence the supply of FDI. There are three prevailing models. First, the economic statecraft perspective emphasizes the government’s use of economic means to achieve political objectives (Kastner 2009; Norris 2016). Second, the state capitalism model reverses the direction of influence by arguing that powerful stateowned enterprises (SOEs) shape the country’s foreign policy for their own commercial interests (Downs 2019). Third, the state-mobilized globalization approach views the state as operating within a network of political leaders, national bureaucracies, local governments, and SOEs that together shape firms’ FDI behavior (Ye 2020). The BRI is viewed as a mechanism to counteract the fragmented structure of China’s authoritarian state. A common thread running through each of these models is the distinctive role of the state and SOEs to Chinese foreign investment. Regardless of the model one adopts, there remains the question of which policy priorities are most likely to influence FDI behavior. Chief among these has been access to strategically important natural resources to fuel China’s rapid growth (Broadman 2006; Tuman and Shirali 2017). Others, however, point to the changing nature of China’s priorities, as manifested by the government’s “Investment Catalogue” (Buckley 2018). But underlying the specific policies is a recognition of SOEs’ privileged access to capital at favorable rates through the state banking sector, their preferential access to capital markets (Sutherland 2009; Karreman and van de Knaap 2012), and the government’s role in guiding and coordinating their investments to fulfill strategic policy aims. Consequently, these supply-side characteristics differ from those typically viewed as salient to private MNCs’ foreign investment behavior, suggesting a different set of demand-side host country attributes may matter for attracting Chinese foreign investment (Liu et al. 2017; Tuman and Shirali 2017; Chen and Lin 2018). These new attributes are likely to be most apparent in the context of BRI infrastructure investments since these types of investments are especially sensitive to host country political risk. I therefore turn to an overview of prevailing explanations for the BRI. 1.2.3

Explanations for the Belt and Road Initiative

The initiation of the BRI shortly after Xi became president and asserted more centralized control over China’s fragmented authoritarian political system has led some commentators to view the BRI through a geopolitical lens (Swaine 2015; Wang 2016; Yu 2017). They see the BRI and the establishment of the Asian Infrastructure Investment Bank

https://doi.org/10.1017/9781009385879.001 Published online by Cambridge University Press

Introduction

9

(AIIB) in 2014 as initiatives aiming to counter US moves such as the pivot to Asia and the TransPacific Partnership (Chatzky and McBride 2020), and reflecting a more assertive Chinese statecraft aimed at challenging US hegemony (Ferchen 2016; Economy 2018). From this perspective, the China-Pakistan economic corridor, granting access to Middle East oil and gas reserves via the port at Gwadar, is primarily driven by the need to reduce China’s vulnerability to shipping via the Straits of Malacca which accounts for 60 percent of total maritime trade, including 80 percent of China’s oil imports (CSIS 2020).12 This could leave China vulnerable to a strategic blockade of the straits in the event of a conflict. Even if national security objectives influence certain projects, others consider economic objectives to be the primary driver of the BRI. There are three main economic rationales typically identified for China to initiate the BRI. First, excess capacity in the wake of the government’s financial stimulus during the global financial crisis contributed to “funding [that] has been heavily skewed towards state-linked construction firms facing collapsing domestic demand” (Jones and Hameiri 2020). A second economic motivation regards the effort to reduce regional income disparities between China’s western and central regions in relation to the coastal provinces (Ye 2020). Improving connectivity and enhancing trade with neighboring countries will lift China’s less developed regions. A third economic objective regards the effort to upgrade China’s productivity via investments in digital technologies and the foreign adoption of its technical standards as signaled by the launch of the Digital Silk Road (Hemmings 2020; Naughton 2020; de la Bruyère 2021; Ding 2021; Kania 2021). China’s aging population, the need to maintain agricultural output levels (requiring adequate labor that would be unavailable for manufacturing or services sectors), and limits to the growth of its exportdependent manufacturing industry all call for greater levels of productivity and innovation to sustain China’s growth targets which are critical to social stability and the CCP’s hold on power (Sasaki et al. 2021). To this end, BRI projects have explicitly incorporated technical standards into their construction since 2015. The effort to include technical standards into BRI projects stems from the “Vision and Actions” statement mentioned earlier (NDRC 2015).13 This policy framework statement identifies five priorities for enhancing cooperation and connectivity in 12

CSIS 2020: “How much trade transits the South China Sea?” 13 The document was drafted by the National Development and Reform Commission (NDRC) together with the Ministry of Foreign Affairs (MOFA) and the Ministry of Commerce (MOFCOM). See http://de.china-embassy.org/det/zt/yidaiyilude/201503/ t20150330_3126178.htm (accessed March 13 2022).

https://doi.org/10.1017/9781009385879.001 Published online by Cambridge University Press

10

China’s Chance to Lead

the context of the BRI, including: (1) policy coordination; (2) expediting international commerce by adhering to agreed Chinese technical standards, not only in traditional modes such as rail, road, and sea, but also in border and customs controls, power supplies, and telecommunications; (3) reducing costs and risks along supply chains via unimpeded trade; (4) financial integration under Beijing’s guidance; and (5) closer people-to-people ties via scholarships and building relationships with overseas Chinese in bilateral relations. But the multiple issue areas identified in the Vision and Actions statement raise doubts among outside observers for Chinese Communist Party (CCP) leadership to assert strong centralized coordination among the thousands of BRI projects. Moreover, despite efforts at asserting more centralized political control, China’s domestic political system remains fragmented, undercutting the ability for a centralized coordinated strategy (Cai 2017; de Jonge 2017; He 2019; Jones and Hameiri 2020). While the Chinese government may provide guidance on where infrastructure spending should be directed, SOEs retain the power to decide which projects to participate in (Jones 2019). In this view, Chinese SOEs are best characterized as quasi-autonomous, profit-seeking firms, not simply instruments of economic statecraft (Jones and Zou 2017). Coordination failures at the apex of the Chinese state, between Beijing and the provinces, as well as between the provinces themselves problematize the assumption that SOEs are effectively controlled by the central government (Blanchard 2018; Gong 2019). The persistence of a fragmented development financing system gives greater weight to the role of recipient countries in driving BRI project implementation patterns. Regardless of whether one views China’s BRI as strongly or weakly controlled by the CCP leadership, there is a clear gap with regard to identifying systematic features of recipient countries and their influence on the selection and implementation of BRI projects. Indeed, the high stakes involved with the spread of Chinese technical standards coupled with the huge investments made through traditional infrastructure spending have invigorated scholars to understand countries’ varying demand for Chinese BRI spending (Vangeli 2019; Jones and Hameiri 2020). Recent work has found that Chinese investments in BRI countries differ from non-BRI countries with regard to the institutional environment (Kang et al. 2018), and the political environment more specifically (Liu et al. 2017). A number of case studies provide detail to support these general findings (Chung and Voon 2017; Blanchard 2018; Chen 2018; Gong 2019; Jones and Hameiri 2020; Hutchinson and Yean 2021). However, there remain some important gaps.

https://doi.org/10.1017/9781009385879.001 Published online by Cambridge University Press

Introduction

1.2.4

11

Gaps in the Existing Literature

Existing work on FDI – both twentieth- and twenty-first-century approaches – has focused on a democracy-non-democracy continuum. To the extent institutional variation is examined, it has typically focused on characteristics found among democracies, such as veto points and electoral systems. However, two-thirds of developing countries are autocracies. Recent scholarly work on the political structures of autocracies illuminates their heterogeneity (Svolik 2012; Truex 2016; Geddes et al. 2018). To the extent scholars have examined the relationship between autocratic political arrangements and socio-economic outcomes, attention has primarily focused on freedom of the press, civil liberties, and human rights (Maerz et al. 2020). But just as varying political characteristics have contributed to varying types of market institutions among advanced democracies (Hall and Soskice 2001; Iversen and Soskice 2019), varying types of autocratic political arrangements are likely to yield different types of economic outcomes. Yet, there remains little work that has systematically examined these relationships.14 With regard to Chinese foreign investment, including work on the Belt and Road Initiative, analyses tend to lump autocracies together (Liu et al. 2017; Tuman and Shirali 2017) or focus on country-specific dynamics (Hillman 2020; Jones and Hameiri 2020). The treatment of autocracies as relatively homogeneous, or falling along a linear democracy-non-democracy continuum, is also the prevalent approach to examining Chinese influence across other issue areas beyond foreign investment, as with the study of United Nations General Assembly voting patterns (Strüver 2016; Dreher et al. 2018). In sum, there is a clear need to identify and analyze how systematic institutional differences among autocratic regimes may impact Chinese BRI spending patterns. 1.3

Summary of the Argument

Although developing countries share many of the same types and magnitudes of market failures, they display varying levels of participation in the BRI. To explain this variation, I consider the factors driving demand on the part of recipient countries, supply factors on the part of China, and how the convergence of demand and supply factors varies by political regime. I conduct this analysis with regard to infrastructure spending first, and then theorize the implications for the spread of Chinese standards. 14

Notable exceptions include Jensen et al. (2014) and Carney (2018).

https://doi.org/10.1017/9781009385879.001 Published online by Cambridge University Press

12

China’s Chance to Lead

With regard to the demand side, two factors are highly salient to political incumbents’ decision to seek BRI infrastructure spending, including: (1) the clientelist structure of the regime, which affects the volume of resources sought for redistribution to clients in exchange for political support; and (2) the public-private orientation of the corporate sector, which affects whether the provision of clientelist resources is under state or private control. Together, these factors influence the demand for Chinese spending and the power of government officials to address that demand. With regard to the supply side, China’s party-state promoted the rapid deployment of infrastructure spending to alleviate excess capacity; but in the long term it seeks to leverage infrastructure projects to boost its total factor productivity by moving low-end manufacturing offshore and by promoting the adoption of Chinese technical standards. To achieve both of these objectives, speed matters. Alleviating excess capacity requires speed so as to enable heavily indebted infrastructure SOEs to pay their debts and minimize layoffs; speed, in the form of a first-mover advantage, also matters for promoting the adoption of Chinese technical standards before other standards get adopted. Thus, China’s SOEs have had access to abundant, cheap financing in order to promote rapid infrastructure development, with private firms engaging in DSR projects that occur in tandem with or following the initiation of infrastructure projects. While both SOEs and private firms have varying degrees of autonomy in deciding whether to engage with specific projects, in general they possess strong incentives to pursue them. Thus, the key question regards the compatibility of China’s supply factors with the demand characteristics of host country political regimes. I argue electoral autocracies have the highest compatibility with Chinese infrastructure spending. Electoral autocracies rely heavily on clientelism with resources distributed via SOEs which produces mutual interests with China in speed, opacity, and assigning greater weight to political priorities in relation to firm profitability. With regard to speed, electoral autocrats face regular elections that require a large and speedy delivery of clientelist resources to ensure regime support; this matches China’s need to quickly alleviate excess infrastructure capacity so firms can pay their debts and minimize layoffs and to win the race for the adoption of technical standards. With regard to opacity, electoral autocrats have an interest in shielding the targeted distribution of funds from public scrutiny; China also favors opacity to prevent collective bargaining by BRI recipients and to shield BRI agreements from third-party scrutiny. Finally, both electoral autocrats and China’s party-state align in assigning greater weight to political priorities relative to profit maximization,

https://doi.org/10.1017/9781009385879.001 Published online by Cambridge University Press

Introduction

13

as would occur with private firms. This convergence of interests yields the expectation that electoral autocracies will host the largest share of Chinese infrastructure spending. Liberal democracies, which have private-oriented corporate sectors and do not rely heavily on clientelism, will display the lowest convergence. Closed autocracies and electoral democracies are predicted to be in the middle. Political rulers in closed autocracies do not possess the same concern about speed because they do not hold elections. Political incumbents in electoral democracies must adhere to higher standards of transparency than their autocratic counterparts, and they place a higher priority on the interests of private firms (i.e., maximizing profits). The convergence of interests between electoral autocracies and China is amplified by two factors. First, rulers of electoral autocracies with a heightened fear of losing power will more avidly seek Chinese spending as a source of patronage to clientelist networks. Second, the relatively greater prevalence of the private sector in electoral autocracies in comparison to closed autocracies while retaining the state’s residual control rights (in contrast to electoral and liberal democracies) makes them more appealing to China for utilizing public-private partnership funding models in the future as China looks to reduce state funding over time. With regard to the adoption of technical standards by foreign firms and countries, this can occur either via de facto adoption whereby a particular standard is adopted due to its market dominance or via formal adoption which arises out of negotiations and agreements reached in the context of standards development organizations (SDOs). However, the market penetration of a standard will influence formal adoption decisions, thus elevating the importance of de facto adoption. China is aggressively pursuing the de facto adoption of its technical standards in three ways. First, China is promoting the adoption of its technical (digital) standards by directly incorporating them into its infrastructure projects. Second, China promotes the delivery of technology packages whereby a recipient country’s adoption of numerous technology products effectively locks in Chinese standards, making it difficult and costly to switch to an alternative standard even if it is better. Third, China is promoting the development of global value chains that incorporate Industry 4.0 technologies and standards by encouraging the redeployment of low-end manufacturing facilities to foreign markets, which is enabled by the building of industrial parks in the context of the BRI. Among political regimes, electoral autocracies are posited to display the highest rate of adoption of Chinese technology products and standards and to be most likely to embed themselves in China-led GVCs due to their avid pursuit of Chinese infrastructure spending and their lack

https://doi.org/10.1017/9781009385879.001 Published online by Cambridge University Press

14

China’s Chance to Lead

of domestic rivals to Chinese technical standards. In this way, the rapid adoption of China’s de facto standards by electoral autocracies bolsters China’s negotiating position for the formal adoption of its technical standards in the context of SDOs. 1.4

Organization of the Book

The book begins by providing the context for China’s Belt and Road Initiative in Chapter 2. Developing countries possess market failures that limit their ability to address their infrastructure financing and development needs. This chapter begins by identifying these market failures. It then discusses the main alternatives for addressing these needs and why China stands out as particularly well positioned to address them via the BRI. Chapter 3 empirically establishes the relationship between countries’ market failures and their level of infrastructure spending. The findings demonstrate that countries with more developed banking systems spend more on infrastructure, as expected. But the most noteworthy result is that domestic spending declines as financing from international sources increases, suggesting a substitution effect for foreign investment and lending with infrastructure spending. This motivates the need to investigate foreign sources of spending more closely. To this end, I test the most likely explanations for Chinese BRI spending based on the FDI literature relating to Western MNCs. The results do not yield strong significant results, motivating the need for an alternative explanation. In Chapter 4, I present a novel argument to explain varying levels of Chinese spending across countries in the context of the BRI. The chapter begins by discussing the characteristics of developing countries that influence their demand for infrastructure spending followed by the characteristics of Chinese foreign spending that affect its supply. The demand-side characteristics include the structure of the political regime, the corresponding structure of the clientelist network, and the publicprivate orientation of the corporate sector. Together, these yield regimespecific demand characteristics for infrastructure spending. On the supply side, I focus on the interests of three actors, including the Chinese Community Party, Chinese state-owned entities (e.g., state-owned enterprises, the China Development Bank, the Export-Import Bank of China, and the Silk Road Fund), and private firms. I then consider the compatibility of recipient countries’ demand characteristics with China’s supply characteristics to generate specific predictions about countries’ varying receptivity to Chinese infrastructure spending. Following these predictions, I then draw implications arising from two additional factors,

https://doi.org/10.1017/9781009385879.001 Published online by Cambridge University Press

Introduction

15

including political leaders with an insecure hold on power and the rising importance of private investment. The latter portion of the chapter turns to the impact of Chinese infrastructure spending on the adoption of Chinese standards in the context of the Digital Silk Road. The adoption of Chinese standards can occur either via de facto or via formal standardization. The former refers to market dominance; the latter refers to agreements in the context of standards development organizations. I first consider the relative benefits of each type of standardization method to China’s interests and theorize de facto standardization offers the greater benefit. I then turn to a more detailed consideration of the factors that affect countries’ demand for Chinese standards which is critical to Chinese de facto standardization. The demand factors include commercial and development policy factors as well as political factors (e.g., surveillance of political opposition). I theorize the spread of Chinese technical standards depends on the initial desire for Chinese infrastructure spending. Thus, electoral autocracies are expected to be the most avid participants in the BRI and the most enthusiastic adopters of Chinese technical standards, especially electoral autocracies with political leaders whose rule is insecure. Chapter 5 turns to empirics regarding the demand side; that is, the characteristics of countries which may host BRI projects. First, I discuss the measurement of political regimes. I then demonstrate that infrastructure spending is broadly similar across political regimes, thus calling for a novel approach to explain why Chinese spending varies across countries if it is not due to differences in infrastructure spending needs. I then discuss the prevalence of clientelism across political regimes followed by the public-private orientation of the corporate sector across political regimes. Chapters 6 through 9 empirically assess my argument for why Chinese foreign spending varies across countries. Chapter 6 begins with an overview of China’s foreign spending in the context of the BRI using a well-established dataset for this topic – the China Global Investment Tracker (CGIT).15 Analysis of the data by political regime reveals electoral autocracies are the major recipients, by far. I then conduct multivariate analyses to control for alternative explanatory factors. The results confirm the strong significant relationship between electoral autocracies and Chinese foreign construction spending. Electoral autocracies with unstable political rule display the strongest relationship to Chinese foreign construction spending. I then assess whether these patterns predate 15

The CGIT dataset is the primary data source for analyses of the BRI conducted by the OECD (2018) and the World Bank (Chen and Lin 2018).

https://doi.org/10.1017/9781009385879.001 Published online by Cambridge University Press

16

China’s Chance to Lead

the BRI with a broad overview of China’s global investments between 2005 and 2019. While there is some weak evidence for a pre-BRI relationship, the results indicate electoral autocracies become newly important to Chinese foreign spending following the launch of the BRI at the end of 2013. Given the importance of the BRI to Chinese foreign spending, Chapter 7 assesses how the characteristics of BRI projects vary across political regimes, including the number of projects initiated, project completion rates, the use of Chinese financing, and the prevalence of public-private-partnerships. To examine these characteristics, I turn to a newly constructed dataset that provides project-specific information for over 2,000 BRI projects located outside China.16 Analysis of this new dataset reveals that electoral autocracies host the largest share of BRI projects, they have the highest completion rate, and they host the largest share of BRI projects that are structured as public-private partnerships. Overall, the findings of this chapter corroborate and extend those of Chapter 6 that identify electoral autocracies as having a disproportionately important role in the BRI, especially those in which the leader possesses an insecure hold on power. To provide context to the findings of the previous chapters, and to identify how political regimes engage differently with China’s BRI, Chapter 8 conducts a structured comparison of five country cases. The cases are selected on the basis of representing the political regimes studied in prior chapters and because they each host a major BRI project implemented by a Chinese state-owned enterprise. The country cases include the United Arab Emirates (closed autocracy), Djibouti (strongly durable electoral autocracy), Malaysia (weakly durable electoral autocracy), Indonesia (electoral democracy), and Greece (liberal democracy). Each case begins with a background discussion of the country’s political system, an overview of Chinese foreign spending in the country, and then a focused analysis of a specific BRI project. The conclusion discusses similarities and differences across the cases. Overall, the qualitative findings presented in this chapter are consistent with the quantitative findings of the previous chapters and provide evidence for the causal link between political regimes and varying levels of Chinese foreign spending as part of the BRI. Chapter 9 turns to the export of Chinese technologies and standards, especially in relation to new and emerging digital technologies. The establishment of technical standards can occur either through negotiations 16

The dataset comes from Refinitiv, formed by a collaboration between Thomson Reuters and the Blackstone Group. Ten variables for each project are manually coded.

https://doi.org/10.1017/9781009385879.001 Published online by Cambridge University Press

Introduction

17

and decisions in the context of international professional associations (formal standardization) or by increasing the use of a standard in the marketplace by selling more products and services that adhere to it (de facto standardization). While China is gaining influence in standards development organizations, it has achieved greater success through the de facto adoption of its standards, especially among countries that participate in the BRI. The construction of railways, industrial parks, power stations, and other infrastructure creates opportunities to introduce bundles of technologies that lock in recipient nations to Chinese standards. Smart cities technologies are a particularly effective method of achieving the export of Chinese technologies and standards due to swelling urbanization rates across the developing world. In this chapter, I examine the prevalence of these exports by political regime with a new dataset that includes more than 50 Chinese companies with investments in 317 foreign projects in technologies related to the development of smart cities such as artificial intelligence, surveillance, and data storage. The findings indicate electoral autocracies are the most avid recipients of these technologies among developing countries. Because these technologies are reliant upon data transmission networks with adequate speed and volume, I also examine the prevalence of aid and loans affiliated with the ICT sector. I further examine whether these patterns continue during the pandemic when Huawei began an aggressive drive to grow its cloud business. The findings reaffirm the importance of electoral autocracies. Finally, I demonstrate with case studies that the adoption of Chinese technologies and standards is linked to the country hosting BRI projects in the context of an electoral autocracy (Malaysia), but less so in the context of a liberal democracy (Greece). Chapter 10 is the concluding chapter. It begins with a brief summary of the theory and evidence. The findings demonstrate that electoral autocracies are major recipients of Chinese infrastructure spending and digital technologies. This raises the question of whether electoral autocracies display an increasing alignment with China across a wider range of issue areas. To answer this question, I analyze United Nations General Assembly votes between 1990 and 2019. The results indicate that since the launch of the BRI, the relative likelihood of electoral autocracies voting with China increased at a higher rate than closed autocracies or electoral democracies. This finding provides further motivation and context for considering the theoretical, policy, and business implications of this study. The primary theoretical contribution of this study is to move beyond twentieth-century explanations for FDI that focus on demand and supply factors relating to private capital and democracies. The theory and

https://doi.org/10.1017/9781009385879.001 Published online by Cambridge University Press

18

China’s Chance to Lead

evidence presented in this book advances our understanding of FDI by developing and testing new theory relating to state-owned entities and autocracies. Specifically, the evidence indicates a stronger compatibility between China and electoral autocracies in comparison to other regimes. The second theoretical implication extends the political regimes framework of this book to a consideration of other economic phenomena. The third theoretical implication relates to the capacity for China to enhance international coordination due to network effects that enhance the depth and breadth of its foreign relations. I discuss four main policy implications. First, I consider how infrastructure development needs to be more attentive to the political incentives of autocratic leaders, especially in electoral autocracies. This regards both the design of infrastructure financing arrangements and understanding the limits to which policy can work due to how it conflicts with leaders’ underlying political incentives. With China now offering an important new source of infrastructure development financing, Western multilateral development banks (MDBs) must adjust their policy approach to developing countries to be competitive with Chinese proposals. Second, although the prevailing private-led approach to standards-setting (as in the EU and the United States; Büthe and Mattli 2011) has proven very successful over the past several decades, it possesses certain fundamental limitations that impede its competitiveness relative to China’s stateled standards-promoting regime, such as an unwillingness on the part of private firms to invest in markets often considered too risky (developing countries). Consequently, the private-driven model is unable to compete effectively against China in many developing countries. This is concerning for Western actors given these economies have displayed faster growth rates over the past two decades than advanced economies and developing countries now account for a larger share of global GDP than advanced economies. This suggests Western governments need stronger public-private collaborations to compete against China in these politically risky locations. Third, this study offers novel insights into corporate social responsibility among autocracies. Clientelist benefits typically improve the working conditions of employees and the environment of designated neighborhoods; in other words, they may be viewed as promoting socially responsible corporate behavior when in fact they are simply mechanisms to promote political support of incumbent rulers. Consequently, the measurement of CSR among developing countries may be misconstrued as socially responsible corporate behavior when in fact improvements in social and environmental outcomes are driven by political expediency. Fourth, the findings of this book also offer a new perspective on the debt-trap diplomacy debate by demonstrating that

https://doi.org/10.1017/9781009385879.001 Published online by Cambridge University Press

Introduction

19

recipient countries possess greater agency and variation (depending on their regime type) in their borrowing decisions with China. Additionally, if host country leaders perceive China as exploiting its lending arrangements, then they will be reluctant to work with China in the future which is contrary to China’s long-term interest of using the BRI (and the export of digital technologies and standards) to promote its own economic growth. In short, the view of China as a predatory lender is incorrect. The chapter also discusses three business implications. First, this study offers helpful insights into how political risk varies among autocracies. The conventional approach tends to group these regimes together, but this book offers a novel way to distinguish between them to the benefit of foreign investors. Second, I discuss the benefits of public-private collaboration from the perspective of how it can benefit managers, paralleling the earlier discussion from the perspective of policymakers. The third implication concerns the business education curriculum. This was originally conceived and developed in the liberal democracies of the United States and the UK. But China’s new model changes the two basic foundations on which business operates, shifting attention away from private ownership and democratic rule-making to state ownership and autocratic decision-making. The way managers of private firms analyze business opportunities and risks needs to be fundamentally re-examined, especially if they want to participate in countries that are likely to display the fastest growth in coming decades. The book ends on an optimistic note about the likely benefits to flow to the poor in developing countries as a result of growing geopolitical rivalry for influence among this set of countries in the coming decades.

https://doi.org/10.1017/9781009385879.001 Published online by Cambridge University Press

2

Market Failures and China’s Chance to Lead

Since the turn of the century, the demand for infrastructure investments in developing countries has grown faster than economic output, and therefore tax revenues, creating a widening gap between infrastructure needs and available public funding. This infrastructure financing gap has invigorated the search for alternative sources of financing. The private sector is the first obvious choice with banks as the main option.1 Yet despite high savings rates in many developing countries (especially in Asia), there remains a maturity mismatch problem between shortterm bank deposits and long-term project financing (Yoshino 2012). Alternative private financing could also come from nonbank financial institutions funds (e.g., pension funds, insurers, sovereign wealth funds) and capital markets (e.g., infrastructure bonds, corporate bonds, and listed equities). Although long-term institutional investors possess very large and growing long-term liabilities and a growing desire for a diversified portfolio of long-term assets, only a very small fraction of their financial resources has been allocated to infrastructure. Major barriers include limited regulatory capacity to provide a steady stream of attractive projects to private financiers (e.g., lack of suitable projects, poor procurement processes, project size), as well as an inadequate intermediation process and market structure in host countries (e.g., lack of secondary markets, weak capital markets, expensive investment vehicles) (Inderst  2009; Della Croce 2011). But the massive and growing demand for infrastructure financing in developing countries suggests they should be searching for solutions to these problems. Given the slow pace of reform, many countries have 1

Developed countries often rely on a higher share of private financing than developing countries. For example, about 70 percent of infrastructure is funded by private sources in the UK (HM Treasury 2014). But in developing countries, public funding still accounts for about 70 percent of total infrastructure expenditures according to World Bank estimates, while 20 percent is financed by private sources and the rest by development banks (Delmon and Delmon 2011; Bhattacharya et al. 2012).

20

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

Market Failures and China’s Chance to Lead

21

turned to China rather than multilateral development banks to fill their infrastructure financing needs. Why? Developing countries possess market failures that limit their ability to address their own infrastructure financing and development needs, leading them to search for alternative sources of financing. Market failures occur when markets fail to allocate resources in a way that maximizes the general welfare of society. The First Fundamental Theorem of Welfare Economics specifies three conditions for markets to allocate resources so as to maximize society’s general welfare (Arrow 1951; Debreu 1959). First, all goods and services must be traded at publicly known prices. Second, all consumers and producers must behave competitively (i.e., they are price takers). Third, an equilibrium exists. These conditions are posited to yield a “Pareto-optimal” allocation of resources such that any other allocation that makes one actor better off would also make another actor worse off. Market failures occur when the first and/or second conditions are violated, yielding an inefficient (or Pareto suboptimal) allocation of resources. Within a national economy, there are four general categories of market failures that arise from violating the first and/or second conditions of the First Fundamental Theorem of Welfare Economics. The first category concerns the undersupply of capital for strategic investments that private actors are unwilling to make in order to promote economic development. This can occur for projects that promote the economic development of a geographic region or industry. The second category involves an undersupply of finance due to business cycles that create intertemporal dynamics that push the economy onto a Pareto-inefficient path (Stiglitz 1974, 1991). Agents are unable to coordinate their expectations and preferences due to information asymmetries and high screening costs (the cost of obtaining information about other actors’ preferences and expectations is too high), or issues of free riding (agents do not change their preferences/expectations out of fear that other agents will benefit from their actions). The third category concerns innovation development including financing for startups and SMEs due to the lack of a track record of profitability and the uncertainties associated with the innovation process (Griffith-Jones and Tyson 2013). The fourth category regards missionoriented objectives that address a particular societal challenge such as “putting a man on the moon” as in the case of Project Apollo (Mowery 2010; Foray et al. 2012). If we extend the boundaries of an economic system beyond a single national economy then additional market failures arise as the first and second conditions of the First Fundamental Theorem of welfare economics are more likely to be violated. The most notable market failure concerns

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

22

China’s Chance to Lead

the harmonization of regulatory and technical standards so as to reduce transaction costs. The creation of the European Common Market offers a vivid illustration of the challenges (and potential benefits) associated with reducing these costs. Additional market failures at the international level are associated with an inability to coordinate monetary and fiscal policies, which are also manifested clearly in the European context. The market failures that China’s Belt and Road Initiative is best positioned to address include: (1) the undersupply of capital to promote economic development or infrastructure financing; and (2) reducing transaction costs via coordination on regulatory and technical standards across countries. With respect to each of these topics, the remainder of this chapter discusses the specific types of market failures that occur, the underlying reasons for these market failures, and the opportunity this creates for China to address these failures. Before proceeding to the next section, it is important to note that market failures occur not only with regard to financing, but also the capacity to execute large, complex infrastructure development projects. For example, many Mideast monarchies do not need external financing but nevertheless turn to China to complete major infrastructure projects. Thus, the undersupply of “capital” to promote economic development can refer to either financing or physical and/or human capital (Stiglitz 1989; Meier and Stiglitz 2000). This makes the broader measure of Chinese infrastructure spending (used in the chapters that follow) a more valid measure than one that focuses on the subset of infrastructure financing.2 That said, the market failure of greatest concern for most developing countries regards inadequate financing and is therefore the focus of the discussion that follows. 2.1

Infrastructure Financing and Development

2.1.1

Traditional Infrastructure Financing Options

The main infrastructure financing options governments typically rely on include public and/or private financing. However, inadequate public funds and market failures contributing to the undersupply of private 2

Often, the host country will use a public-private partnership model in which a fraction of the project is paid for by the host country and then China finances the rest. In other words, Chinese financing occurs along a continuum with China financing zero percent of the cost up to 100 percent. Most of the time, Chinese financing is somewhere between these two extremes. For this reason, the broader measure of Chinese spending is of interest. But for most countries, it occurs in tandem with the host country’s need to address an infrastructure financing shortfall typically due to market failures.

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

Market Failures and China’s Chance to Lead

23

capital have prompted the search for alternative funding sources. I discuss the conventional sources of infrastructure financing and then turn to a discussion of the market failures that have prompted the search for alternate sources of financing. 2.1.1.1 Public Financing The traditional approach to funding infrastructure projects involves the government allocating funding from general budget appropriations financed by: (1) tax revenue or government debt; (2) revenue bonds tied to specific projects; or (3) infrastructure investment by state-owned enterprises, including national development banks. Infrastructure appropriations from the government budget typically enable a higher level of public accountability. Public procurement can be structured to enable transparent bidding and awarding of contracts, with competition that fosters cost-effective infrastructure proposals. A lower cost of government debt in comparison to private sector debt can also make government appropriations a cost-effective financing option. However, there are several limitations. For example, tax revenue may be inadequate to meet infrastructure financing needs, fiscal rules such as debt or deficit limits can limit infrastructure development, and political pressures may lead to the financing of particular projects regardless of a cost-benefit analysis or the awarding of procurement contracts to favored political allies. Privatization of existing assets can offer an additional source of public financing for new infrastructure projects, but a strong regulatory framework is needed if it is a natural monopoly, for example. SOEs can offer additional financing options. SOEs are legally distinct entities that are wholly or partially owned by the government. In some cases, the core function of an SOE is to operate and maintain infrastructure assets, such as a state-owned port authority or utility operator. Nonfinancial SOEs can finance infrastructure investments through several sources, including retained earnings, capital contributions or payments from the government for non-commercial services the SOE is directed to provide (e.g., dredging a harbor), bond issuance, selling equity upon listing on a stock market, and borrowing from banks. With proper corporate governance in place, SOEs can be more efficient and make better investment decisions than the general government sector. However, SOEs also possess several drawbacks. First, SOEs may operate monopoly infrastructure and lack incentives to improve efficiency, services, and facilities. Second, the ability of SOEs to issue debt and sell equity raises concerns about budget accountability measures. Finally, moral hazard may arise as SOEs can generate large contingent liabilities for governments that find it politically unpalatable to let the SOE or an infrastructure project fail.

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

24

China’s Chance to Lead

In addition to nonfinancial SOEs are state-owned banks. They possess many of the characteristics of nonfinancial SOEs in that they can raise capital from a variety of sources that are separate from the government budget, but they also have additional capabilities of attracting deposits and making loans. Their unique power to mobilize savings for infrastructure investment commonly leads to strong government oversight but also magnifies the potential for moral hazard and the potential for political influence on their lending decisions. National development banks (e.g., the China Development Bank) are a special type of state-owned entity that are established with the explicit mission of providing credit to sectors of the economy that private financial institutions are considered to be underservicing. Because they operate with the benefit of an explicit government guarantee, they possess a credit rating that is tied to that of the government which grants them access to lower financing costs than private banks in their jurisdictions. Export-import banks (e.g., the Export-Import Bank of China) possess similar benefits but with a mission to promote the development of the country’s export of goods and services. Developing countries often face a variety of challenges in relying on public financing to fund infrastructure development, such as an inadequate tax base to fund needed infrastructure from budgetary appropriations, banking systems that are inefficient or dysfunctional (e.g., an inadequate deposit base), underdeveloped and illiquid capital markets that fail to attract private investors, a sovereign debt rating that limits the capacity to raise financing from international investors, weak regulatory oversight, and political interventions with distortionary effects. These various market and governance failures (to be discussed below) have contributed to public financing constraints that lead governments to look to the private sector to supplement their infrastructure financing needs. 2.1.1.2 Private Financing Private investor involvement in infrastructure projects alleviates some of the public financing burden and often improves the efficiency of project completion and performance of the asset by incorporating the latest technological advances and managerial techniques into its construction and servicing. There are two sources of private financing for infrastructure projects – debt and equity. Debt financing is by far the larger portion. It is primarily composed of bank loans, although bonds provide a small but growing amount. Bank loans dominate financing in the initial phase of an infrastructure project due to several advantages they possess over bonds or other structured financing solutions. First, banks are more

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

Market Failures and China’s Chance to Lead

25

likely to have the necessary expertise and incentives to effectively monitor the project’s planning and construction phases. Second, banks possess greater flexibility in the disbursement of funds. Third, banks can more easily accommodate debt restructurings due to unforeseen events in comparison to bonds, which is complex and time-consuming. Although bond financing is rare in the construction phase, it becomes more common in the operational phase when debt restructurings are less common. Finally, the crucial monitoring role banks perform in the planning and construction phases often attracts other groups of investors with fewer monitoring capabilities. However, banks’ short-term liabilities limit the maturity of assets they can safely hold. Infrastructure projects typically require long-term funding due to the relatively long time between construction and the generation of positive cash flows. Equity investors in private projects are either primary or secondary investors. Primary investors directly participate in decisions regarding the construction of the infrastructure asset, such as construction companies. Some projects may also raise equity via an initial public offering. Once projects are in operation with a proven revenue stream, equity is often sold in the secondary market to investors with a lower appetite for risk, such as pension funds or other institutional investors, enabling primary equity investors to free up capital to invest in new infrastructure projects. Ineffective legal and regulatory arrangements coupled with underdeveloped capital markets limit the use of equity in developing countries. Indeed, a major challenge to reaping the benefits of private financing is the prevalence of market failures among developing countries that limit private financing options and raise their costs. These market failures are particularly challenging to address with regard to infrastructure projects given their complexity and unique characteristics. Consequently, some form of public support is typically necessary to promote private sector participation. 2.1.2

Market Failures for Infrastructure Financing

Which market failures are responsible for developing countries’ infrastructure financing gaps? Within the general market failure category regarding the undersupply of capital to promote economic development, infrastructure assets possess several common characteristics that differentiate them from other asset classes, making them especially susceptible to certain types of market failures that impede the matching of investment demand with financing supply from private funders. First, infrastructure projects often possess “public good” characteristics that

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

26

China’s Chance to Lead

provide non-excludable and non-rival goods or services. In these cases, private actors may not provide a good or service because a fee cannot be charged for consumption of a non-excludable good or service, as with street lighting or defense. Second, infrastructure projects may generate “positive externalities” whereby the social returns to provision outweigh private returns to provision. For example, socially desirable infrastructure projects with positive externalities, such as hospitals or schools, may not be funded by private actors because they cannot internalize benefits that would compensate for the development costs and make the investment worthwhile (Arrow 1999). Third, “negative externalities” in which the social costs of provision of an infrastructure asset outweigh the private costs of provision, contributing to overproduction by private actors without government regulation, as with fisheries or the use of radio frequency. In these instances, the government must have an adequate regulatory capacity to permit private actors’ participation. Fourth, a “natural monopoly” in which the market can accommodate only one player efficiently may cause the government to limit or deny private participation as with electricity transmission, for example. A strong regulatory framework is necessary to open up this space to private participation. Fifth, private lenders may be unwilling to fund the construction of large-scale infrastructure projects if the complexity and scale of such projects exceeds the capabilities of domestic firms (Stiglitz 1991). Additionally, large and complex infrastructure projects typically involve a large number of actors, such as construction companies, operators, government authorities, private investors, insurers, and the citizens most directly affected. But because infrastructure often involves natural monopolies such as an airport or seaport, governments typically want to retain ultimate control to prevent an abuse of monopoly power. This requires complex legal, financial, and organizational arrangements to ensure the incentive-compatibility of contracts, the nature of contingencies, and the proper distribution of payoffs and risk-sharing. The high level of expertise necessary to delineate the contractual and financial complexities of such projects may exceed the capabilities of domestic firms, limiting the participation of private lenders. Sixth, infrastructure investments often generate cash flows only after many years, with the initial planning and construction phases of the project subject to high risks. For example, the availability of land allocated for the project may be subject to legal dispute or changes in the composition of the government may lead to unilateral changes to the project contract or even its cancellation. Additionally, the unique services

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

Market Failures and China’s Chance to Lead

27

provided by infrastructure investments make them less liquid. Together, the temporal risks associated with cash flow generation, elevated risks in the initial planning and construction phases, and illiquidity elevate the riskiness of infrastructure assets for private investment. Even if the project by itself appears to be financially viable, the quality of institutions and the rule of law may further impede the participation of private investors given the likelihood for unexpected events to occur during the long duration of such projects (typically twenty-five years or more). An additional market failure common to developing economies is incomplete or non-existent property rights (De Soto 2000). This impedes the development of markets and the capacity to finance infrastructure projects due to incomplete information about agents’ ability to repay loans and the existence of imperfect competition if property rights are well defined for only a subset of agents, contributing to the violation of two conditions of the First Fundamental Theorem of Welfare Economics discussed earlier. To attract private investors to finance infrastructure projects in developing countries requires host governments to develop a pipeline of properly structured projects that offer attractive returns. Additionally, there must be a coherent and trusted legal framework for infrastructure projects. If disputes arise between public and private actors, there must be clear legal codes and established precedent for protecting private investors. But even if these project and legal risks are addressed, an additional major challenge for attracting infrastructure investment to developing countries is political risk (Osei-Kyei and Chan 2017). The arbitrary exercise of political power can take many forms, such as sudden cuts to the prices private infrastructure operators are permitted to charge, land availability, new regulations, or changes to contracts by a new government. Even if legal frameworks are updated to address these risks, it takes time to demonstrate their effectiveness in order for private investors to have confidence that they function effectively and independently from political interference. Political risks are particularly acute in developing countries since the market failures mentioned above are especially prevalent, requiring public support for the successful realization of the infrastructure project. This support may take the form of direct financial assistance, some form of insurance, the allocation of land, or resolving disputes between parties. Given the scope and prevalence of market failures that limit the ability of governments to address their widening infrastructure financing gaps, government officials have turned to three alternative funding sources – public-private partnerships, multilateral development banks (MDBs), and foreign states.

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

28

China’s Chance to Lead

2.1.3

Alternative Financing Options

Why is China well positioned to address developing countries’ infrastructure financing gaps? To address this question, I discuss alternatives to the conventional public and private sources of infrastructure finance, including public-private partnerships, MDBs, and foreign states such as China. The discussion illustrates that for many developing countries, China is the most preferred option. 2.1.3.1 Public-Private Partnerships For developing country governments, private partners can help to alleviate some of the financing costs while also introducing efficiency gains compared to purely public procurement. Private sector participation falls along a continuum with regard to the extent of private participation and the allocation of risk. At one end, private partners are only involved in the operation and maintenance of an asset that is owned, built, and financed by the government; thus, the government assumes most of the risk. At the other end, the public entity may only be involved in purchasing the asset at the end of its operations by the private partner; in this case, the private actor assumes most of the risk. Between these two extremes are varying public-private arrangements according to which actor owns the asset and is responsible for its engineering and design, construction or renovation, operation and management, and financing. Ideally, only those risks which the private partners either control or are able to insure against should be allocated to them. In general, risks can be placed into two categories including project-specific and market-wide risks. Private partners have greater ability to address project-specific risks, such as construction, operation, and maintenance of an asset. Market-wide risks should typically remain with the government, including political, regulatory, legal, interest rate, and exchange rate risks, as well as land acquisition (Gatti 2015). In developing countries in particular, infrastructure projects entail significant political risks since governments retain the power to renegotiate contracts and the legal system is typically unable to protect investors against this (which often occurs following a change in the ruling party). Because cash flow generation often occurs only after a long period of time, the potential for political risks is heightened. In a survey of global PPP experts, the top three factors (out of sixteen) considered important to attracting private investments in PPPs in emerging economies include political support for PPPs, government support for private sector investments, and political stability (Osei-Kyei and Chan 2017).

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

Market Failures and China’s Chance to Lead

29

2.1.3.2 Multilateral Development Banks To fill their infrastructure financing gap, many governments turn to multilateral development banks, such as the World Bank or one of the regional development banks. The structure of contemporary MDB lending arrangements can be traced to the response to the Third World debt crisis of the early 1980s when financial aid to indebted governments was used to incentivize market-liberalizing reforms via conditionalities, which refer to the provision of financial resources to governments in exchange for the adoption of particular policies (Dobbin et al. 2007). By the end of the 1980s, nearly 30 percent of World Bank disbursements were made via policy-conditional “structural adjustment” loans. These were complemented by IMF initiatives to promote not only currency stability but also market liberalization. These policies, collectively known as the “Washington Consensus,” included the privatization of state-owned industries, the removal of trade barriers, opening up to foreign investment, and balancing the national budget. In the early 1990s, regional development banks began following this approach, with Eastern European countries subject to similar conditions following the collapse of the Soviet bloc (Williamson 1994). These initial conditions were expanded in the mid-1990s with the “augmented Washington Consensus” to also include institutional reforms such as central bank independence, corporate bankruptcy legislation, and “pro-poor” conditions such as minimum targets for social spending (Kuczynski and Williamson 2003). In some cases, conditions were extended to the privatization of public pension systems or the removal of capital controls. The number of conditions eventually attached to any particular loan could be very expansive. For example, in its IMF loan request during the 1997 Asian Financial Crisis, the Indonesian government committed to more than 100 policy conditions. These imposed policy constraints often required politically painful reforms, prompting calls for changes to the voting rights of developing countries in these organizations and for alternative MDBs that were more aligned with the interests of developing country governments. These complaints were bolstered by evidence that the countries that had most fervently implemented the Washington Consensus recipe, such as those in Latin America and Eastern Europe, were failing to thrive (Stiglitz 2002), and suffering from increases in unemployment and inequality (Forster et al. 2019). It was also impossible to ignore that China had achieved remarkable long-term economic growth and impressive social indicators despite ignoring Washington Consensus prescriptions (Rodrik 2006). In 2006, economist Dani Rodrik commented, “it is fair to say that nobody really believes in the Washington Consensus anymore” (Rodrik 2006: 974).

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

30

China’s Chance to Lead

Despite this backtracking, the IMF’s advocacy of deep structural reforms persists. For example, during the Eurozone bailouts of the 2010s the IMF and European Union institutions collaborated in the design of reform packages that included not only austerity measures but also civil service reorganizations, labor market liberalization, and the privatization of state assets and enterprises (Fitoussi and Saraceno 2013). The main aims of the World Bank and other multilateral development banks still seek to promote liberalizing reforms, but with subtler tactics. For example, in the “billions to trillions” initiative, they proposed using official development assistance to spur private portfolio investment in developing country infrastructure.3 To attract such investment, developing country governments need to “de-risk” their investment environments by removing regulatory barriers, privatizing assets, and providing subsidies and guarantees (Gabor 2021). However, evidence suggests investors simply gravitated toward safer investments in wealthier developing countries (Attridge and Engen 2019). To address their inability to reform the World Bank and IMF, developing country governments led by China founded two new multilateral development banks in 2014, the New Development Bank and the Asian Infrastructure Investment Bank. Additionally, governments accumulated large international reserves both individually and as part of regional financial agreements in order to manage balance-of-payments crises without having to turn to the IMF and its painful conditionalities. But perhaps the most significant change to emerge from the growing demand for infrastructure investment and the desire to avoid IMF and World Bank conditionalities is a flood of bilateral foreign aid and lending from new donors and lenders, with China leading the way. 2.1.3.3 Foreign States In contrast to the World Bank, IMF, and other Western organizations, new donors and lenders such as China and India do not include a string of policy conditions. Lending by China, for example, builds on a reputation as a donor with a strong respect for the sovereignty of other countries. As former President Festus Mogae of Botswana remarked, “I find that the Chinese treat us as equals. The West treats us as former subjects” (Smyth 2007). This sentiment stems from principles that Chinese aid has emphasized since Premier Zhou Enlai’s visit to Africa in 1964, including sovereignty, equality, and mutual respect. India’s aid program, which began in the 1950s, is similar with an emphasis on respect for 3

See https://thedocs.worldbank.org/en/doc/622841485963735448-0270022017/original/ DC20150002EFinancingforDevelopment.pdf (access May 15 2022).

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

Market Failures and China’s Chance to Lead

31

equality and mutual benefit, mutual noninterference in domestic affairs, and peaceful coexistence (Price 2005). These emerging lenders also exemplify successful alternative development paths that diverge from the policies advocated by the Washington Consensus and its successors. This has granted developing country governments the ability to choose among various infrastructure financing options. Likewise, bilateral lending programs grant lending governments greater flexibility in their lending decisions since they are not bound by the constraints inherent to MDBs that possess an assortment of stakeholders with the power to influence lending decisions. In bilateral lending arrangements, lending governments and affiliated organizations (SOEs) can act more autonomously so as to pursue projects and structure financing terms in ways that are more aligned with their own specific needs and preferences. Consequently, even though developing countries, such as China, may possess greater influence in the context of a MDB (such as the NDB or AIIB), they often opt for bilateral lending. 2.1.3.4 Chinese Infrastructure Financing and Development Characteristics In comparison to Western lenders, China’s infrastructure financing and development initiatives are distinguished by three characteristics. First, they do not come with conditionalities, enhancing the attractiveness of Chinese financing for governments reluctant to implement the kinds of liberalizing reforms espoused by Western organizations. As mentioned above, China seeks lending arrangements that are compatible with the needs of recipient country governments as the leaders of those countries view them. This then raises the question of what are the primary interests of recipient country leaders, which will be discussed further below. Second, Chinese SOEs and state-owned agencies (e.g., China Development Bank and the Export-Import Bank of China) are the entities primarily involved in implementing these financing arrangements for construction projects which increases the potential for and value of government-to-government discussions. For example, if the project or a group of projects are considered sufficiently important to each country’s national interest, visits and discussions between the two countries’ leaders may occur, as happened with Djibouti, Malaysia, and Greece (discussed in Chapter 8). Chinese state-owned entities stand ready to swiftly implement any agreements reached by government officials. Third, Chinese financing agreements are typically opaque. From China’s perspective, this is beneficial insofar as it reduces public scrutiny and accompanying critiques from Western organizations. Opacity also helps China weaken the ability of recipient governments from making comparisons to other country agreements in order to strengthen their

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

32

China’s Chance to Lead

own negotiating position. Although opacity promotes market failures and ought to be disfavored by recipient governments from the perspective of maximizing a nations’ aggregate welfare, host country leaders may strongly desire opacity if the funds are to be used for projects that benefit some groups (e.g., political leaders’ allies and supporters) rather than others (political opponents). Given that China does not question or impose conditions on the terms of their lending provisions, political leaders of recipient countries can exert considerable discretion in deciding which domestic actors are most likely to benefit. Although developing countries share many of the same types of market failures, they display varying levels of interest in participating in the BRI. To explain this variation, it is necessary to identify the primary motivation for governments to seek infrastructure financing and the conditions under which China would be an attractive option. To determine China’s compatibility with the interests of potential recipient governments, it is necessary to identify its motivations and capacity to address foreign countries’ infrastructure financing needs. 2.2

China’s Motivation and Capacity to Address Infrastructure Financing Needs

Although China may be the most preferred source of infrastructure financing for many developing countries, we must also consider what motivates China to provide such abundant financing and related resources devoted to infrastructure development. But even if China possesses the motivation to provide abundant resources, it may lack the capabilities to fulfill those ambitions. Thus, it is also necessary to consider China’s capacity for infrastructure financing and development. I therefore discuss China’s motivation and capacity for initiating infrastructure financing and development in the context of the BRI in turn. 2.2.1 Motivation China has several motivations for initiating the BRI. First, transportation costs for shipping goods will be reduced which is of great interest to China, especially since it became the world’s largest trading nation in the world in 2013.4 These benefits will also help countries beyond the designated BRI routes since their exporters will also use the upgraded infrastructure. 4

Note that I frame this discussion of China’s motivations and capacity with regard to when it initiated the BRI at the end of 2013.

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

Market Failures and China’s Chance to Lead

33

Second, China is attempting to reduce the economy’s dependence on domestic infrastructure investment to fuel growth. The motivation to reduce this dependence was amplified following China’s delivery of one of the largest stimulus packages in recent economic history to prevent the economy falling into recession during the global financial crisis. A lasting side effect was the creation of massive excess capacity in many industrial sectors from steel to cement, often dominated by SOEs. For example, China’s annual steel production jumped from 512 million tons in 2008 to 803 million tons in 2015. The extra 300 million tons exceeded the combined production of the United States and the EU.5 Excess capacity would squeeze profits, lead to massive layoffs, elevate debt levels, and make the country’s financial system vulnerable. To address the excess capacity problem, the BRI would help Chinese construction companies, equipment suppliers, and other businesses that thrived on the country’s building boom to look to foreign markets for new opportunities. While part of the initial motivation was to shift the operations of Chinese producers of infrastructure materials and equipment to other BRI markets, recipient countries were often hesitant to welcome such investment due to a desire to protect their own domestic firms. Consequently, attention shifted to requiring recipient countries to use Chinese manufactured materials and workers. For example, a study of loan practices by the China Development Bank and the Export-Import Bank of China in 2013–2015 showed that 70 percent of overseas credit was made on the condition that at least part of the funds be used to purchase Chinese equipment and involve Chinese labor (Kynge 2015). Thus, an important proximate motivation for the BRI was to address China’s excess capacity problem in its infrastructure development sectors, though in the longer term it would facilitate a shift away from infrastructure investment-led growth. Third, the government has sought to improve market mechanisms so as to increase total factor productivity. Ongoing efforts to improve market mechanisms that provide for more efficient capital allocations can be accomplished by encouraging less competitive firms to exit and by reducing state support for inefficient firms (OECD 2018).6 These issues 5

World Steel Association data, https://worldsteel.org/wp-content/uploads/2015-WorldSteel-in-Figures.pdf (accessed 8 January 2022). 6 In his speech to the 19th Party Congress, President Xi explicitly directs attention to the need for more effective market mechanisms when discussing China’s development strategy: “We should pursue supply-side structural reform as our main task.… We should endeavor to develop an economy with more effective market mechanisms, dynamic micro-entities, and sound macro-regulation. This will steadily strengthen the innovation capacity and competitiveness of China’s economy” (Xi, J., 2017b, page 26) and quoted in OECD (2018).

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

34

China’s Chance to Lead

are difficult to address in countries where SOEs play an important role since market discipline may be reduced. The implementation of the BRI aims to create new markets, facilitate trade as well as investment, shift production capacity to where there is ready demand or where production factors are cheaper. This would enable the government to reduce excess capacity, concentrated primarily in SOEs, and promote a transition away from infrastructure-led investment growth to more manufacturing and services-oriented growth. Fourth, China’s western and central regions have lagged behind the coastal provinces in terms of GDP per capita. For example, the coastal metropolis of Shanghai is about five times wealthier than the inland province of Gansu, which is part of the old Silk Road.7 Despite concerted government efforts to promote the development of China’s western provinces over the past two decades (the twelve western provinces account for 70 percent of the country’s land area and one-third of its population), such as the “western development strategy” implemented in 1999, their contribution to gross domestic product (GDP) only increased by about 4 percent to slightly above 20 percent.8 One acute side effect of heavy state subsidies in these western provinces has been a high concentration of state-owned enterprises and low penetration of private firms. To invigorate the development of these economies and attract private investment, the BRI can be used to promote the integration of Western provinces into wider regional economies (e.g., Pakistan, Central Asia, Southeast Asia, and Russia). Fifth, the BRI can help China secure access to energy imports through the construction of new pipelines and deep-water ports. China’s energy demand increased more than 500 percent since 1980, making the country the largest energy consumer and, as of 2014, the largest net importer of oil in the world. However, around 80 percent of China’s oil imports (and over 60 percent of China’s total maritime trade) pass through the Straits of Malacca, making the country vulnerable to a blockade in the event of a crisis (CSIS 2020). Sixth, infrastructure development of foreign countries may promote their economic growth, in turn elevating demand for China’s goods and services. For example, infrastructure development of ports, railways, and industrial parks may in turn attract additional foreign investment from both China and other countries. At the same time, infrastructure 7

See National Data (stats.gov.cn) (accessed March 12 2022). 8 See www.china-briefing.com/news/investing-in-chinas-western-provinces-how-to-readthe-new-encouraged-catalogue/#:~:text=Over%20the%20last%20two%20 decades,percent%2C%20according%20to%20government%20data (accessed March 2022).

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

Market Failures and China’s Chance to Lead

35

spending can promote the deployment of Chinese digital technologies that can lock the recipient country into Chinese standards to the longterm benefit of Chinese firms. This will be discussed further in the next section titled “Transaction Costs and Coordination Failures.” Finally, in the long-term the BRI can help China’s quest for greater international stature for the renminbi as a global reserve currency. Emerging markets that have experienced their own currency volatility have backed China’s efforts, with 25 countries signing bilateral swap agreements as of 2014. By 2014, RMB cross-border trade settlements accounted for 22 percent of China’s trading volume. With the aim of financing projects with loans denominated in Chinese currency, China founded the Asian Infrastructure Investment Bank in 2014 and joined the European Bank for Reconstruction and Development in 2015. These efforts contributed to the IMF adding the renminbi to the basket of global currencies that make up the Special Drawing Right, or SDR, in 2016. 2.2.2 Capacity In addition to several motivations for pursuing BRI projects, China’s capacity to implement them influences whether and how they will be completed. I discuss four main capacity issues. First, the unprecedented growth of China’s economy has been accompanied by persistent trade surpluses, contributing to large foreign currency reserves. In 2014, China’s foreign exchange reserves were the highest in the world, at US$3.95 ­trillion, accounting for one-third of the world’s total.9 Most of these reserves are invested in dollar-denominated assets, principally US treasury securities. China’s surplus dollar reserves offer an abundant source of capital that China can tap to fund BRI projects. Abundant dollar reserves are important because BRI countries will remain unable to obtain adequate RMB funds to finance infrastructure projects so long as China retains restrictions on the capital account and currency convertibility (Liang 2020). Moreover, China’s current account surpluses with BRI countries and the inadequate development of the RMB-denominated bond market for the sale of Panda bonds (RMBdenominated bonds issued in the Chinese onshore market by issuers incorporated outside China) further limit the ability of BRI partner countries to obtain adequate RMB in order to meet their RMB debt obligations. Thus, in the absence of the availability of RMB to finance BRI projects, abundant dollar reserves are necessary. This, of course, poses risks given the long investment horizon of BRI projects and the 9

See www.safe.gov.cn/en/2014/0707/1121.html (accessed March 14 2022).

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

36

China’s Chance to Lead

potential for currency mismatch (e.g., Chinese banks lack access to dollars to continue dollar-denominated lending midway through a project or BRI countries lack adequate dollar reserves to repay their loans in dollars). It requires China to retain adequate US dollar reserves and to continue to accumulate dollars while managing the value of the RMB. Second, given the need to tap dollar reserves to finance BRI projects, the primary funders for the BRI are state-owned financial organizations controlled by Beijing. These state-owned banks together accounted for 81 percent (US$765 billion) of BRI funding as of the end of 2018 (He 2020, Figure 6). They include two policy banks, the China Development Bank (26 percent of the total) and the Export-Import Bank of China (19 percent), which account for 45 percent of the total, with the four big state-owned commercial banks making up the remaining 36 percent, including the Bank of China (17 percent of the total), the Industrial and Commercial Bank of China (15 percent), the China Construction Bank (2 percent), and the Agricultural Bank of China (2 percent). Chinese SOEs provide the next largest tranche of financing, with capital raised via equity financing in China’s capital markets (9 percent of the total). This is followed by the sale of bonds for BRI projects by China’s financial institutions (the state-owned policy and commercial banks) and nonfinancial SOEs (4 percent of the total). The next main provider of financing is the Silk Road Fund which was established by the Chinese government in 2014 as a state-owned fund exclusively focusing on investments for the BRI. By the end of 2018, it accounted for only 2 percent of total funding. Chinese government-sponsored bilateral funds also account for about 2 percent of total funding. Officially, these funds are jointly funded, but in practice China’s state-owned banks are the major sponsors. There are nearly a dozen of these funds targeting projects in different regions, such as the China-Africa Fund for Industrial Cooperation, and the ChinaASEAN Investment Cooperation Fund (He 2020; Table 7). Finally, multilateral financial institutions finance approximately 2 percent of the total. These include the Asian Infrastructure Investment Bank, the New Development Bank, as well as funds established by the People’s Bank of China in cooperation with regional development banks such as the African Development Bank. While the proportion of total financing is small for many of these lenders, the nominal amounts are not insignificant. For example, just 1 percent of total financing equates to USD7.6 billion. Third, Chinese SOEs involved in infrastructure development have acquired massive economies of scale and scope as well as considerable experience implementing large, complex projects. The development of these industries and companies has occurred over a long period of time, with huge investments that would not be achieved by the market

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

Market Failures and China’s Chance to Lead

37

alone, especially in the context of a developing country. The capacity of these SOEs to engage in infrastructure development in BRI projects was explicitly targeted as a goal of China’s SOE reforms following the 18th Party Congress in 2012. This involved merging the country’s centrally controlled SOEs to promote their competitiveness in fulfilling the BRI, to reduce surplus capacity, and to advance “supply-side reforms” in the domestic market (i.e., facilitating firm exits from industries and/ or bankruptcies to promote more efficient capital allocation). Consequently, the number of central SOEs dropped from 189 in 2002 to 96 at the end of 2018. Since 2013, more than 70 Chinese SOEs have made it onto the Fortune Global 500 list. Few other developing countries are capable of producing infrastructure companies with comparable size and capabilities. Fourth, the heavy reliance on Chinese SOEs for both the financing and development of infrastructure projects in BRI countries bestow certain advantages (and disadvantages) typically unavailable to Western organizations. First, SOEs are capable of engaging in nontransparent bidding, negotiations, financing, and construction. This is advantageous in terms of enabling Chinese firms to reach agreements with recipient governments that might not occur otherwise. On the one hand, Chinese firms have an incentive to engage in opaque processes so as to preserve the fragmentation of borrowers and thereby retain their own bargaining power, as well as to reduce accountability to external stakeholders and western critics. At the same time, such opacity may be favored by recipient governments that seek to shield the allocation of BRI financing from public scrutiny (e.g., by domestic political opponents). Second, Chinese SOEs monitored by Beijing will be more responsive to addressing the central government’s political objectives. While SOE reforms over the past two decades have increased the emphasis on profitability, this does not equate to profit maximization. Depending on the project, the government may exercise greater pressure to implement it at the expense of profitability. In other words, profitability may be constrained by the pursuit of political goals. For example, Chinese enterprises gave up the protection of government guarantees from receiving countries for their investment in high-speed rail projects in Malaysia, Thailand, and Indonesia in order to win contracts over highly competitive Japanese companies (He 2020). Third, the involvement of central SOEs in BRI projects enables government-to-government negotiations that can allow for a larger set of issues to be considered in negotiations than would occur between a private company and a foreign government. For example, governments could agree to sales of military hardware in exchange for payment terms concerning an infrastructure project. Fourth, SOEs can

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

38

China’s Chance to Lead

often provide financing at a lower rate than comparable private firms due to the backing of the government, with correspondingly lower sovereign borrowing rates. However, this can easily lead to moral hazard as these entities then engage in risky lending with the expectation the government will bail them out. Another important disadvantage associated with these special advantages of SOEs is the unwillingness of private firms to collaborate with them in these projects. In the event of disputes, which often occur in the context of large, long-term, and complex infrastructure projects, private Chinese firms are at a considerable disadvantage to state-owned entities. Moreover, the prioritization of political objectives above profit maximization creates substantial risks for private firms that cannot count on the government to save them. Opacity in the bidding, negotiations, financing, and construction of these projects tends to magnify the risks for private firms. 2.3

Transaction Costs and Coordination Failures

A second important market failure plaguing developing countries’ ability to attract financing for major infrastructure investments is transaction costs that impede international trade, investment, and information flows. In this section, I first discuss how coordination on common standards can reduce transaction costs and how this creates an opportunity for China. I next turn to a discussion of the specific technologies, industries, and markets (and related products and services) in which China can play a leadership role in overcoming coordination failures. I then examine why the moment is ripe for claiming leadership in developing and promoting the adoption of standards in these technologies, industries, and markets due to macro trends in the global economy and in the current stage of technological progress and innovation. Next, I discuss why China is the country to help solve these market failures (reducing transaction costs via standardization) and why the BRI is the way do this in comparison to alternative approaches such as working through international standards development organizations. 2.3.1

Coordination Failures and China’s Chance to Lead

While infrastructure financing gaps involve political solutions to market failures that yield Nash equilibria that are often Pareto suboptimal, coordination at the international level is about reducing transaction costs between countries by moving them collectively, though to varying degrees, toward the Pareto frontier. The point on the Pareto frontier they move toward depends on the distribution of power between countries

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

Market Failures and China’s Chance to Lead

39

being either asymmetrical or symmetrical. Where asymmetrical power prevails, as with de facto standards due to market power (e.g., physical infrastructure such as railways), China standards are likely to prevail. But for technology standards, which more frequently are decided by formal negotiations in the context of professional associations (e.g., the International Telecommunications Union), as with 5G standards or artificial intelligence, it is far less likely that Chinese standards will be adopted. However, the spread of Chinese technology standards by incorporating them into hard infrastructure (railways, ports, logistics) allows China to point to the extensive adoption of Chinese technologies as indicators of the high quality of their standards, bolstering their case in international standards deliberations and strengthening the global marketability of their products.10 In the context of a national economy, Arrow (1969, p. 516) suggests transaction costs could be overcome by collective action, either in the form of firms or as social rules and norms where “norms of social behavior, including ethical and moral codes … are reactions of society to compensate for market failures.” But at the international level, in which such social rules and norms are typically absent, a hegemonic power such as the United States typically plays a critical role in promoting coordination so as to reduce transaction costs, such as varying standards used for physical infrastructure (e.g., railway gauges), customs and duties regulations at border crossings, as well as information asymmetries (e.g., opacity regarding infrastructure tenders or transportation subsidies or even railway schedules). The growth of transaction costs due to rising infrastructure needs across developing countries creates opportunities for an actor with the de facto (market) power to establish common standards and reap the gains. At the international level, China can act as a focal point on which businesses and governments coordinate. By providing infrastructure across numerous countries, China can play an important coordinating role that is beneficial to both private capital and political rulers within individual countries on a bilateral and multilateral basis. This can occur with regard to coordinating projects across different countries, such as building an industrial park with a road or railway that passes through another country which then connects to a port (e.g., Ethiopia and Djibouti). China can also promote coordination of standards to reduce shipping costs, as with railway gauges, customs clearances, and logistics. Coordination can also involve common technology standards and the collection of data 10

See Chen et al. (2018), www.uscc.gov/sites/default/files/Research/SOSi_China’s%20 Internet%20of%20Things.pdf (accessed March 14 2022).

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

40

China’s Chance to Lead

on a shared digital platform, yielding benefits via network effects for all participants, such as improving the efficiency of transportation networks. To identify China’s emerging and future potential to play an international coordinating role in reducing transaction costs, we must identify: (1) the technologies, industries, and markets that stand to benefit from standardization; (2) why the contemporary period is an opportune moment for establishing such standards; (3) why China is well positioned to capture this opportunity; and (4) why the BRI is the means to do this. 2.4

Which Technologies, Industries, and Markets Would Benefit from Standardization?

Technical standards have long facilitated integration of markets, serving as the foundation upon which industries, services, and products are interconnected locally, regionally, and across the globe. Technical standards refer to the specifications by which processes and technologies are designed to improve the quality, security, and interoperability of various goods and services. Standardizing processes and technologies that may otherwise hinder the movement of goods and information allows for transaction costs to be reduced. Railways illustrate the opportunities available through the promotion of a uniform set of standards that can reduce transaction costs such as operational, regulatory, financing, and technical costs (Nitsche 2020; Wagener et al. 2020). Operational transaction costs are due to the lack of systematic design and coordination of railways across countries. For example, there are over sixty cities operating westbound block trains from China to Europe. All these routes are managed by local governments. The lack of systematic top-level design and regional coordination contributes to inefficiency in terms of railway capacity utilization and resource allocation, with volatile and uncertain transportation times. Additionally, different gauges, train length regulations, and inadequate or aging railway infrastructure to permit feeder connections off the main China–Europe route elevate these costs. For example, the difference in railway track gauges between former USSR countries (1,520 mm), China (1,435 mm), and Western Europe (1,435 mm) requires the transshipment of cargoes or the exchange of bogies (the undercarriages) at border crossing stations. Transaction costs are also increased due to regulatory, legal, and financing obstacles. For example, border and customs formalities, the absence of standardized shipping documents and nontransparent laws and regulations can all slow down shipping times. Financing issues can stymie the lowering of transaction costs if the costs of upgrading

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

Market Failures and China’s Chance to Lead

41

infrastructure exceed their financing capacity. For example, Chinese provincial governments subsidize block train costs from under 50 percent to about 75 percent of the unsubsidized costs in an effort to promote regional development (World Bank, 2019). However, these subsidies are being reduced over time in order to make block trains more economical. The growing traffic and the need for more efficient operations require a large infrastructure investment that exceeds the financial resources available to most investors or governments, especially of smaller countries. A very promising way of reducing many of these costs is through the widespread deployment of digital technologies across global supply chains, which involve a large number of participants such as manufacturers, forwarders, shippers, customs agents, insurers, and customers. For example, containers can be equipped with sensors to locate the GPS position of the container, as well as the temperature, intrusions, and shocks. Blockchain technology offers the possibility for full data interoperability by enabling paperless information exchange between all parties involved, instant financing, and end-to-end tracking of container movement in a trusted and secure manner. Costs are thereby reduced by eliminating intermediators and business processes optimized. Indeed, digital technologies and their affiliated standards hold tremendous promise for reducing transaction costs and other types of market failures as smart manufacturing, smart infrastructure, and smart cities spread across the global economy. Smart Manufacturing is at the core of the Industry 4.0 concept. The term Industry 4.0, coined in 2011 by a German initiative of the federal government with universities and private companies, refers to adaptable manufacturing systems in which production processes automatically adjust for multiple types of products and changing conditions. This allows quality, productivity, and flexibility to increase while also enabling the production of customized products at a large scale and in a sustainable way with better resource utilization. From a technological evolution perspective, there are four stages widely identified as industrial revolutions (Kagermann et al. 2013). The first three took around two centuries and are associated with: (1) the introduction of water and steam-powered mechanical manufacturing facilities; (2) the application of electrically powered mass production technologies through the division of labor; and (3) the use of electronics and information technology (IT) to support further automation of manufacturing. Industry 4.0 technologies that address operational and market needs can be thought of as “front-end technologies,” including new manufacturing activities based on emerging technologies (Smart Manufacturing), the way products are offered (Smart Products), novel delivery methods

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

42

China’s Chance to Lead

of raw materials and products (Smart Supply Chain), and the new ways workers perform their activities based on the support of emerging technologies (Smart Working) (Frank et al. 2019). “Base technologies” provide connectivity and intelligence for front-end technologies by allowing the latter to be connected in a complete integrated manufacturing system. Base technologies include the Internet of Things (IoT), cloud services, big data and analytics, and 5G telecommunications infrastructure. Smart infrastructure refers to the integration of connected sensors and big data analytics with physical infrastructure to enable real-time monitoring, efficient decision-making and enhanced delivery (Weiss 2009; Ogie et al. 2017). The potential benefits of smart infrastructure include decreased maintenance costs (monitoring of critical infrastructure such as bridges, tunnels, and dams), reduced damage and disruption costs (traffic congestion and power blackouts), increased quality and value of service (on-demand use, flexible tariffs, automatic toll collection), as well as protecting human life (fewer road accidents, more rapid response to disasters, waste and water services). Together, these benefits can contribute to sustainable urban growth (Morimoto, 2010). As with smart manufacturing, smart infrastructure depends on base technologies, such as cloud services, big data and analytics, and 5G telecommunications infrastructure. The advent of smart manufacturing and smart infrastructure together contribute to the emergence of smart cities, which offer the most comprehensive application of new digital technologies and standards. Harrison et al. (2010) define a Smart City as an urban area that exploits operational data (data extracted from traffic, power consumption, etc.) in order to optimize city operations. They stress the importance of three features of the city: (1) the near-real-time data obtained from physical and virtual sensors; (2) the interconnection between different services and technologies inside the city; and (3) the intelligence from the analysis of the data and the process of optimizing and visualizing it to improve the sustainable economic growth and quality of life of its residents. Smart city technologies can be placed into four broad categories, including business, citizen, environment, and government-related domains (Yin et al. 2015; Sanchez-Corcuera et al. 2019). Businessrelated domains extend from agriculture to entrepreneurship, logistics, and transactions. Agriculture refers to the transformation of agricultural systems to ensure food security while accounting for climate change as growing urban populations reduce the availability of land used for crops and agriculture. Smart city entrepreneurship emerges from the new services and data generated by citizens that creates the opportunity for new companies to be created, especially in services and knowledge

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

Market Failures and China’s Chance to Lead

43

economies. Smart cities are also creating new opportunities in logistics via the implementation of complex processes based on cloud computing technologies to make the management of information easier, which reduce the cost of logistics management systems, and which consider the environmental impact such as pollution and noise. Finally, smart cities contribute to more efficient transactions, which are at the core of business processes. Blockchain technology offers the potential to simplify and automate them via its capacity to hold a distributed list of records using cryptographic techniques in a verifiable and permanent way. Citizen-related domains refer to a wide range of services extending from education and healthcare to traffic management. Smart city education provides citizens with twenty-first-century skills necessary to drive smart city innovations. Mobile devices that enable mobile health applications can provide remote monitoring and emergency assistance and tailor services to the needs of the disabled and elderly. Smart city traffic management systems can help reduce traffic congestion, the number of accidents, pollution, and time spent on the road. Environment-related domains include city planning, pollution control, renewable energy, smart electricity grids, and waste management. Smart cities with a set of interconnected devices to acquire data from the city and provide for its analysis would be useful for future urban planning and enhance energy efficiency, conservation, and the generation of renewable energy. Smart city architecture based on sensor and video analytics can measure the emissions produced by traffic, keep citizens informed of its condition, and suggest measures to reduce it. To promote the use of renewable energy, smart energy grids can enable bidirectional communication between residences and the general grid in order to lower power consumption and avoid large energy demand peaks. With regard to waste management, smart containers are capable of detecting their filling status and notifying trash collectors when they are nearly full, thereby promoting the optimization of city resources and reducing the environmental impact. Government-related domains include city monitoring, e-government, emergency response and public safety, and public services. Smart cities enable city monitoring due to pervasive and ubiquitous computing and digitally instrumented devices built into the very fabric of urban environments (e.g., fixed and wireless telecommunication networks, digitally controlled utility services and transport infrastructure, sensor and camera networks, building management systems) in order to monitor, manage and regulate city flows and processes. Smart cities would also enable E-government which can enable city stakeholders, particularly citizens, to efficiently participate in decision-making processes. Emergency response

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

44

China’s Chance to Lead

aims to reduce the harmful effects of all hazards, including natural disasters, road traffic issues such as congestion or accidents and speed the response times of emergency services such as ambulances and police cars. Public services could also be improved via smart cities that enable the co-creation of urban apps based on the collaboration of different stakeholders (citizens, private companies, research institutes, and public administrations). 2.5

Why Is the Time Ripe for Standardization?

A critical base technology enabling smart manufacturing, smart infrastructure, and smart cities is 5G telecommunications infrastructure. In comparison to previous advances in wireless technology, 5G promises far greater reach and significantly faster computing speeds and responsiveness. It will radically expand the number of devices and services that are interlinked. Hans Vestberg, CEO of telecom giant Verizon Communications, has remarked that 5G will “transform people, businesses and society” on the scale of a new Industrial Revolution.11 Instead of building cell towers located several kilometers apart, 5G technology will require installing hundreds of thousands of base stations every few hundred meters yielding huge infrastructure investment opportunities. For example, McKinsey forecasts that 5G deployments will cover around 25  percent of the global population by 2030, at an estimated cost of USD 700 billion to USD 900 billion.12 Countries that lead the way in the deployment of 5G infrastructure and in deciding the standards used will see huge economic benefits and could dictate how the technology is used around the world. Indeed, the types of technical standards used to ensure interoperability among an ever-growing ecosystem of connected objects and social networks will become increasingly important. They will not only define the technical parameters by which future technologies develop, but also who will be best positioned to reap the benefits. The adoption of bundled technologies (e.g., via integrated projects that combine multiple technologies) offers a particularly effective method to deliver an advantage for the companies and countries that produce them due to locking-in the technical standards that make it difficult and expensive to switch to an 11

“The Fourth Industrial Revolution Will Be Built on 5G, Says Verizon CEO,” Technology Breaking News, Jan. 11, 2019, https://tinyurl.com/y6z52vfc (accessed March 14 2022) 12 See www.mckinsey.com/~/media/mckinsey/industries/technology%20media%20and%20 telecommunications/telecommunications/our%20insights/connected%20world%20 an%20evolution%20in%20connectivity%20beyond%20the%205g%20revolution/mgi_ connected-world_discussion-paper_february-2020.ashx (accessed March 14 2022).

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

Market Failures and China’s Chance to Lead

45

alternative standard even if it is superior (Russel and Berger 2021). An important vehicle to export and lock in technology standards is through an ecosystem of smart projects that are complementary such as smart ports, railways, industrial parks with smart manufacturing, and smart cities. First-movers thereby gain an important advantage by locking-in their own standards and by locking-out competitors. The advent of Industry 4.0 has sparked a race on the part of governments and firms worldwide to research and develop new applications that can give their companies and economies a competitive advantage. For example, shortly after the introduction of the smart manufacturing concept by Germany in 2011, the US government began a series of nationallevel discussions, actions, and recommendations, titled the “Advanced Manufacturing Partnership,” to ensure the United States would be at the forefront of the next generation of manufacturing (Rafael, Shirley, and Liveris 2014). This was followed by similar initiatives by the French and UK governments in 2013, South Korea and the European Commission in 2014, and Japan and China in 2015 (Liao et al. 2017). Industry-wide initiatives were also launched, often in collaboration with universities and research centers. The rapid increase in the number of well-resourced and highly motivated actors competing to develop the next generation of technologies and affiliated standards speaks to the very high stakes. Competition to decide the next generation of technology standards is important not only to domestic national economies, but also to globally competitive manufacturing via global value chains (GVCs). These refer to the division of labor among firms and countries that participate in different stages of production, from basic assembly on one end through to product development and branding on the other. The standard-setting and IP-owning firms sit at the top of the GVC hierarchy, setting prices and creating product specifications. They are not heavily engaged in the physical production of goods, focusing instead on intangible production. Regardless of whether the specifications are proprietary or open-source, technological standards set the parameters for components manufacturers and final product assemblers that perform specialized manufacturing tasks. Standard-setting firms embed themselves in GVCs by establishing arms-length relationships with suppliers, but also by ensuring quality consistency, reliability of component and assembly manufacturing supply chains, and creating managerial consistency along these supply chains (Von Hagen and Alvarez 2011). The goal is for a company to have its patented technical solution to be adopted as an international standard, which entitles them to earn revenue by licensing this standards-essential patent (SEP). Revenue from SEP licensing can be substantial. For example, Qualcomm earned

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

46

China’s Chance to Lead

around $8 billion from licensing SEPs in 2021. Given the high stakes in terms of potential revenue, the significant implications for which technologies will dominate future markets, as well as the power to influence the trajectory of technological innovations together contribute to significant competitive pressures. First movers capable of setting the standards reap tremendous rewards where those that lag behind must bear the costs of adaptation. As Werner von Siemens said in the late 1800s, “he who owns the standards, owns the market” (Koch 2017). Competitive pressures to set technical standards are greater for faster growing markets, such as emerging economies. Since the turn of the century, emerging economies have experienced GDP growth at a rate more than twice that of advanced economies, with expectations that this pace will continue for the foreseeable future (IMF World Economic Outlook, April 2022). This sustained rapid growth has contributed to emerging economies now accounting for nearly two-thirds of global GDP (PPP adjusted), having switched positions with advanced economies since 1990 when the latter accounted for two-thirds of global GDP. A subset of emerging economies is poised to experience a demographic boom over the next several decades, elevating their potential growth and the size of the opportunity for companies that succeed in setting their technical standards. These countries will be home to large youth populations better prepared to engage with the variety of smart technologies discussed earlier. Over the next thirty years, the UN expects the global population to increase by 2 billion people, from 7.7 billion currently to 9.7 billion in 2050.13 Nine countries will make up more than half the projected growth, including India, Nigeria, Pakistan, the Democratic Republic of the Congo, Ethiopia, Tanzania, Indonesia, Egypt, and the United States (in descending order of the expected increase). From a regional perspective, the population of sub-Saharan Africa is projected to double by 2050 (99 percent increase), followed by Northern Africa and Western Asia (46 percent), Central and Southern Asia (25 percent), and Eastern and South-Eastern Asia (3 percent). A final demographic shift placing additional pressure on companies and governments to win the standards war is urbanization rates. According to the United Nations, these rates are projected to be highest in Africa and Asia followed by Latin America and the Caribbean up to 2050, increasing by about 20 percent in Asia and Africa, follow by Latin America and the Caribbean at around 10 percent. This large urban migration will contribute to an increase in manufacturing-related 13

The World Population Prospects 2019: Highlights, which is published by the Population Division of the UN Department of Economic and Social Affairs.

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

Market Failures and China’s Chance to Lead

47

industries that are integrated into Smart Manufacturing GVCs, in addition to growing demand for smart infrastructure and smart cities that can accommodate growing urban populations in a sustainable way. 2.6

Why Is China Well Positioned to Promote Standardization?

China is uniquely positioned to address the market failures due to a lack of coordination across regulatory and technical standards for two primary reasons. First, China possesses the motivation to address these issues for its own domestic political needs. Second, China possesses the capacity to address these needs. Together, China’s domestic political priorities combined with its particular capacity characteristics influence how it addresses market failures, as manifested by the manner in which the Belt and Road Initiative is being implemented. I discuss these topics in turn. 2.6.1 Motivation To retain its hold on power, the Chinese Communist Party (CCP) has sought to maintain social stability with economic growth the primary factor contributing to this end since opening up in 1978. The effort to implement regulatory and technical standards across countries via the BRI is best understood as a means to achieving this end. During the 19th Central Committee Meeting of the Communist Party of China in 2020,14 President Xi Jinping affirmed his commitment to maintaining a high growth rate, stating that it is perfectly possible to double the size of the economy and per capita income by 2035 (Sasaki et al. 2021). Over the past four decades, China sustained remarkable average annual growth of 10 percent while transforming from a rural agricultural society to an urban industrial one, and from a planned to a market-based economy. However, since the global financial crisis in 2008, China has experienced a sharp slowdown in growth in output per worker. In 2015– 2018, average GDP growth fell below 7 percent for the first time since 1991, to a large extent due to slowing growth in total factor productivity (TFP) (Brandt et al. 2020; Sasaki et al. 2021). Aggregate TFP growth slowed from 2.8 percent in the ten years prior to the global financial crisis to 0.7 percent in 2009–2018. One clear way to improve productivity growth is for China to foster innovation. By some measures, China’s innovation capacity has improved 14

At this important meeting of the CPC, the five-year plan for 2021–2025 and the longterm goals for 2035 were discussed.

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

48

China’s Chance to Lead

steadily in recent years, placing the country 14th on the Global Innovation Index (Dutta et al. 2020). However, China remains quite distant from the global technology frontier and thus has substantial remaining potential for catch-up growth (Brandt et al. 2020). For example, China’s per capita GDP is still only about 20 percent of that of the United States. If China follows the catch-up process that other East Asian economies have followed (Japan, South Korea, Taiwan, and Singapore), it will be possible for China to double the size of its economy and per capita income by 2035 (Sasaki et al. 2021). However, China differs in three important ways from the East Asian Tigers that reduce the likelihood of achieving this potential. First, China has a large population and the government has set a goal of a food selfsufficiency rate of 95 percent in order to ensure food security. Improvements in productivity growth depend on the movement of labor across industries (from agriculture to manufacturing and/or services) as well as improvements to labor productivity in each sector. To sustain a food self-sufficiency rate of 95 percent while doubling the size of the economy, either labor productivity in agriculture must be increased to allow labor to move to manufacturing or services industries or labor productivity in the manufacturing and services sectors must grow at a rate well above the rate achieved by the East Asian Tigers, or some combination of both of these. Second, China must contend with an aging population which could affect both the reallocation of labor across industries (from agriculture to manufacturing or services), labor productivity within industries, and contribute to a decline in the savings rate. At the same time, the size of the Chinese economy and capital controls in place will likely require domestic investment to be financed by domestic savings, which is how the other East Asian Tigers financed capital deepening during the catch-up process. For China, the slowdown in capital accumulation will require higher TFP growth in order to achieve catch-up (higher labor productivity). Third, China already accounts for a large share of global manufacturing, reducing the ease of increasing its manufacturing-related exports. According to estimates by the Bank of Japan (Sasaki et al. 2021), it is likely that the GDP share of China’s manufacturing sector will remain relatively unchanged from 28 percent in 2020 to about 26 percent in 2035 as it seeks to achieve its growth target. Over this time, value added in the manufacturing sector will increase by about 1.8 times, contributing to a further increase in its current global manufacturing share of about 25 percent in 2020. However, tensions between the United States and China, the declining growth rate of global trade, and its already

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

Market Failures and China’s Chance to Lead

49

large share of global manufacturing-related exports suggest it will not be possible to match the high export-led growth of the East Asian Tigers in their catch-up process. To compensate for this, the Chinese government could stimulate domestic demand by continued urbanization with a corresponding rise in incomes. In fact, the Chinese government aims to increase the urbanization rate from 60.6 percent in 2019 to 65 percent by 2025 during the period of the 14th Five-Year Plan (2021–2025). However, increasing urbanization while also maintaining agricultural output levels and simultaneously increasing China’ global share of manufacturing exports requires raising labor productivity. Alternatively, or at the same time, China could raise its share of global manufacturing exports by substituting export growth from advanced economies to other emerging economies. These three challenges require the Chinese government to prioritize investments into innovation in order to enhance TFP growth. This was emphasized in President Xi’s speech to the 19th Party Congress in 2017: We need to raise total factor productivity and accelerate the building of an industrial system that promotes coordinated development of the real economy with technological innovation, modern finance, and human resources. We should endeavor to develop an economy with more effective market mechanisms, dynamic micro-entities, and sound macro-regulation. This will steadily strengthen the innovation capacity and competitiveness of China’s ­economy. (Xi 2017, page 26)

One tool to promote innovation was announced in 2015 with the Made in China 2025 (MIC 2025) plan that aims to move Chinese electronics manufacturers up the chain of value creation. There are two parts to this plan. First, MIC 2025 promotes the development of smart manufacturing capabilities as part of a Chinese version of the Industry 4.0 initiative. Second, it seeks to bolster Chinese firms’ competitiveness in controlling the patents, brands, and data that allow top global firms to generate revenue from global production without directly engaging in physical manufacturing. In the context of global value chains, grabbing a larger share of output does not mean physically producing “core components,” but rather possessing the right to do so to the exclusion of competitors. As mentioned earlier, standards stand at the core of achieving sustainable competitive advantage in global value chains. The long-term objective of MIC 2025 is to place Chinese firms at the apex of global value chains by helping them to develop intellectual and creative design capabilities indicative of American MNCs while also retaining the specialized manufacturing capabilities that are a hallmark of German and Japanese firms.

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

50

China’s Chance to Lead

But China’s promotion of setting technical standards is not only about winning the standards war in the global marketplace. It is also a tool that the government can use to force the upgrading of the country’s industrial base. In contrast to the standard setting processes dominated by private actors in the United States and EU (Büthe and Mattli 2011), the Chinese government plays a far more active role in setting technical standards high, thereby forcing manufacturers to upgrade their processes. As one Chinese standards official described it, “Our situation is very different from yours. We need better quality and therefore higher technical standards. It is a national requirement and not a matter of company decisions.”15 2.6.2 Capacity China has had a sustained commitment to R&D investment, which has grown at an annual rate of more than 10 percent for over a decade.16 China’s ratio of R&D investment to GDP is remarkably high for an emerging economy, ranking above the EU in 2018, and nearly doubling the ratio of Brazil.17 Coupled with this is China’s long-standing interest in developing technologies that are precursors to their current smart cities initiative as a vast population swiftly migrated from rural to urban settings. China’s contemporary smart cities initiative traces its origins to digital cities policies introduced in the 1990s. New digital mapping technologies, such as global positioning systems, geographic information systems, and remote sensing, were implemented to increase the data available to government policymakers. In the mid-2000s, municipal services and administration were brought into the digital age with new information technology hardware and software capabilities. China’s smart cities policies really began to resemble their current form with the 12th Five Year Plan covering the 2010 to 2015 period. The document specifically encouraged the planning and development of ‘smart cities’ via standardization and interoperability of government information systems while also outlining plans to expand data collection and analysis capabilities. From 2012 onward, China’s Ministry of Housing and UrbanRural Development (MOHURD), China’s main government authority in charge of urban planning and management, has been selecting cities in which to experiment and implement smart cities initiatives. By 2015, nearly 300 MOHURD-sponsored localities had smart cities initiatives. 15 Quoted in Rühlig (2020). 16 Based on data released by the National Bureau of Statistics of China. 17 Based on data from the OECD and UNESCO, as reported in Sasaki et al. (2021).

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

Market Failures and China’s Chance to Lead

51

China’s 13th Five Year Plan, covering 2016 to 2020, spotlights the importance of smart cities to China’s economic development. It places renewed emphasis on the interoperability of municipal governance platforms and the adoption of a unified set of standards to enable the orderly development of smart cities. It promotes the need for the new generation of information technologies, including cloud computing, the Internet of Things (IoT), pervasive mobile networks, big data systems, and artificial intelligence to improve data gathering and analysis capabilities available for urban planning and governance. China has since made significant progress in developing and deploying the technologies needed to support smart cities projects with a vibrant ecosystem of private companies leading the way. With regard to China’s IoT industry, it has grown at an 81 percent annual rate between 2010 and 2018 compared to 60 percent worldwide.18 China’s IoT market was valued at 15 percent of the global market in 2018 (USD 209.8 billion out of USD1,400 billion for the global market). The leading companies involved in IoT and mobile infrastructure technologies include several of China’s largest private enterprises such as Huawei, Alibaba, and Tencent, surveillance specialists Hikvision and Dahua, and state-owned giants China Mobile and China Unicom. The relatively faster growth of the IoT industry in China is indicative of othersSmart cities–related technologies. For example, China’s big data market has grown at a 61 percent annual rate between 2015 and 2018 compared to 30 percent for the global market over the same time period.19 In 2018, China’s big data market accounted for 12 percent of the global market (USD 5.17 billion compared to USD 42 billion for the global market). Even more than IoT, Chinese private firms dominate this technology category including well-known names such as Alibaba, Tencent, and Huawei, as well as other private players such as Neusoft, Inspur, and supercomputer developer Sugon. 18

China’s IoT industry grew from approximately RMB 200 billion (USD 28.0 billion) in 2010, shortly after IoT was identified as a core development priority, to nearly RMB 1.5 trillion (USD 209.8 billion) in 2018. This is according to a 2018 report from the China Economic Information Service. Worldwide, the IoT market was valued at USD 240 billion in 2010 and USD 1,400 billion in 2018, equating to a 60 percent annual growth rate according to a survey conducted by Frost & Sullivan (source HKExnews). 19 China’s big data industry grew from approximately RMB 11.6 billion (USD 1.82 billion) in 2015 to RMB 32.9 billion (USD 5.17 billion) in 2018 according to iResearch and the China Academy of Information and Communications Technology, a think tank affiliated with China’s Ministry of Industry and Information Technology. The global big data market was valued at USD 22.6 billion in 2015, which increased to USD 42 billion in 2018 according to Wikibon and SiliconANGLE.

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

52

China’s Chance to Lead

With regard to cloud services, China’s public cloud market is about one-third the size of North America’s as of 2019 (USD 21 billion compared to USD 63 billion). But China’s market grew 142 percent between 2018 and 2019 compared to 17 percent for the North American market.20 Cloud computing technology leaders include well-known private companies Alibaba, Tencent, Huawei, and Baidu, as well as the stateowned telecommunications behemoth China Telecom. 2.6.2.1 Smart Cities and Artificial Intelligence In 2006, China announced the national policy priority of developing the artificial intelligence industry. It grew slowly but steadily over time but began to take off in the latter part of the 2010s contributing to an AI market value that exceeds the North American AI market. For example, between 2016 and 2018, China’s AI market grew 142 percent annually to USD 9.3 billion.21 This compares to an annual growth rate of 26.5 percent for the North American AI market, valued at USD 8.6 billion in 2018.22 This rapid growth is due, in part, to a State Council announcement in July 2017 of ambitious long-term growth targets for China’s AI industry with the aim of developing it into a global leader by 2030 worth RMB 1 trillion (USD 140 billion). To help address concerns that a shortage of high-end AI talent may limit progress in the future, more than 180 Chinese universities were approved by China’s Ministry of Education to set up new AI-related majors for undergraduate students by 2019, the fastest-growing discipline in the country.23 In 2017, President Xi Jinping also handpicked the Xiong’an New Area, a site in Hebei province about 100 kilometers southwest of Beijing, to be developed into China’s “city of the future.” It is being designed as a sustainable smart city that is expected to attract research, development, and manufacturing firms in high-tech industries, including information technology, biotechnology, and new materials. Smart city technologies such as roads designed for smart cars and renewable energy infrastructure are being used to augment its attractiveness as a research and manufacturing hub. A 2019 planning document issued by the Central Party Committee 20

Data for China’s cloud services market comes from the China Academy of Information and Communications Technology. Information about the North America market comes from goldsteinresearch.com and Statista. 21 The AI market size in China has grown from RMB 15.4 billion (USD 2.42 billion) in 2016 to RMB 59.3 billion (USD 9.3 billion) in 2018 according to AskCI Consulting and Eastmoney.com. 22 In North America, the AI market was valued at USD 6.8 billion in 2017 and increased to USD 8.6 billion in 2018 according to Fortune Business Insights and Statista estimates. 23 www.scmp.com/tech/policy/article/3064956/ai-fastest-expanding-discipline-chinasuniversities-180-more-approved (accessed April 3 2022)

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

Market Failures and China’s Chance to Lead

53

identifies it as one of China’s “key development zones,” second only to the Shenzhen Special Economic Zone and the Shanghai Pudong New Area.24 To fulfill the State Council’s 2030 ambition for the development of the AI industry, Chinese policymakers have been willing to support private entrepreneurial firms as national champions. In August 2017, the central government established a “national team” in AI that initially included Alibaba (for smart city), Tencent (medical imaging), Baidu (autonomous driving), and iFlyTek (natural language processing); SenseTime (smart vision) was added a year later. The list was expanded to fifteen companies in 2019; Ping An (inclusive finance) is the only state-owned firm on the list.25 Each firm is charged with the development of a technology platform which grants it official state support and endorsement of its expertise while also requiring it to maintain and share, to some extent, the technology with others, including potential competitors. Partnering with the government may seem to be at odds with the independent, entrepreneurial culture of a start-up, but there is a very real pragmatism driving these relationships. In exchange for assisting the government in achieving its broader policy objectives, firms are granted quasi-monopoly status (or at least preferential treatment) in their specific technical domain. This approach builds on the model developed between the government and private firms governing Chinese cyberspace which is almost completely separated from the global, US-dominated internet. By walling off social media platforms and other popular internet services used around the world, Chinese companies are able to develop their own expertise and technical capabilities that adhere to a different set of Chinese standards. While walling off global AI competitors and other smart city–related technologies cannot be accomplished in the same way as the Great Firewall, promoting the development and enforcement of Chinese technical standards offers an effective alternative. Indeed, the combination of the CCP’s control of information, industrial-policy protectionism, and dynamic and competitive enterprises fosters a mutual interest by the government and private firms in developing and promoting the adoption of Chinese technical standards relevant to smart cities. This is manifested most clearly by Alibaba, which is the star player on the AI national team. Like Tencent and Baidu, it started without 24

Central Committee of the Chinese Communist Party and State Council of the People’s Republic of China, “[Guiding Opinion Regarding Support of Deepening Reforms and Expanding the Hebei Xiong’an New Area](中共 中央 国务院关于支持河北雄安新区全 面深化改革和扩大开放的指导意),”24 January 2019, www.gov.cn/zhengce/2019-01/24/ content_5360927.htm (accessed April 3 2022). 25 The additional firms include: YiTu (Vision Computing), Huawei (Software/Hardware), Hikvision (Video Perception), JD.com (Smart Supply Chain), Megvii (Image Perception), 360 (Cybersecurity), TAL (Smart Education), Xiaomi (Smart Home), and Mininglamp (Smart Marketing).

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

54

China’s Chance to Lead

Chinese government financing in 1999. State ownership was introduced in 2012 when a consortium led by the China Investment Corporation bought 20 percent of Alibaba that was owned by Yahoo. The government has been pulling Alibaba closer ever since, as dramatically played out with the revocation of Ant Group’s IPO (a subsidiary of Alibaba) in December 2020 and the concurrent falling out between Jack Ma and the CCP. As Xu Shijun, head of the government affairs department at Alibaba Cloud’s Smart Digital Division, remarked, “Alibaba has proposed a deep partnership [with the government], not only in business but also in funding. We’re advancing from the past model which was about constructing a single project to today’s model of multidimensional participation in planning, design, construction, service, and operation.”26 The relationship between Alibaba and the government offers a notable illustration of public-private collaboration yielding the successful implementation of a smart city project. Hangzhou’s “City Brain” platform was developed by Alibaba in 2016. It is a cloud-based system that collects data from sensors, videos, social media, and traffic throughout the city which then gets analyzed and deployed via adaptive, responsive technologies. For example, the city was ranked as the fifth most congested city in China but fell to 57th following the launch of the platform.27 It has helped to reduce emergency response times by up to 49 percent.28 Other cities have pursued similar initiatives, with China now hosting numerous cities regarded as among the most ‘intelligent’ in the world, such as Beijing, Shanghai, and Shenzhen. Chinese firms can now confidently point to Chinese smart cities as glowing examples of what they can offer to cities outside China. 2.7

Why Is the BRI the Best Way to Promote Chinese Standards?

An overarching aim of the BRI is to lift China’s TFP through the upgrading of its manufacturing capabilities and by helping China set international standards. The BRI helps with these aims in several ways. First, the BRI facilitates the shifting of lower-end manufacturing to foreign markets through the expansion of a China-centered global value chain. The development 26

Chen Jing, “Zhihui chengshi zhu baixing guoshang ‘shuzi shenghu’ ” [Smart Cities Help the People to Live “Digital Life”], Jingji Ribao, August 30, 2019. www.gov.cn/ xinwen/2019-08/30/content_5425766.htm (accessed April 5 2022). 27 Michelle Toh and Leonie Erasmus, “Alibaba’s ‘City Brain’ Is Slashing Congestion in Its Hometown,” CNN Business, modified 15 January 2019, www.cnn.com/2019/01/15/ tech/alibaba-city-brainhangzhou/index.html (accessed April 5 2022) 28 Jenny W. Hsu, “Alibaba Cloud Launched ‘ET City Brain 2.0’ in Hangzhou,” Alizila.com, 20 September 2018, www.alizila.com/alibaba-cloud-launched-city-brain-2-0-hangzhou/ (accessed April 5 2022).

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

Market Failures and China’s Chance to Lead

55

of industrial parks in foreign markets facilitates the ease of shifting these facilities overseas with infrastructure, regulatory conditions, and services that are attractive to Chinese companies. By enhancing the attractiveness of moving lower-end manufacturing offshore, Chinese companies are encouraged and expected to engage in higher-end manufacturing. Second, the BRI pushes Chinese firms to focus on controlling the patents, brands, and technical standards that are the most valuable parts of GVCs and enable firms to decide the structure of global production networks. In other words, the BRI not only incentivizes Chinese firms to focus on higher quality manufacturing, but to also develop and disseminate the standards for the manufacture of those products. Given the early stage of development of the technologies related to Industry 4.0, Smart infrastructure, and smart cities, there is a wide spectrum of technical standards up for grabs. By incorporating Chinese technical standards into BRI projects and manufacturing facilities, Chinese firms can claim a strong, first-mover advantage in the adoption of their standards to the exclusion of their competitors, thereby positioning Chinese standards as the de facto, if not the formal, standards. A strategically important method for the de facto adoption of Chinese standards is the provision of integrated technology package deals that combine multiple technologies such as smart cities, smart ports, and Port-Park-City projects that create an ecosystem of technologies. Through these kinds of packages, many different technologies may be adopted such as energy smart grids, cloud computing networks and data centers, AI-driven surveillance and facial recognition technology, municipal services such as traffic control and emergency call centers, and other integrated platforms and features. These kinds of package deals offer an important vehicle for Chinese companies to export and lock in their technologies, making it difficult and expensive to switch to an alternative standard, even if it is superior. Third, China can point to the extensive adoption of its technologies as indicators for their high quality, bolstering their case in international standards deliberations and the global marketability of their products (Chen et al. 2018). In 2015, the State Council published its “Plan to Deepen Reform of Standardization Work” which explicitly directs Chinese companies to integrate Chinese standards into foreign projects as well as equipment and infrastructure exports as part of the BRI.29 The plan also 29

国务院办公厅 [Office of the State Council], “国务院办公厅关于印发 贯彻实施《深化标 准化工作改革方案》行动计划(2015–2016 年)的通知” [Notice of the General Office of the State Council on issuing and implementing the “Deepening Standardization Work Reform Plan” Action Plan (2015–2016)], 30 August 2015, www.gov.cn/zhengce/ content/2015-09/10/content_10154.htm (accessed April 6 2022).

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

56

China’s Chance to Lead

commissioned an analysis of standards across BRI countries, called for 500 priority Chinese standards to be translated into foreign languages as part of an official “Standards Foreign Language Action Plan,” and advocated for standardization centers to be established in BRI host countries. The widespread adoption of Chinese standards strengthens the argument that standards development organizations should formally adopt their standards, ceteris paribus. Fourth, the deployment of 5G networks in the context of smart infrastructure projects and interlinked smart manufacturing facilities can generate a trove of valuable data that are sent back to China that can be mined by Chinese firms and policymakers. By establishing control over the flow of critical information and data as well as goods and services, Chinese firms and policymakers can develop dynamic strategies tailored to specific markets to develop and sustain their competitive advantage. 2.8 Conclusions Market failures have impeded the ability of developing countries to achieve their economic growth objectives. China’s Belt and Road Initiative helps to address two of the main types of market failures. The first type is the inability to attract an adequate supply of infrastructure financing. The second type is transaction costs that impede trade, investment, and information flows among countries. China is well positioned to address these needs via the BRI due to its own motivation and capacity. Its proximate motivation stems from a desire to reduce excess capacity coupled with the longer-term aim of sustaining adequate growth to match the development track records achieved by other East Asian Tigers. To achieve this growth requires lifting its total factor productivity by promoting the adoption of its technical standards and shifting low-end manufacturing offshore. China possesses a unique capacity to fulfill these ambitions thanks to its surplus foreign exchange USD reserves and its commitment to developing new emerging technologies involved in the development of Industry 4.0, smart infrastructure and smart cities. In summary, the large and growing needs of developing countries combined with China’s motivation and capacity create a singular opportunity for China to establish a new international leadership role. But developing countries may vary in their embrace of China’s BRI. Explaining this varying participation is the topic of Chapters 3 and 4.

https://doi.org/10.1017/9781009385879.002 Published online by Cambridge University Press

3

Measuring Infrastructure Needs and Foreign Infrastructure Investment

The previous chapter discussed the motivation for developing countries to seek infrastructure spending from China. It also discussed China’s motivation to address countries’ infrastructure needs, providing an opportunity for China to develop a new leadership role in the global economy. This chapter seeks to first empirically establish the relationship between countries’ market failures and their level of infrastructure spending. It then seeks to test the most likely explanations for Chinese foreign spending based on explanations relating to FDI from Western MNCs. I first discuss the measurement of countries’ infrastructure financing needs. This is a challenging measure to establish for both conceptual and empirical reasons. Conceptually, it is problematic to accurately claim a country has a specific infrastructure spending target. Empirically, reliable data for country-level infrastructure spending gaps are only available for fifty countries, including only thirteen low- and lower-middle-income countries. An alternate approach is to use actual infrastructure spending as a percentage of GDP. This mitigates the conceptual problems to some extent by directly measuring countries’ infrastructure spending priorities; it also addresses the empirical problem because data are available for 120 countries. I then assess the relationship between infrastructure spending and indicators for a variety of market failures relating to countries’ capacity to finance infrastructure projects. The expectation is that infrastructure spending will be lower as market failures rise; put differently, infrastructure spending as a percentage of GDP will be higher with better financial development. The results for the banking industry are consistent with this expectation, but the most noteworthy result is the substitution effect observed for foreign investment and lending with respect to infrastructure spending. As foreign investment and bank lending rise, infrastructure spending declines. This finding further motivates the need to examine the relationship between Chinese foreign spending and host country infrastructure spending. 57

https://doi.org/10.1017/9781009385879.003 Published online by Cambridge University Press

58

China’s Chance to Lead

Based on the established literature regarding FDI inflows primarily involving Western MNCs, I test whether the most likely explanations can account for Chinese foreign spending in the context of the BRI. First, I distinguish between Chinese foreign construction and investment inflows. The FDI literature typically restricts its focus to the latter, though infrastructure spending is commonly examined with a very similar, if not the same, set of variables (e.g., Henisz 2002; Jiang et al. 2019). Second, I perform tests to assess whether there are significant relationships between these explanatory variables and Chinese foreign construction and investment spending. Notably, the results do not yield convincing results that these variables can account for Chinese foreign spending patterns. This finding supports the need for a novel approach to Chinese foreign spending in the context of the BRI, which is the topic of Chapter 4. 3.1

Measuring Countries’ Infrastructure Financing Needs

How much infrastructure a country needs depends on the objective pursued. The objective pursued in turn depends on the political context of the country, its economic growth aspirations, and the attention devoted to social and environmental issues. A country’s political context can yield vastly different investment priorities. For example, the political contexts of North and South Korea following World War II when the Korean peninsula was divided into two independent states contributed to vastly different views of their infrastructure investment needs despite similar levels of development. Likewise, a country’s economic growth aspirations are likely to vary depending on its natural resource endowments. For example, Singapore and Brunei are both small city-states that neighbor Malaysia, both were formerly British colonies prior to independence, and both are autocracies. However, Brunei is endowed with an abundance of hydrocarbon reserves while Singapore is not. Consequently, Brunei’s economy and growth aspirations have been heavily influenced by its dependence on hydrocarbon revenue which has impeded the development of a diversified economy and accompanying infrastructure investments. Singapore, by contrast, was forced to make capital and infrastructure investments that would attract foreign companies and which have contributed to the modernization and diversification of its economy. A country’s attention to social and environmental issues can further influence its infrastructure needs. Consider Maldives in comparison to Bhutan. Both have small populations of less than a million. Maldives is a middle-income country, whereas Bhutan is lower-middle income, which

https://doi.org/10.1017/9781009385879.003 Published online by Cambridge University Press

Measuring Infrastructure Needs

59

suggests the infrastructure gap is lower for the former. Maldives is also regarded as an autocracy that holds semi-competitive elections, whereas Bhutan is regarded as having more democratic political institutions which typically correspond to a higher valuing of social and environmental priorities. Together, these characteristics suggest Bhutan should have larger infrastructure financing needs, ceteris paribus. Although Bhutan is well known for its dedication to environmental preservation in the context of a land-locked Himalayan environment, rising sea levels threaten to make 80 percent of the Maldives uninhabitable by 2050, contributing in turn to a far larger share of investments into infrastructure to mitigate this threat. The main point is that it is difficult to draw generalizations about countries’ infrastructure investment needs. It is often heavily dependent on the specific circumstances of individual countries. Moreover, there is a complex relationship among various factors that drives the utility of infrastructure investment. For example, most infrastructure is in the form of networks, which creates threshold effects and returns that vary with the stage of completion of the network and the number of users (e.g., a subway system). Additionally, the benefits of infrastructure investments are often dependent on complementary inputs. For example, investments in transportation and electricity services depend not only on roads and power plants, but also on the proximity of the users to markets. Further, infrastructure may be built in pursuit of goals other than growth, such as promoting social equity, environmental preservation, public health, political goals, or even personal enrichment. For this reason, researchers such as Rozenberg and Fay (2019, p. 3) at the World Bank caution against using precise estimates for countries’ infrastructure spending needs: “This complex relationship implies that it is not possible to determine an optimal level of infrastructure.” Finally, to the extent estimates of infrastructure gaps are calculated for individual countries, the most recent and comprehensive database that is publicly available (the G20’s Global Infrastructure Outlook database; Heathcote and Mulheirn 2017) only includes fifteen low- and lowermiddle-income countries.1 Thus, there are both conceptual and empirical issues with existing measures of infrastructure needs available for a large sample of developing countries. However, comprehensive and accurate data have recently been compiled for what countries actually spend on infrastructure which helps address the empirical problem. It can also mitigate, to some extent, 1

The most recent updates extend the coverage from fifty countries in the 2017 report to fifty-six countries total.

https://doi.org/10.1017/9781009385879.003 Published online by Cambridge University Press

60

China’s Chance to Lead

the conceptual issue regarding the optimal level of spending insofar as actual spending is likely to reflect the specific priorities of that country even if it does not capture the gap. Additionally, infrastructure gap measures may suggest a country with a very high level of infrastructure spending, such as Laos (at 11.86 percent of GDP in 2011) or the Republic of Congo (at 6.49 percent of GDP in 2011), does not have a gap. But this assessment would fail to identify the extent to which a country exceeds the gap. And in the view of a country’s political leaders, the country may still have a gap that justifies the desire for infrastructure spending from foreign sources such as China. The measurement of their actual infrastructure spending allows us to account for such cases and to provide a quantitative measurement of the conceptual issues (i.e., countries’ individual spending priorities) that underlie these high spending levels. For these reasons, I use data for country-level infrastructure spending provided by Fay et al. (2019). Based on various data sources, they construct high, low, and preferred estimates for 120 countries for infrastructure spending in 2011. The year for which these spending amounts are measured is appropriate for gauging the demand for Chinese infrastructure spending following the introduction of the BRI in 2013. The high estimate relies on measures for gross fixed capital formation (GFCF) of the general government combined with private participation in infrastructure (from the World Bank’s PPI database) to capture private investment. The low estimate is estimated from data measuring gross fixed capital formation on civil engineering. The preferred estimate, which is the measure I use, is based on national treasury systems using the World Bank’s BOOST database, which collects the full fiscal accounts of 55 countries and painstakingly analyzes the data. This is combined with the PPI dataset to account for private expenditures. The authors then propose several refinements which extend the coverage to 120 countries based on information available from the GFCF data. I adopt the most conservative of these refinements in the data used below for countries’ spending levels. 3.2

Infrastructure Spending and Market Failures

This section examines the relationship between infrastructure spending and indicators for the kinds of market failures that would keep this spending lower than would occur without such market failures. For example, a country with an underdeveloped equity market where private firms could raise funds for large infrastructure projects may display a lower level of infrastructure spending than a country with a well-developed equity

https://doi.org/10.1017/9781009385879.003 Published online by Cambridge University Press

Measuring Infrastructure Needs

61

market. Thus, the general expectation is that countries with larger market failures will have lower levels of infrastructure spending (if the infrastructure gap were analyzed, then this correlation would be expected to be positive). Indicators for market failures regarding the capacity for infrastructure financing come from the World Bank’s Global Financial Development Database. Indicators are placed into one of five categories, including Access, Depth, Efficiency, Stability, and Other. The Other category includes variables for loans from international banks as well as whether the country experienced a banking crisis. I supplement these with indicators that do not come from the Global Financial Development Database (GFDD), including sovereign risk rating, tax revenue as a percentage of GDP, and three political risk indicators including a country’s ICRG investment profile, and two measures for countries’ political constraints to assess political risk stemming from policy uncertainty (polconiii and polconv). Polconv uses the same methodology as polconiii but also takes two additional potential veto players into account: sub-federal units and the judiciary. Given the two potential additional veto players, the values of the polconv are generally higher than those for polconiii.2 Finally, I include interaction terms for tax revenue as a percentage of GDP with each of the three political risk variables to assess whether reduced levels of infrastructure spending due to political risk are moderated by a country’s taxation base; that is, political risk that deters foreign investors may be amplified by evidence of low tax revenue. Table 3.2 reports the results of OLS regressions for each explanatory variable with respect to infrastructure spending as a percentage of GDP. All of the variables in the GFDD were tested; many variables had too few observations or did not return significant results. I restrict the reported findings to those variables which offer helpful insights into the relationship between market failures and infrastructure spending. I focus on domestic indicators first. The results indicate that more developed deposit-banking systems (as measured by private credit extended 2

The Political Constraints III and V (polconiii and polconv) indices are developed by Henisz (2002). In essence, these measures of political constraints estimate the feasibility of policy change by using data on the number of independent branches of government (executive, lower and upper legislative chambers) with veto power over policy in addition to the distribution of preferences of the members of the different branches of government. The distribution of preferences is measured by the fractionalization of the different branches of government. If there are no veto players and preferences are aligned, the executive can change policy easily and the indices take a value of zero. In contrast, the more veto players exist and the more fractionalized the branches are, the higher the values of the political constraints indices.

https://doi.org/10.1017/9781009385879.003 Published online by Cambridge University Press

62

China’s Chance to Lead

by them and by their total assets as a percentage of GDP) correlate positively with a higher level of infrastructure spending; private credit extended as a percentage of GDP yields the highest adjusted R-squared value of any variable (0.103) with total assets as a percentage of GDP not far behind (0.081). The results also indicate that greater stock market volatility correlates negatively with infrastructure spending, suggesting wariness of the part of investors who may perceive the country as risky (as signaled by market volatility); the adjusted R-squared value is also relatively high (0.097). This interpretation is consistent with the results for the banking crisis dummy variable, which also indicates less infrastructure spending when there is a banking crisis although the adjusted R-squared value is small (−0.007). In addition to these domestic indicators with significant results are the variables that do not display significant results. Interestingly, several indicators for the stock market do not produce significant results, including value traded on the stock market excluding the top 10 traded companies to total value traded (%), market capitalization excluding top 10 companies to total market capitalization (%), and stock market turnover ratio (%). Also interesting to note is the lack of significant results for sovereign risk rating, which is consistent with the results reported by the OECD (2018), and for tax revenue as a percentage of GDP. But the most striking results concern the indicators relating to international lenders and investors. The results for the three different indicators of international debt securities (outstanding international private debt securities to GDP (%), outstanding international public debt securities to GDP (%), and outstanding total international debt securities to GDP (%)) also produce significant negative coefficients with respect to infrastructure spending. The adjusted R-squared values range from 0.046 to 0.065. They suggest a substitution effect; that is, domestic spending declines as financing from international sources increase. We can observe the same relationship for both indicators for loans from nonresident banks (loans from nonresident banks (net) to GDP (%), and loans from nonresident banks (amounts outstanding) to GDP (%)). With respect to the three political risk variables, only polconv displays significant results, indicating a decline in infrastructure spending as political risk rises according to this measure. Finally, I test several interaction effects to assess whether an indicator’s effect is conditional on another variable. I specifically examine whether the impact of tax revenue as a percentage of GDP on infrastructure spending is conditional on a country’s perceived level of political risk, using each of the three political risk indicators. This is to assess whether a substitution effect may occur, as discussed above with regard

https://doi.org/10.1017/9781009385879.003 Published online by Cambridge University Press

Measuring Infrastructure Needs

63

to the international debt securities and loans from nonresident banks. The results indicate there is no significant relationship. An important takeaway from these results concerns the significant negative relationships observed between the indicators for international lenders and investors and infrastructure spending (Table 3.1). These findings suggest a substitution effect whereby domestic governments will reduce their own infrastructure spending as foreign investors and lenders increase their commitments, enabling host country governments to use those funds for other priorities or to reduce overall spending. In either case, it motivates the need to investigate further the relationship between foreign spending and recipient governments. 3.3

Chinese Foreign Spending

3.3.1

Measuring Chinese Construction versus Investment Spending

To investigate Chinese foreign investment and construction spending, I use the China Global Investment Tracker (CGIT) dataset compiled and managed by Derek Scissors. To my knowledge, it offers the most up-to-date and comprehensive coverage regarding Chinese firms’ foreign spending. The CGIT dataset includes all verifiable investment and construction transactions worth $100 million or more since 2005.3 Data come from primary sources provided by MOFCOM, Chinese corporate sources and partners, and independently derives numbers for Chinese foreign spending. The dataset is revised biannually to account for revisions by these sources. As mentioned, the data distinguish between construction and investment activities. Construction activity is valuable but does not involve ownership. It is typically performed by massive state-owned enterprises, such as Sinomach, China Communications Construction, State Construction Engineering, and others. Investment includes transactions that involve the purchase of a foreign asset, such as a corporate acquisition or a greenfield investment. The dataset used for this study is from spring 2020 and includes 1706 investments worth more than $1.2 trillion. It also includes 1749 construction projects worth over $830 billion from 2005 up to the end of 2019, which marks a natural end point due to the onset of the COVID-19 pandemic in January 2020. This dataset is used in other studies of the BRI, such as OECD (2018) and Chen and Lin (2018). 3 The CGIT dataset does not include trade, bonds, loans, aid, bank deposits, and other forms of storing foreign exchange overseas that do not involve ownership or services in the host country.

https://doi.org/10.1017/9781009385879.003 Published online by Cambridge University Press

https://doi.org/10.1017/9781009385879.003 Published online by Cambridge University Press

Access Bank accounts per −0.000* 1,000 adults (0.000) Value traded exclud−0.002 ing top 10 traded (0.010) companies to total value traded (%) Market capitalization excluding top 10 companies to total market capitalization (%) Depth Private credit by deposit money banks to GDP (%) Deposit money banks’ assets to GDP (%) Outstanding international private debt securities to GDP (%) Outstanding international public debt securities to GDP (%)

Infrastructure Spending (% of GDP)

−0.004 (0.018)

0.007*** (0.002)

Table 3.1  Market failure indicators and infrastructure spending

0.006** (0.002) −0.028** (0.013)

−0.021** (0.009)

https://doi.org/10.1017/9781009385879.003 Published online by Cambridge University Press

Other Loans from nonresident banks (net) to GDP (%) Loans from nonresident banks (amounts outstanding) to GDP (%) Constant 2.206*** (0.211) N 67 Adjust R2 0.011

Outstanding total international debt securities / GDP (%) Efficiency Stock market turnover ratio (%) Stability Stock price volatility

1.845*** (0.391) 19 0

1.926** (0.783) 19 0

1.942*** (0.147) 115 0.103

1.938*** (0.168) 115 0.081

1.747*** (0.218) 28 0.002

1.814*** (0.139) 45 0.046

1.850*** (0.145) 45 0.061

−0.018*** (0.007)

1.639*** (0.145) 41 0

−0.000 (0.000)

2.484*** (0.304) 37 0.097

−0.043*** (0.015)

1.796*** (0.218) 28 0.054

−0.194** (0.076)

1.894*** (0.133) 56 0.065

−0.020*** (0.007)

https://doi.org/10.1017/9781009385879.003 Published online by Cambridge University Press

2.292*** (0.160) 120 −0.007

−0.622*** (0.160)

2.134*** (0.741) 34 0

−0.034 (0.067)

2.138*** (0.519) 78 0

−0.009 (0.029)

Note: *, **, and *** indicate significance at the 0.10, 0.05, and 0.01 level.

N Adjust R2

Constant

polconv*Tax revenue as% of GDP

ICRG investment profile*Tax revenue as% of GDP polconiii*Tax revenue as% of GDP

polconv

polconiii

ICRG Investment Profile

Tax revenue as% of GDP

Non-GFD Database Indicators Sovereign risk rating

Banking crisis dummy (1=banking crisis, 0=none)

Infrastructure Spending (% of GDP)

Table 3.1  (cont.)

2.129*** (0.502) 82 0

−0.010 (0.063)

−1.523** (0.747)

2.653*** 2.810*** (0.402) (0.391) 117 107 0.013 0.045

−1.297 (0.994)

0.171 −0.073 −0.037 (0.220) (0.074) (0.068) 0.469 (0.480) −4.174 (3.473) −2.591 (2.678) −0.026 (0.028) 0.220 (0.203) 0.110 (0.154) −1.235 3.351** 2.935** (3.745) (1.298) (1.180) 57 77 70 0 0.005 0

Measuring Infrastructure Needs

3.3.2

67

Testing Likely Explanations for Chinese Foreign Spending

In this section I test whether country-level infrastructure spending correlates with Chinese foreign construction and/or investment spending. I also consider two of the most common explanations for foreign investment in the context of Western MNCs, namely, the host country’s income level and geographic distance. There are several reasons why a country’s level of development, as represented by GDP per capita, is regarded as a likely determinant of inward FDI by Western MNCs (Bénassy‐Quéré et al. 2007). First, countries at a low level of development likely possess different factor endowments to the countries from which FDI originates, thus justifying foreign investment. Second, foreign firms are attracted by citizens’ purchasing power in a host economy, which raises the potential for product differentiation and higher profits. Third, GDP per capita can proxy for productivity and real wages. Foreign MNCs with higher productivity than local firms may be able to capture a larger difference between labor costs and prices charged for products produced in low-income economies. Additionally, GDP per capita can be a proxy for the quality of institutions. Daude and Stein (2001) show inward FDI to be significantly influenced the quality of institutions, with five out of six governance indicators provided by Mundial et al. (2019) displaying significant relationships, including political instability and violence, government effectiveness, regulatory burden, rule of law and graft. Only the voice and accountability indicator yields a non-significant relationship to FDI. The quality of a country’s institutions can affect FDI for several reasons. First, good institutions raise a country’s productivity potential which may attract foreign investors (Acemoglu et al. 2001, 2002). Second, good institutions can reduce the costs to FDI, such as corruption (Wei 2000). Third, the high sunk costs involved with FDI makes it especially vulnerable to many types of uncertainty such as weak enforcement of property rights, policy changes, or a weak legal system in general. What is notable about these justifications for GDP per capita is that many of them are of particular salience for foreign investors originating from high-income economies. While the impact of institutional quality may be more relevant for Chinese investors, its salience again depends on whether we are interested in explaining foreign investment by private firms or construction spending which is typically initiated by SOEs. A second common explanatory variable for FDI is geographic distance. This has long been regarded as negatively impacting economic

https://doi.org/10.1017/9781009385879.003 Published online by Cambridge University Press

68

China’s Chance to Lead

exchange between countries by elevating transaction costs, particularly relating to transportation and communication (Tesar and Werner 1995). Studies on trade have confirmed this relationship (e.g., Rose 2000; Frankel and Rose 2002), but scholars of FDI argue that setting up a foreign subsidiary could circumvent these problems. Others argue that even if this is the case and economic transactions are mainly intrafirm, then distance will still negatively impact FDI (Helpman 1984). Although costs associated with transportation and communication have declined over time, the effect of distance has not fallen and may have actually increased (Ghemawat 2001; Bénassy‐Quéré et al. 2007). This has led to new studies that point to the role of information asymmetry. Geographical distance raises this asymmetry due to increases in the costs of information gathering, search costs, and information exchange between investors (Coval and Moskowitz 1999). Recent work has confirmed that information flow is a significant factor in attracting FDI and that it reduces the distance between countries, regardless of income level (Ly et al. 2018). A potential problem for the salience of distance and the role of information is the kind of information that private investors find useful in relation to SOEs. Private investors looking for greenfield investments may place greater value on information availability with regard to technically complex manufacturing facilities and due to their relatively weak position in relation to the host government. For Chinese SOEs, however, this information deficit may be compensated for by governmentto-government negotiations and by the implicit guarantee the Chinese government provides to its SOEs which reduces the risk they face when investing in locations with an informational deficit. The Chinese foreign investment and construction data (CGIT dataset) includes continuous variables with right-skewed distributions. The accepted approach to estimating effects for this type of data is to use the fixed effects Poisson estimator (Wooldridge 1999, 2010).4 This is the same econometric model used by the World Bank (2018). Two dependent variables are analyzed, including construction (contract) and investment (FDI) measures that come from the CGIT dataset. To conduct robust analyses of potential relationships, I examine two alternative measures of the dependent variables. In Panel A of Table 3.2, I use the sum of foreign spending for each country across the 2014 to 2019 period as the dependent variable. Panel B uses a panel specification where country-year data are as the dependent variables. 4

Specifically, the fixed effects Poisson estimator is fully robust for estimating the conditional mean parameters. See Wooldridge (1999, 2010).

https://doi.org/10.1017/9781009385879.003 Published online by Cambridge University Press

https://doi.org/10.1017/9781009385879.003 Published online by Cambridge University Press

N Pseudo R2

Infrastructure spending (% of GDP)*Ln(GDP per capita)*Distance to China Constant

Infrastructure spending (% of GDP)*Ln(GDP per capita) Infrastructure spending (% of GDP)*Distance to China Ln(GDP per capita)*Distance to China

Distance to China

14.554*** (2.258) 87 0.120

−0.009 (0.010) −0.032 (0.158) −0.652*** (0.231)

Infrastructure spending (% of GDP)

Ln(GDP per capita)

(2)

(1)

Panel A

−13.667 (21.380) 87 0.143

−0.144 (0.151) 4.254 (3.109) 2.411 (2.340) −0.005 (0.005) 0.021 (0.017) −0.470 (0.346)

Contract

−1.021 (0.722) 1.106 (4.572) −0.551 (3.867) 0.089 (0.078) 0.120 (0.079) −0.124 (0.502) −0.011 (0.009) 13.162 (35.471) 87 0.160

(3)

Contract

11.679*** (2.399) 87 0.162

−0.024* (0.014) 0.619* (0.324) −0.941*** (0.244)

(4)

FDI

−20.549 (33.807) 87 0.178

−0.137 (0.213) 5.143 (4.910) 2.562 (3.810) −0.005 (0.005) 0.018 (0.022) −0.496 (0.555)

(5)

FDI

Total Sum of Contract or FDI from 2014 to 2019 by Country Contract

Table 3.2  Potential explanatory variables for Chinese foreign spending

−1.921 (1.321) −0.568 (5.512) −3.233 (4.603) 0.179 (0.129) 0.219 (0.146) 0.141 (0.622) −0.021 (0.014) 31.309 (40.870) 87 0.211

(6)

FDI

https://doi.org/10.1017/9781009385879.003 Published online by Cambridge University Press

19.650 (12.062) N N 225 0.038

Note: *, **, and *** indicate significance at the 0.10, 0.05, and 0.01 level.

Country Fixed Effects Year Fixed Effects N Pseudo R2

Infrastructure spending (% of GDP)*Ln(GDP per capita)*Distance to China Constant

Infrastructure spending (% of GDP)*Ln(GDP per capita) Infrastructure spending (% of GDP)*Distance to China Ln(GDP per capita)*Distance to China

Distance to China

10.228*** (1.457) N N 225 0.031

−0.006 (0.042) −0.066 (0.080) −0.287** (0.119)

Infrastructure spending (% of GDP)

Ln(GDP per capita)

(6)

(5)

Panel B 0.025 (0.716) −1.364 (1.486) −1.284 (1.334) 0.034 (0.048) −0.035 (0.077) 0.139 (0.164)

Contract

Contract

Table 3.2  (cont.)

12.539* (7.620) 1.927 (2.242) 1.534 (1.962) −1.637 (1.014) −1.452* (0.854) −0.238 (0.253) 0.189* (0.114) −4.946 (17.428) N N 225 0.046

(7)

Contract

5.411** (2.172) N N 122 0.029

0.033 (0.045) 0.187 (0.117) 0.005 (0.214)

(8)

FDI

Panel Data, 2014–19

8.539 (18.075) N N 122 0.093

−1.116 (0.869) −0.471 (2.168) 0.108 (2.115) 0.253** (0.106) −0.092 (0.090) 0.017 (0.252)

(9)

FDI

0.704 (13.784) −0.022 (3.069) 0.501 (2.769) 0.013 (1.852) −0.299 (1.542) −0.035 (0.349) 0.027 (0.207) 5.124 (24.191) N N 122 0.094

(10)

FDI

Measuring Infrastructure Needs

71

The explanatory variables include the individual variables mentioned above, including infrastructure spending (% of GDP) in 2011, Ln(GDP per capita) in 2013, and geographic distance to China. The table also displays results for interactions between each of these variables. When considering the potential relation between infrastructure spending and Chinese foreign spending, it is helpful to control for GDP per capita in order to offer a comparison for countries that controls for their level of development. As displayed in Table 3.2, the only significant results at a 5 percent level or lower are for the Distance to China variable for the contract and FDI variables (models 1 and 4) in Panel A. For the country-year specification in Panel B, model (5) indicates a significant negative relationship for the Distance to China variable but no other models display a significant relationship. These results are suggestive of a relationship, but they are sensitive to the specification used since the results become non-significant when interactions are included. Overall, the more noteworthy finding is the absence of significant relationships for most of the variables. These surprising findings force us to consider alternative explanations for why countries host varying levels of Chinese foreign spending. 3.4 Conclusions The aim of this chapter has been to empirically measure countries’ infrastructure spending needs, to identify the types of market failures that display significant relationships to these spending needs, and to then assess whether common explanatory factors used in the context of Western foreign investment can account for Chinese foreign spending in the context of the BRI. To this end, the chapter first discussed the challenges associated with identifying a valid measure of countries’ infrastructure financing needs for both conceptual and empirical reasons. The best solution is to use countries’ actual infrastructure spending as a percentage of GDP. I then proceeded to test whether countries’ infrastructure spending correlates with a wide variety of indicators for the prevalence of market failures in a country. One of the most striking findings of this analysis is the strong relationship between foreign lending and debt securities purchases and infrastructure spending, suggesting a substitution effect that enables host country governments to use domestic sources for other priorities or to pay down government debt (or prevent its increase). The important takeaway point is the significance of foreign lending and investment to recipient countries’ infrastructure spending decisions.

https://doi.org/10.1017/9781009385879.003 Published online by Cambridge University Press

72

China’s Chance to Lead

Finally, I tested whether common explanatory factors for foreign investment by Western investors could account for Chinese foreign spending patterns, in addition to countries’ infrastructure spending levels. The findings suggest there may be some explanatory power for the Distance to China variable, but this effect is not particularly robust. The lack of significance for these variables calls for an alternative approach to explaining the pattern of Chinese foreign spending across countries in the context of the BRI.

https://doi.org/10.1017/9781009385879.003 Published online by Cambridge University Press

4 Theory

Why Countries Vary in Their Participation in the Belt and Road Initiative

Although developing countries share many of the same types of market failures, they display varying levels of participation in the BRI. Widely used explanatory factors for foreign investment by Western MNCs fail to account for these Chinese foreign spending patterns, as shown in the prior chapter. This calls for an alternative approach. To explain this variation, I begin with a discussion of recipient country characteristics that impact the demand for Chinese spending, including the political regime, clientelism, and the public-private orientation of the corporate sector. I then discuss the supply-side factors that influence Chinese foreign spending, including the Chinese Communist Party (CCP), state-owned entities (e.g., SOEs), and private firms. In the third section, I consider the compatibility of these demand and supply characteristics. The key prediction is that electoral autocracies will display the strongest compatibility with Chinese foreign construction spending. This is amplified when the leaders of these regimes have a weak or insecure hold on power. The fourth and final section turns to the implications for the adoption of Chinese standards, which also predicts electoral autocracies will most avidly adopt Chinese standards stemming from their eagerness for Chinese infrastructure spending. I now turn to a more detailed discussion regarding demand and supply factors, their compatibility in the context of different political regimes, and the implications for the adoption of Chinese standards in turn. 4.1

The Demand Side: Recipient Country Characteristics

Solutions to market failures are typically framed with respect to how to achieve economic development following the model of today’s advanced democracies. But Douglas North (1994, 359) regards the neoclassical approach to markets and their failures as problematic due to its modeling of a frictionless and static world: “When applied to economic history and development it focused on technological development and more recently human-capital investment but ignored the incentive structure embodied 73

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

74

China’s Chance to Lead

in institutions.” Stiglitz (1999) echoes this view: “governments face information and incentive problems no less than does the private market.” To explain whether governments are likely to support economic development, Acemoglu and Robinson (2000) argue we should look at “agents who have political power and fear losing it.” Specifically, they advise, “we should look more to the nature of political institutions and the determinants of the distribution of political power.” While the political economic arrangements of advanced democracies may be viewed as preferable for achieving Pareto optimal solutions, these solutions are often incompatible with the political incentives of incumbent rulers in weakly democratic or nondemocratic states. Moreover, these models neglect alternative solutions that have demonstrated considerable success, such as India and China. While these emerging giants’ development models may encounter problems as they approach the technological frontier, they have displayed considerable success in rapidly moving closer to it. China has clearly shown that an alternative approach aligned with the incentives of nondemocratic rulers can produce beneficial outcomes even if it is Pareto suboptimal in the long run. But research so far does not help us answer the original question: Why do countries with equivalent types and degrees of market failure vary in their receptivity to Chinese spending? I argue it depends on political rulers’ ability to leverage it for their own political gain. Chinese foreign infrastructure spending affects recipient countries in two main ways. First, it promotes economic growth which in turn promotes regime stability (Miller 2012). Second, it increases the supply of resources that political incumbents can redistribute via their clientelist network. In developing countries with abundant market failures, political rulers rely on clientelism to build and maintain their hold on power. This includes the targeted distribution of public resources in exchange for political support. The relative supply of resources to distribute depends on the public versus private control of the corporate sector. The size of the clientelist network and the demand for patronage resources vary in qualitatively distinct ways depending on the structure of the political regime, as displayed in Table 4.1, and together these affect the demand for Chinese infrastructure spending. I make a two-part argument. First, the political regime type and its affiliated clientelist network structure raises the demand for Chinese infrastructure spending on the part of electoral autocracies and electoral democracies in comparison to closed autocracies. In the former regimes, political incumbents face election pressures due to semi-competitive and fully competitive elections and must manage larger clientelist networks than leaders of closed autocracies that hold non-competitive elections

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

Theory

75

Table 4.1  Political regimes, clientelism, and the public-private orientation of the corporate sector Clientelist Network Structure

Public-Private Orientation of Corporate Sector

Closed Autocracy

Narrow

Public dominant

Electoral Autocracy

Dominant party-led

Public-oriented

Electoral Democracy

Competitive

Private-oriented

Liberal Democracy

Weak

Private dominant

Regime Type

Outcome

Example

Narrow stateresourced clientelism Dominant party state-resourced clientelism Competitive, oligarch-led clientelism Programmatic

UAE

Malaysia and Djibouti Indonesia

Greece

if they are even held; China’s abundant infrastructure spending can help address this demand. Second, electoral autocracies rely more on state-owned entities (e.g., SOEs) for the distribution of resources via their clientelist network, which is more compatible with China’s SOEled infrastructure spending arrangements due to their shared interest in opacity and the capacity for government-to-government negotiations. This enables rulers of electoral autocracies to exert stronger controls over larger volumes of Chinese infrastructure spending than leaders of electoral democracies. Consequently, electoral autocracies display the highest demand for Chinese infrastructure spending. I discuss political regime types, clientelist network structures, and the public-private orientation of the corporate sector in turn.1 4.1.1

Political Regimes

Political regimes vary in qualitatively distinct ways with regard to their reliance on clientelism and the structure of the clientelist network. Over the past two decades, scholars in comparative politics have devoted increasing attention to the institutional differences between political regimes (Geddes 2003; Magaloni et al. 2013; Wahman et al. 2013; 1

Regime types set a lower bound on the Nash equilibrium, but the upper bound could be Pareto optimal; for example, from Olson’s argument, leaders of closed autocracies could act for the economy’s aggregate welfare but typically they do not. Electoral autocracies lift the floor on the Nash Equilibrium and so on.

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

76

China’s Chance to Lead

Carney 2018; Miller 2020). They have noted that closed autocracies display important, qualitatively distinct governance characteristics from electoral autocracies, and both categories of autocratic regimes differ in qualitatively distinct ways from democracies. Likewise, electoral democracies differ from liberal democracies in qualitatively distinct ways. Indicators of national political systems typically fall along a democracy-​ nondemocracy continuum (e.g., Acemoglu et al. 2008, 2019; Delis et al. 2020). However, if political regimes differ from one another based on qualitatively distinct and theoretically meaningful characteristics, then a categorical approach may be warranted. To illustrate this point, consider two widely cited measures of national legal systems: quality of law and legal tradition. The former is a continuous variable that measures “perceptions of the extent to which agents have confidence in and abide by the rules of society, and in particular the quality of contract enforcement, property rights, the police, and the courts, as well as the likelihood of crime and violence.”2 Legal tradition, such as common or civil law, refers to “the strategy of social control that seeks to support private market outcomes, whereas civil law seeks to replace such outcomes with state-desired allocations” (La Porta et al. 2008). Due to their qualitatively distinct approaches to handling economic affairs, common law and civil law are classified as categorical variables. Likewise, continuous measures of democracy fail to capture the qualitatively distinct political characteristics of each regime type that contribute to changes in the reliance on clientelism versus programmatic distribution policies. Specifically, the switch from an autocracy that does not hold elections to one with semi-competitive elections generates the need for a broad-based institutionalized ruling party apparatus that can administer the systematic distribution of clientelist goods and services. Likewise, the introduction of genuinely competitive elections in democracies amplifies pressures to hold politicians accountable for how public finances are spent so as to reduce the incidence of corruption and related clientelist practices. The scope and public versus private orientation of clientelist networks varies with respect to political regime. Scholars from the democratization literature have developed measures for a four-regime typology that is anchored in a widely accepted theoretical categorization of political regimes (e.g. Diamond 2002; Rössler and Howard 2009; Schedler 2013). The regimes include closed autocracy, 2

See the definition provided by the World Bank: http://info.worldbank.org/governance/ wgi/pdf/rl.pdf (accessed April 10 2022).

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

Theory

77

electoral autocracy, electoral democracy, and liberal democracy (Lührmann et al. 2018).3 Regimes are identified as either democratic or autocratic depending on whether they hold multiparty elections and whether the elected government is accountable to its citizens.4 Accountability is considered effective if democratic institutions adhere to the six institutional guarantees enumerated by Dahl (1971, 1998),5 including elected officials, free and fair elections, freedom of expression, alternative sources of information, associational autonomy, and inclusive citizenship. Electoral democracies meet these minimum requirements, but liberal democracies are distinguished by additional institutional features that protect individual and minority rights. These include legislative and judicial checks on executive power as well as protections for individual liberties via an effective rule of law that treats all citizens equally. Examples of electoral democracies include Indonesia, the Philippines, India, and Brazil. In autocracies, rulers are not accountable to citizens by Dahl’s standards. The key difference along the authoritarian spectrum is whether multiparty elections are held to decide the office of the chief executive and seats in the national legislature (Schedler 2013). In closed autocracies, there are no competitive elections (Brownlee 2009; Rössler and Howard 2009; Donno 2013). To the extent political competition exists, it occurs only among members of the governing coalition, such as within the ruling party (e.g., the Communist Party of Vietnam) or within the ruling family (e.g., Saudi Arabia). Information and resources are monopolized by incumbent rulers, thus denying opportunities to potential opponents to criticize, spread information about, or mobilize opposition against it (Geddes 1999; Gandhi and Przeworski 2006). For example, the press is normally owned and controlled by the political elite and therefore refrains from independently monitoring and reporting on questionable government activities (Kern and Hainmueller 2009). Examples of closed autocracies include Saudi Arabia, Brunei, and China. 3

In my previous book, Authoritarian Capitalism (2018), I used the following regime categories: narrow authoritarian regime, single party authoritarian regime, dominant party authoritarian regime, and democracy. The narrow and single-party authoritarian regime categories together comprise the closed autocracy category. The dominant party authoritarian regime category is equivalent to the electoral autocracy category. And the democracy category corresponds to the electoral and liberal democracy categories. 4 Accountability is considered effective if democratic institutions adhere to the six institutional guarantees enumerated by Dahl (1971, 1998), including elected officials, free and fair elections, freedom of expression, alternative sources of information, associational autonomy, and inclusive citizenship. 5 Dahl (1971, 1998) provides the most comprehensive and most widely accepted theory of what distinguishes a democracy from an autocracy.

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

78

China’s Chance to Lead

In electoral autocracies, on the other hand, there are at least de jure multiparty elections for the executive and/or legislature. But these institutions are undermined by mechanisms that allow rulers to evade electoral accountability (Diamond 2002; Schedler 2002, 2013; Gandhi and Lust-Okar 2009; Levitsky and Way 2010). Thus, electoral autocracies fail to meet the standards of democracy enumerated by Dahl.6 Despite permitting opposition political parties, the extent of this competition is restricted by the ruling party, including such measures as limits on press freedom, the manipulation of electoral rules, and eligibility requirements to run for office. These restrictions are designed to preserve an entrenched majority of legislative seats for the ruling party, and to ensure it retains control of the government. Although political opposition is permitted, critical information that could be used by political opponents is curtailed (e.g., information about the ethnic distribution of homeowners, as in Singapore), and access to resources is closely monitored and tightly controlled by the ruling party. Examples of electoral autocracies include Malaysia, Singapore, Sri Lanka, and Ethiopia. 4.1.2 Clientelism Clientelism is an asymmetric, recurring relationship between patrons and clients (Powell 1970; Lemarchand 1972; Scott 1972; Kaufman 1974). Clientelism involves political patrons offering inducements to individuals, such as voters or business owners, in exchange for their support. Inducements include goods and services often reflected in public welfare spending as targeted redistributions from the government, but may also come from private sources (Hicken 2011). By their very nature, clientelist benefits are excludable and are denied to some individuals and organizations, thus their distribution (and threat of denial) promotes loyalty and ongoing support for political incumbents. Examples include cash transfers, offers of employment in the bureaucracy, preferential lending arrangements to favored businesses, and employment policies that favor specific segments of the population. In contrast to clientelism, programmatic redistribution involves the allocation of goods and services that cannot be withdrawn if the beneficiary stops supporting the political incumbent. The crucial difference between clientelist versus programmatic distribution is not the good or service itself, but the criteria used to distribute it (Stokes et al. 2013; Kuo 2018). If a political incumbent offers to subsidize rental payments 6

This characterization of electoral autocracies is consistent with influential work on this topic by Schedler (2002, 2006, 2013) and Levitsky and Way (2010).

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

Theory

79

to everyone meeting certain criteria and is unable to withdraw them despite recipients not voting for the incumbent, they are programmatic. If those payments are dependent on how someone votes, they are clientelist. Clientelism is especially prevalent in developing countries where low incomes raise the value of such benefits (Brusco et al. 2004). As countries develop, the value of clientelism declines for two main reasons. First, development increases the physical and occupational ability of citizens, thereby weakening the power of traditional patron–client networks by reducing the ability to develop, monitor, and maintain them (Kitschelt and Wilkinson 2007; Kitschelt et al. 2010). Second, an increasing supply of resources is needed to support clientelism as incomes rise. This is not a problem if political leaders have access to resources that keep pace with rising incomes, but access to clientelist resources typically becomes more constrained as incomes rise. Additionally, resistance to clientelism is likely to increase when the supply of resources entails transfers from middle-/upper-class citizens to the poor (Stokes 2007). Preserving clientelist policies as the economy develops can reduce economic performance (Keefer 2007), and elevate levels of perceived corruption (Kitschelt et al. 2010), in turn amplifying discontent with clientelist strategies among middle/upper class citizens as well as business owners and managers. Indeed, the association between clientelism and corruption is so strong that indicators of corruption have been used as proxies for the level of clientelism (Persson et al. 2003; Keefer 2007). Research has shown that businesses report higher levels of corruption in their interactions with government officials in clientelist systems as compared with more programmatic systems (Keefer 2007; Kitschelt and Wilkinson 2007). Overall, clientelism causes distortions to the economy by promoting and sustaining market failures by contributing to the violation of two conditions of the First Fundamental Theorem of Welfare Economies, including information asymmetry and noncompetitive markets, thereby generating a Pareto suboptimal Nash equilibrium. 4.1.3

Clientelism and Political Regimes

Comparative politics researchers have firmly established that clientelism varies across political regimes (Magaloni 2006; Blaydes 2010; Hicken 2011; Miller 2020). In essence, political regimes institutionalize these mechanisms, thereby locking in Nash equilibria that are Pareto suboptimal. In autocracies that do not hold multiparty elections, such as Saudi Arabia or Vietnam, patronage is targeted at political and

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

80

China’s Chance to Lead

business elites in order to secure their loyalty and ongoing support for the regime.7 In these regimes, clientelist benefits may also be offered to segments of the citizenry, but this is typically accompanied by intimidation tactics which reduce the need for clientelist resources. Thus, in closed autocracies, clientelist benefits are extended to a relatively narrow subset of the population. Sometimes autocrats choose to hold multiparty elections, as in electoral autocracies. Typically, such elections are introduced so the government can either qualify for foreign aid or access international lending facilities (Carothers 2002; Schedler 2006), or to promote regime survival by facilitating the cooption of elites or by revealing information about citizens’ concerns. Autocratic elections that portray an image of invincibility discourage divisions within the ruling party and deter challengers from outside the party (Lust-Okar 2006; Magaloni 2006). Elections also reveal information about citizens’ concerns and thereby offer a way to monitor the performance of local leaders (Magaloni 2006; Blaydes 2010). But allowing elections can invigorate political opponents. Since 1946, autocrats have left office in about one out of five elections (Hyde and Marinov 2012). To ensure elections are in their favor, autocrats manipulate electoral processes with a wide variety of techniques, such as gerrymandering, candidate qualifications to run for office, and control of the media. But even more common is for autocrats to cultivate loyal support and punish opponents through clientelism. The awarding of patronage to those who vote for the incumbent and the denial of resources to those who don’t (and the threat of doing so) is pervasive in such regimes. Indeed, scholars regard clientelism as the central mechanism affecting autocratic elections (Lust-Okar 2006; Blaydes 2010; Miller 2020). For autocratic leaders to retain power in the face of persistent electoral threats from opposition parties requires the ruling party to develop more extensive clientelist networks than exists in autocracies without multiparty elections, and this in turn leads to a more institutionalized ruling party apparatus. The development of a ruling party’s administrative apparatus is necessary in order for autocratic rulers to broaden their clientelist networks and to control and distribute resources to supporters of the regime in a targeted and systematic way that optimizes the incumbent’s political benefit. In order for rulers to 7 Closed autocracies such as oil monarchies can have higher GDPs per capita because they have abundant resource to sustain patronage resources. In the absence of this, they must transition to another regime type or increase intimidation tactics.

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

Theory

81

extend their appeal to a wider swath of the citizenry, private business typically gains more opportunities, generating more jobs and improving employment opportunities and working conditions for the poor. Malaysia, for example, is noted for its huge infrastructure projects which offer an economic and political windfall to regime supporters (Gomez and Jomo 1999; Gomez et al. 2017). Resistance to clientelism tends to increase when the supply of resources entails transfers from middle-/upper-class voters to poorer voters (Stokes 2007). Resistance becomes further entrenched by elevated perceptions of corruption and the distortionary effects of clientelism on economic performance (Keefer 2007; Kitschelt et al. 2010). Consequently, the advent of genuinely competitive multiparty elections, as in democracies, leads to politicians being held more accountable via greater transparency in public spending and more effective oversight by a stronger and more independent judiciary. These features raise the costs of clientelism sufficiently to make it politically unappealing. For example, McMann et al. (2020) find that the mere introduction of elections, as in semicompetitive autocracies, increases most forms of corruption. But once the quality of elections begins to improve, as in democracies, most forms of corruption decline. The key implication is that clientelism predicts electoral autocracies and electoral democracies will want more financing than closed autocracies. 4.1.4

Public versus Private Control of Resources across Political Regimes

SOEs can either be used to reduce market failures or they can magnify and sustain them. Whether and how this happens depends on the incentives motivating leaders to bolster their hold on power by utilizing public resources to fulfill their political objectives. I argue the reliance on SOEs by autocrats in electoral autocracies explains why those regimes attract more Chinese spending than electoral democracies despite comparable electoral pressures to attract funds for their clientelist networks. According to the First Fundamental Theorem of Welfare Economics, government intervention may be able to increase social welfare when the market for any good yields a natural monopoly, has public goods characteristics, or goods with large externalities (Putnin‚š 2015). However, political imperatives can push leaders to exploit SOEs and the resources they control to help them achieve political objectives. The means and objectives of government intervention vary by political regime, which in turn vary in the extent they grant political rulers ownership and control

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

82

China’s Chance to Lead

of the corporate sector.8 Autocracies manifest higher levels of state controlled entities that can be directed to support rulers’ hold on power. But the manifestation of state control varies depending on the institutional structure of the authoritarian regime. In closed autocracies, political rulers rely on a narrow support coalition to retain power for a potentially indefinite period of time. They also rely heavily on the centralized control of resources to distribute patronage, co-opt rivals, and to secure compliance from the masses. Through state-owned enterprises that control the supply of essential inputs necessary for commerce (energy, financing, transportation, communications, basic chemicals and materials, construction, real estate), the state ensures private owners of capital remain dependent on the political incumbents that control SOEs. In short, private capital serves the interests of political rulers. In autocracies, state control of the corporate sector is vital to leaders’ ability to allocate resources to political allies (and deny it to opponents) and to preserving their hold on power (Carney 2018). Although private ownership may be prevalent in closed autocracies (e.g., China), it is typically confined to small and medium-sized enterprises with the state retaining control over those firms and industries that are vital to downstream economic activities. For example, SOEs that control the supply of essential inputs necessary for commerce (energy, financing, transportation, communications, basic chemicals and industrial materials) enable political incumbents to keep private business owners dependent on them. Via these mechanisms of resource control, political rulers attempt to ensure private capital does not deviate too far from their interests. Consequently, incumbent rulers can use the resources of stateowned enterprises for priorities that may not align with the interests of private capital (e.g., more abundant capital available at lower interest rates for SOEs than private firms). In electoral autocracies, the ruling party must develop more extensive patronage networks than exist in closed autocracies due to electoral pressures associated with semi-competitive elections, and this in turn leads to a more institutionalized ruling party apparatus. In order for rulers to extend their appeal to a wider swath of the citizenry, private business typically gains more opportunities, such as better access to loans or participation in public procurement projects often via subcontracts in the context of large infrastructure projects. The development of the 8

This follows the argument of Carney (2018), but modifies political regime categories. The emphasis in that book was on varieties of authoritarian capitalism. Here, it is on China’s influence. Thus, I focus on types of democratic regimes among emerging economies as well.

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

Theory

83

economy and the expansion of opportunities for private business contribute to a decline in SOEs’ crowding out effects. However, the ruling party will retain residual rights of control, as the controlling shareholder of a partially state-owned enterprise for example, or as the primary contractor responsible for a government-funded construction project, or when government-owned banks decide which private firms receive loans. The political need to expand opportunities for the private sector leads to the expectation that SOEs will be less dominant in electoral autocracies than in closed autocracies. Nevertheless, state controls remain necessary in order for autocratic rulers to control and distribute resources to particular regions or constituencies through their patronage networks (see Gomez 1994 on Malaysia; Diaz-Cayeros 1997 on Mexico; Scheiner 2006; Greene 2010; Carney 2018 on East Asia). For example, political incumbents can draw upon and divert the financial resources of SOEs to clientelist networks by offering patronage jobs and preferential treatment to favored businesses (loans, land, lax regulatory enforcement, public procurement contracts, subcontracting opportunities, etc.), and deny them to opponents. The opacity surrounding SOEs from autocracies makes them especially useful tools for achieving political objectives, thereby exacerbating market failures by reducing public information about prices and market competition, the first two conditions of the First Fundamental Theorem of Welfare Economics. Opacity is important to delivering the targeted distribution of resources to clients so as to prevent accusations of corruption or unfair treatment by private business owners and the citizenry. Indeed, research by the OECD (2018) highlights the potential for SOEs to promote clientelism as indicated by the prevalence of perceived corruption.9 According to the OECD report, corruption is “very relevant to public procurement (and particularly in utilities and infrastructure), which is at the heart of efforts to close infrastructure gaps.” Another OECD report (2015, p. 10) remarks, “One of the characteristics that make [the infrastructure] sector especially prone to corruption is the frequent monopoly situation, in which those who control the entities receive large rents.”10 The OECD (2018) goes on to state that a key feature related to the prevalence of corruption in infrastructure is “the relative predominance of SOEs.”

9 Keefer (2007), for example, uses corruption as a proxy for clientelism (see also Hicken 2011). 10 OECD (2015), OECD Guidelines on Corporate Governance of State-Owned Enterprises. www.oecd.org/corporate/guidelines-corporate-governance-soes.htm. (accessed February 20 2022).

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

84

China’s Chance to Lead

For example, Malaysia is noted for its huge infrastructure projects which direct substantial state resources to regime supporters (Gomez and Jomo 1999; Dettman and Gomez 2020). The onset of the Asian Financial Crisis provided an opportunity to challenge the ruling party’s stranglehold on power by business owners not connected to political incumbents’ patronage network (Johnson and Mitton 2003). This challenge marked a dire threat to the political incumbents, but they survived and subsequently nationalized and consolidated the country’s largest banks so as to exert stronger controls over access to state resources such as business loans and public procurement contracts (Carney 2018). But as the economy develops and incomes rise, redistribution from middle/upper class voters to poorer voters fuels a desire for change. Indeed, a danger associated with relying on clientelism to retain power is accusations of corruption by political challengers. This was a critical reason for UMNO’s loss of power in 2018 after uninterrupted rule for over six decades. In electoral autocracies, accusations of corruption, fueled by a lack of transparency in deciding how resources are allocated, lead political challengers, typically funded by private business, to dismantle the state-run patronage system when they claim power. The handover of power to the political opposition often leads to the adoption of electoral democracy which, in turn, contributes to more opportunities for private business. Politicians become more dependent on financial support from the private sector for election campaigns, and weak checks on executive power by the legislature and courts contribute to a continuing reliance on clientelism for building and sustaining clientelist arrangements with private business while reducing the footprint of the state sector (Gomez and Jomo 1999; Kang 2002; Haber 2013; Djankov 2015; Carney 2018). SOEs remain beneficial and widely used in the context of developing countries where private business is incapable of financing large-scale, long-term projects, for example, but their relative dominance will be reduced in comparison to their autocratic counterparts. Through democratization, political competition becomes relatively unrestricted and characterized by members of different political parties occupying political office across time and space. Heterogeneous interests gain representation in the legislature, and checks and balances ensure that no single actor or party monopolizes the control of politically valuable information and resources (Lijphart 2012). Consequently, democratic political systems have more transparency and lower control of corporate resources than autocracies (Carney 2018). Typically, state ownership occurs for economically justifiable reasons, such as a natural monopoly or projects which would not otherwise be funded by private investors

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

Theory

85

(Putnin‚š 2015). The state will become involved in corporate activities when private capital is incapable of or unwilling to pursue them, as with large infrastructure projects that promote economic growth and expand opportunities for private business.11 The control of resources is therefore not monopolized by political incumbents or by the state. Instead, property rights are designed to protect private owners against state encroachments. The reliance on popular support at election time requires the state to minimize its crowding-out effects and its drain on state finances (via taxes), thus leading SOEs to maintain a diminished presence in the economy. Consequently, owners of small and medium-sized enterprises maintain higher levels of representation and influence than in authoritarian settings. Although political incumbents in electoral democracies remain beholden to the interests of private capital, these leaders can still wield government powers to distribute patronage to key political supporters via state-owned enterprises or other organizations allied to political incumbents (e.g., via government tenders). Weak checks on executive power by the legislature and courts enable these types of activities to occur without the threat of punishment. The key constraint is the competitive party system that promotes monitoring and transparency and prevents deep entrenchment of state resources in favor of one political party, as in electoral autocracies with dominant parties. In liberal democracies, SOEs will be even less prevalent due to strong institutional mechanisms for ensuring politicians are accountable to private business interests and the citizenry, thus reducing the ability to use SOEs for patronage. Specifically, judicial and legislative checks on executive power limit the ability of the executive to engage in corrupt, nontransparent patronage via SOEs. This in turn leads to a further erosion of SOEs’ presence in the corporate sector so as not to crowd out opportunities for private business. The state will become involved in corporate activities when private capital is incapable of or unwilling to pursue them, such as large infrastructure projects that promote economic growth and expand opportunities for private business.12 In summary, the size and scope of the state sector varies across regime types, declining from closed autocracies to liberal democracies. The key point for the argument is that clientelism produces a much larger need for patronage resources in electoral autocracies and electoral democracies in comparison to closed autocracies, but the crucial difference between 11

This explains the high level of SOEs in emerging democracies such as India. 12 This accounts for the high level of SOEs in emerging democracies such as India, the Philippines, or Indonesia.

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

86

China’s Chance to Lead

electoral autocracies and electoral democracies is that state control of the corporate sector is much stronger and more encompassing in electoral autocracies than in electoral democracies. 4.2

The Supply Side: Characteristics of Chinese Foreign Infrastructure Spending

To identify the salient supply factors affecting Chinese foreign spending in the context of BRI infrastructure projects, I focus on three actors, including the Chinese Communist Party, state-owned entities, and private firms. 4.2.1

The Chinese Communist Party

As discussed in Chapter 2, the CCP has sought the rapid deployment of infrastructure spending to alleviate excess capacity that was exacerbated by the financial stimulus during the global financial crisis. But in the long-term the CCP seeks to leverage infrastructure projects to boost its total factor productivity by moving low-end manufacturing offshore and by promoting the adoption of Chinese technical standards. The clearest articulation of how China plans to achieve these aims comes from the “Vision and Actions” statement (NDRC 2015). The document was drafted by the National Development and Reform Commission (NDRC) together with the Ministry of Foreign Affairs (MOFA) and the Ministry of Commerce (MOFCOM) by pulling together previous ad hoc statements into a coherent policy framework and guide to future planning among member states forming the Silk Road Community. It specifically says, “The Belt and Road Initiative aims to promote the connectivity of Asian, European, and African continents and their adjacent seas.” To do this, it prioritizes expediting international commerce by adhering to agreed Chinese technical standards, not only in traditional modes such as air, pipelines, rail, road, and sea, but also in border and customs controls, power supplies, and telecommunications and related digital technologies. Additionally, it seeks to reduce costs and risks along supply chains via unimpeded trade and investment. To achieve these objectives, speed matters. Alleviating excess capacity requires speed so as to enable heavily indebted infrastructure SOEs to reduce their debt load and minimize layoffs; speed also matters for promoting the adoption of Chinese technical standards before other standards get adopted by recipient countries and also to influence formal standards negotiations and agreements in the context of international standards development organizations. To this end, the CCP has created

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

Theory

87

subsidized financing mechanisms to enable China’s SOEs to offer cheap and fast infrastructure development proposals that incorporate Chinese technical standards. Chinese private high-tech and manufacturing firms also have incentives to participate in BRI projects due to such standards generating a long-term competitive advantage that locks-in recipient countries and locks-out Western competitors, mirroring the benefits of China’s Great Firewall to the development of indigenous high-tech private firms such as Baidu, Alibaba, and Tencent. But to identify the relative attractiveness of Chinese SOEs and private firms to different political regimes, it is necessary to enumerate the key characteristics that make them attractive or repellent. 4.2.2

State-Owned Entities

China’s state-owned entities have been the primary actors spearheading infrastructure financing and development projects. Their core strength, due to the confluence of characteristics enumerated below, is their unrivaled capacity to establish a first-mover advantage in countries with relatively high political risk. An important enabling factor was the government’s massive stimulus during the global financial crisis, creating the immediate need to find infrastructure projects outside the country to avoid debt repayment defaults and massive layoffs that could jeopardize the stability of China’s financial system and cause broader social instability. China state entities possess a number of characteristics that, in combination, make them particularly well suited to initiating projects that Western private firms, multilateral development banks (MDBs), or even Chinese private firms would be unwilling to pursue in the context of developing countries. First, Chinese state entities such as SOEs possess a double bottom line. They pursue profits, but they often do not maximize them. Instead, they balance profitability with other political objectives specified by the government such as employment stability, regional development, or entering foreign markets. While SASAC reforms starting in 2003 contributed to governance reforms intended to eliminate loss-making businesses and improve profitability, it is important to note they are not profit-maximizing organizations, like private firms. The government retains appointment powers, which ensures that SOE senior managers are moved to different party posts, thus preventing their entrenchment with one organization and ensuring their loyalty to the party first and foremost (Lin et al. 2020). Given the need to be profitable, this is not to say that SOEs will blindly initiate projects without regard for the financial risks. This is reflected in comments to Peter Cai (2017) by a chief

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

88

China’s Chance to Lead

investment officer from one of China’s largest state-owned financial institutions: “I prefer to invest in places like Canada and Australia, where I can get safe and decent returns. However, where I have been ordered to invest in OBOR countries, I will only allocate the minimum amount.” This comment underscores the point that SOEs retain some autonomy in deciding which projects to pursue (Jones and Hameiri 2020). Second, government-to-government negotiations can play an influential role in large, complex infrastructure projects. Coordinating multiple projects, such as port-park-city ecosystems that link to a national or even international rail line is greatly facilitated by high-level government-togovernment deal-making and support. Even China’s massive SOEs benefit from government support with the development of such ecosystems which require coordination among multiple SOEs and actors from other foreign countries and often also MDBs. Government-to-government negotiations and agreements are also important to enabling the respective governments to grant special privileges or to make exceptions for particular projects if they are deemed of sufficient national interest. For example, China’s desire to promote its high-speed rail technology contributed to the offering of a 2 percent interest rate on the Indonesian High-Speed Rail (HSR) project. However, Chinese financiers have demanded a 2.5 percent interest rate on subsequent projects in order to ensure they are profitable (Cai 2017). Such government-to-government deal-making can be especially helpful if the host government faces an imminent election with budget pressures, as was the case for Indonesia’s HSR project. SOEs are more likely to enable and carry out such agreements than private firms that place a stronger emphasis on maximizing profits. Third, Chinese SOEs retain preferential access to cheap financing from state-owned policy banks (and related state-owned entities such as the China Development Bank and the Export-Import Bank of China) and from subsidies earmarked for BRI projects deemed to be of national strategic interest.13 China’s persistent trade surpluses have contributed to large foreign currency reserves, principally held in dollar-denominated assets, typically US treasury securities. The BRI offers a way for Chinese financial institutions to more productively deploy this surplus capital. This financial assistance has often been supplemented by the central government through diplomatic channels and government programs such as the Forum on China-Africa Cooperation (FOCAC). Fourth, Chinese SOEs can accommodate opacity, which can be especially helpful for regimes reliant on clientelism (Gelpern et al. 2021; 13

See Dutton et al. (2020) for additional details.

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

Theory

89

Ghossein et al. 2021).14 Most of the lending is made through bilateral agreements between the lending banks and recipient governments, which enables the terms of the funding agreements to remain private. As mentioned earlier in the discussion on the demand side, recipient country leaders often prefer to hide the targeted distribution of funds to their clientelist network to avoid accusations of corruption. SOEs are more capable of accommodating these requests than private firms accountable to outside shareholders. Fifth, Chinese SOEs dominate the construction and infrastructurerelated industries in China (e.g., transportation such as railways, road construction, and ports; energy such as utilities; and basic materials such as steel and cement), which typically matches arrangements in developing countries, facilitating collaboration that is opaque, and which easily enables government-to-government negotiations and deal-making. Sixth, Chinese SOEs possess enormous economies of scale and scope that grant them cost advantages and enable them to tackle large infrastructure projects. Fortune 500 ranks the largest companies by revenue globally; in 2014, China had seventy-seven SOEs on the list and they are predominantly from industries active in infrastructure development, such as construction, energy, telecommunications, and banking (Lin et al. 2020). Seventh, SOEs can handle risky locations and projects that private firms avoid. Infrastructure investments possess several characteristics that elevate their risk above that for typical foreign investments, including large investment amounts, long payback periods, high monopolization, and with the consumer often being the host government. Infrastructure projects located in developing countries have additional risks such as weak rule of law, policy frameworks, and regulatory enforcement. Added to this is the proclivity for host governments to grab profits of foreign firms through variable taxation, fines, or even outright expropriation (Henisz and Zelner 2010). Collectively, these contribute to high levels of political risk that jeopardize the profitability of such investments (Jiang et al. 2019). However, SOEs possess risk-mitigating features typically unavailable to their private counterparts, such as: (1) the backing of the Chinese government which may engage in government-to-government negotiations and agreements to ensure the project’s successful completion; (2) an alignment of interests with host country political leaders to ensure the provision of resources that can be used for their clientelist networks; (3) SOEs’ 14

Gelpern et al. (2021) “How China Lends” PIIE; Ghossein et al. (2021), “Public Procurement, Regional Integration, and the BRI”.

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

90

China’s Chance to Lead

enormous economies of scale and scope as well as experience developing infrastructure in a similar, institutionally weak, environment; and (4) the Chinese government’s objective to promote its own standards which, over the long term, may compensate for short-term costs. Eighth, China adheres to a policy of non-interference in the domestic affairs of foreign countries. Consequently, China does not impose conditionalities on its infrastructure investments like Western countries or affiliated MDBs, thus reducing host country resistance, especially among autocratic regimes. But this in turn may contribute to Chinese external financial cooperation being more open to influence by recipient country political leaders (Dreher et al. 2018; Ghossein et al. 2021).15 Ninth, SOEs can make long-term commitments, especially common and necessary in the context of large, complex infrastructure projects typically structured as public-private partnerships that last for twentyfive years or longer. For multiple projects that require coordination, such as port-park-city ecosystems, host governments are making decades-long commitments. The government backing of SOEs helps assure the longterm mutual commitment to such projects. Tenth, by the Chinese state actively coordinating the implementation of common standards, Chinese SOEs can enhance the value of infrastructure projects through network effects spanning thousands of projects across dozens of countries. Although SOEs often possess autonomy in selecting and executing projects, they nevertheless follow the same set of technical standards incorporated into them following explicit guidance from the Chinese government. For example, in 2015 the BRI Action Plan stipulated that 500 national standards should be translated into foreign languages to make them accessible to international audiences (SAC, 2015). In 2017, China signed the “Joint Initiative on Boosting Standards Cooperation” to promote the BRI with a dozen countries. In June 2019, China officially announced that it had signed eighty-five cooperation agreements on technical standardization with forty-nine countries and regions along the Belt and Road (Office of the Leading Group for Promoting the Belt and Road Initiative, 2019). While each of these characteristics is important in its own right, when taken together they yield several important takeaways. First, the first eight characteristics16 collectively contribute to greater speed in the 15

Dreher et al. (2018) find that the birth places/regions and ethnic groups of political leaders receive larger amounts of Chinese aid, a result that does not obtain for the portfolio of projects of the World Bank. 16 These eight characteristics include: (1) a double bottom line; (2) benefiting from government-to-government negotiations; (3) privileged access to cheap capital; (4) accommodating opacity; (5) dominating infrastructure-related industries; (6) enormous

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

Theory

91

delivery of infrastructure financing and development, assuring a vital first-mover advantage for the deployment of Chinese technical standards to be discussed in the next section. Second, the Chinese SOE characteristics collectively enhance the delivery of clientelist benefits for political incumbents of recipient countries. SOE dominance in infrastructure combined with their massive size enables them to deliver large patronage opportunities. SOEs’ opacity and government-to-government coordination enables them to work with political leaders to evade public scrutiny of how contracts or subcontracts are allocated. The combination of speed and patronage resources is especially attractive to political leaders of certain regimes confronting elections. 4.2.3

Private Firms

Unlike Chinese SOEs, but very much like private firms in general, Chinese private companies are primarily focused on maximizing profits and minimizing risk. This leads them toward institutional environments with strong investor protections, few crowding out effects, and low policy uncertainty, ceteris paribus. Based on the prevalence of these characteristics among political regimes, liberal democracies are most and closed autocracies least preferred. But among autocratic regimes, electoral autocracies are favored to closed autocracies due to their lower crowding out effects. However, two features of autocratic regimes may enhance their appeal to Chinese private firms relative to democracies. First, Chinese private firms are experienced in working in the environment of a developing economy that lacks strong institutions, strong property rights, and they are familiar with developing products and services that cater to lowincome consumers. They are also familiar with operating in an autocratic country where the government can have considerable influence on business in ways that differ from democratic regimes, such as a lack of privacy protections and the accompanying implications for digital products and e-commerce, and the reliance on a single ruling party and the importance of cultivating good relationships with party officials. For example, Wanda is one of China’s largest commercial property developers that succeeded due to its founder’s relationship with CCP officials. Wanda’s commercial development projects helped generate the growth CCP officials needed for promotion, yielding a mutually beneficial relationship. However, this strategy was predicated on the ruling party’s economies of scale and scope; (7) the willingness to invest in risky locations; and (8) Chinese state policy of non-interference in the domestic affairs of foreign countries.

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

92

China’s Chance to Lead

uncontested hold on power. Using the same approach to purchase a landmark hotel in Madrid (the Edificio Espana), Wanda encountered unexpected problems due to changes in the ruling political party, which ultimately led the company to sell the property soon after the purchase (Meyer et al. 2017). Second, private Chinese firms can benefit from following the lead of SOEs that may have already mitigated some of the political and market risks of the foreign country. For example, the development of an industrial park with modern telecommunications, transportation, and power infrastructure can reduce risks for private firms. Such parks serve several strategic objectives (Zeng 2016). First, they increase demand for Chinese-made machinery and equipment. Second, by producing overseas and exporting to Europe or North America, Chinese companies can avoid trade barriers imposed on exports from China. Third, they assist China’s efforts to offshore low-skill manufacturing and lift manufacturing capabilities at home. Finally, they help create economies of scale for overseas investment, and assist less experienced SMEs to venture overseas. Often, industrial parks are part of a special economic zone that includes policies to reduce business costs (e.g., tax reductions) as well as financial incentives to attract them (e.g., subsidized loans for projects aligned with the park’s commercial objectives). Insofar as electoral autocracies are predicted to attract a larger share of SOE construction spending for projects such as industrial parks, they are also likely to attract a relatively larger share of private investment. With regard to the aim of spreading standards (to be discussed further below), the government and private firms possess a mutual interest in promoting these since they grant an advantage to Chinese firms relative to Western firms. The Great Firewall of the Chinese internet offers a helpful illustration. The CCP’s effort to control information combined with industrial policy protectionism in conjunction with dynamic and competitive high-tech enterprises has allowed Chinese internet companies to develop and thrive in a cyberspace environment that is separated from the global, US-dominated internet. This has contributed to a deep interdependence between the Chinese government and China’s largest internet-based companies, such as Baidu, Alibaba, and Tencent. The overwhelming success of this model provides the template for understanding the deployment of the new generation of digital technologies and standards in the context of the Digital Silk Road (DSR). As discussed in Chapter 2, the Chinese government has made a substantial effort to nurture its own firms at the forefront of novel Industry 4.0, smart infrastructure, and smart cities technologies. It is

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

Theory

93

also investing heavily in building undersea cables and satellite networks to enable high-speed data transmissions linking back to China. And as mentioned earlier in relation to China’s “Vision and Actions” statement, the government is actively promoting the adoption of its technical standards as part of its infrastructure investments. This effort on the part of the government to promote the adoption of Chinese technical standards effectively locks-in foreign countries to Chinese technologies and firms while locking-out foreign competitors. Consequently, private Chinese firms are more willing to invest in developing countries as part of the DSR. At the same time, the pursuit of improving technical expertise and the greater technological readiness of advanced economies to utilize new digital technologies products and services makes liberal democracies especially attractive even if they do not adopt Chinese standards. In these markets, mergers and acquisitions are likely to be a more common form of investment as Chinese private firms seek to acquire technical capabilities. In summary, Chinese private firms are expected to display similar investment behaviors as their Western counterparts in terms of favoring low risk markets in liberal democracies. At the same time, they possess certain advantages and incentives to invest in emerging economies. This is especially the case where SOEs reduce risks and create attractive investment opportunities as with the construction of new industrial parks, or where technology bundles help to lock-in the foreign country to Chinese technical standards, as with smart cities, smart ports, or especially in the context of port-park-city ecosystems. Thus, private firms are more likely to invest in developing countries in tandem with or shortly following the initiation of construction projects by SOEs. 4.2.4 Summary SOEs play a vital role in the supply of Chinese foreign spending in the context of the BRI. They are responsible for implementing the objectives of the CCP and pave the way for private firms to enter developing countries, thereby enabling the offshoring of low-end manufacturing and promoting the adoption of Chinese standards. In this regard, it is vital to focus on the compatibility of Chinese SOEs’ characteristics with host country political regimes in order to identify where Chinese infrastructure spending is most likely to occur, which in turn can facilitate the adoption of Chinese standards by those countries as well.

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

94

China’s Chance to Lead

4.3

Compatibility of Chinese Infrastructure Spending by Political Regime

In this section, I discuss the compatibility of Chinese infrastructure financing and development spending with respect to each political regime. I begin with a brief summary of clientelism, public-private control of resources, and Chinese SOE characteristics in turn. I then discuss how they combine to yield varying outcomes for each political regime. 4.3.1 Clientelism Foreign infrastructure spending can be a very valuable source of patronage resources to support a regime’s clientelist network. In emerging economies in particular, such investments are often directed at infrastructure projects connected to their clientelist networks (Barthel et al. 2014; Burgess et al. 2015; Do et al. 2017). Given the high correspondence between clientelism and corruption, with the latter often used as a proxy for the former (Keefer 2007; Hicken 2011), it is notable that the construction industry stands out for its high level of corruption, with an estimated 10–30 percent of projects costs misappropriated annually worldwide (Matthews 2016). Through the potential to deliver targeted patronage to clients as well as boost economic growth, it is easy to see why foreign investment in infrastructure projects would be highly attractive to political incumbents in developing countries. Due to the structure of the clientelist network increasing in size and scope for electoral autocracies relative to closed autocracies, political rulers in electoral autocracies are expected to display a markedly higher demand for Chinese foreign spending. Political incumbents of electoral democracies have similarly large clientelist networks to fund as their electoral autocracy counterparts, though clientelism is driven more by competitive dynamics between political parties than sustaining the ruling party’s dominance. This competitiveness can force a reduction in the reliance on clientelism over time due to accusations of corruption and as redistribution from the middle/upper classes fosters growing discontent. Consequently, programmatic policies become more prevalent over time, especially in liberal democracies with courts more independent of political influence coupled with strong protections for the media who can police against corrupt activities. 4.3.2

Public versus Private Control of Resources

State versus private control of the corporate sector affects whether the allocation of clientelist resources is controlled by political officials or

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

Theory

95

private owners of capital. Although private ownership is more prevalent in electoral autocracies compared to closed autocracies, electoral autocratic states retain residual control rights, thus ensuring political leaders have the upper hand in controlling redistribution via clientelist networks. The key point for the argument is that clientelism produces a much larger need for patronage resources in electoral autocracies and electoral democracies in comparison to closed autocracies, but the crucial difference between electoral autocracies and electoral democracies is that state control of the corporate sector is much stronger and more encompassing in electoral autocracies than in electoral democracies. 4.3.3

Chinese SOE Characteristics

Reactions to foreign spending by SOEs vary depending on whether the SOE characteristics enumerated in the previous section are viewed as beneficial or problematic to the political leaders of the recipient country. Whether a recipient country views the characteristics enumerated above as strengths or weaknesses depends on its political regime. For example, attentiveness to the political priorities of the government above profits is problematic for liberal democracies and potentially for electoral democracies. Likewise, the focus on speed often comes at the expense of adequate attention to a project’s social and environmental impact on the local community. Again, this is more of a problem for democracies. Third, access to abundant financing is a greater benefit to political regimes with large infrastructure financing gaps and large clientelist networks to fund. Thus, electoral autocracies and electoral democracies will place a greater value on China’s abundant financing than closed autocracies. Fourth, more critical views are likely to be voiced by private business in democracies in relation to SOEs that rely on Chinese workers and subcontractors in contrast to those from the host country. Fifth, opaque government-to-government deal-​ making, a lack of transparency when bidding occurs, and the preferential allocation of subcontracting work can generate fierce criticism and accusations of corruption by political opponents of the ruling party, especially in democracies. Finally, Chinese SOE infrastructure spending is generally disfavored in liberal democracies largely because of fewer market failures that enable alternative financing and development options. Altogether, criticisms of BRI projects are more likely to occur in democracies where opposition parties and press freedoms permit critical opinions to be publicly expressed. In autocracies, by contrast, attention to the weaknesses associated with the characteristics of Chinese SOEs

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

96

China’s Chance to Lead

Table 4.2  The compatibility of Chinese infrastructure spending by political regime Most Compatible Characteristics

Least Compatible Characteristics

Closed Autocracy

Overall Compatibility Example

Opacity, Political Speed, Volume of Moderate priorities above Spending profits, Inattention to socio-​ environmental impact Electoral Opacity, Speed, High Autocracy Volume of Spending, Political priorities above profits, Inattention to socio-​ environmental impact Electoral Speed, Volume of Political priorities Moderate Democracy Spending above profits, Opacity, Inatten­ tion to socio-​ environmental impact Liberal Speed Political priorities Low Democracy above profits, Opa­ city, Volume of Spending, Inatten­ tion to socio-​ environmental impact

UAE

Malaysia and Djibouti

Indonesia

Greece

enumerated above is likely to be muted. Overall, China’s SOEs possess numerous characteristics that make them especially well positioned to deliver infrastructure projects to emerging economies and to autocracies more specifically. 4.3.4

Compatibility across Political Regimes

As displayed in Table 4.2, the demand for resources to fund large state-led clientelist networks in electoral autocracies matches the provision of infrastructure spending by Chinese SOEs with regard to their shared interests in speed, opacity, and alignment of political priorities above maximizing profits. With regard to speed, electoral autocrats face regular elections that require a large and speedy delivery of clientelist funds to ensure regime support; this matches China’s need to quickly alleviate excess infrastructure capacity so firms can pay their debts and minimize layoffs and to win

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

Theory

97

the race for the adoption of technical standards. With regard to opacity, electoral autocrats have an interest in shielding the targeted distribution of funds from public scrutiny; China also favors opacity to prevent collective bargaining by BRI recipients and to shield BRI agreements from thirdparty scrutiny. Electoral autocrats and China’s party-state are also aligned in assigning greater weight to political priorities relative to profit maximization, as would occur with private firms. This allows for government-​togovernment deal-making and can help ensure a high-level commitment to completing infrastructure projects and tailor the project to the needs of the host country’s political leaders. Electoral autocrats also seek a large volume of funds that can be directed toward their loyal supporters and to woo new ones which matches China’s desire and capacity to fund large infrastructure investments. The compatibility of these features (governmentto-government deal-making and funding a large clientelist network reliant on opacity) is further enabled by the dominance of Chinese and host country SOEs in the industries in which developing countries typically seek infrastructure projects, including transportation, energy, and natural resource extraction (see Table 5.6). Finally, the potentially negative socioenvironmental impact of Chinese infrastructure projects is typically muted due to the ruling party’s control of the media. This convergence of interests yields the expectation that electoral autocracies will host the largest share of Chinese infrastructure spending. Liberal democracies display the lowest convergence. Indeed, an increasing number of liberal democracies disfavor and even restrict Chinese SOE investments due to concerns about opacity, corruption, and political objectives overriding purely commercial priorities. Among advanced economies, made up almost entirely of liberal democracies, infrastructure financing gaps are reduced by a reduction of market failures. Thus, there is less demand for Chinese spending since private firms offer an alternate source of funding to cash-strapped governments. Additionally, in liberal democracies the reliance on patronage networks is reduced by institutionalized checks on executive power by the legislature and courts. These stronger institutional arrangements also ensure attention is devoted to an infrastructure investment’s socioenvironmental impact before construction starts. Consequently, the attractiveness of hosting infrastructure investments by Chinese SOEs is greatly diminished in liberal democracies and policymakers and business leaders instead favor investments by Chinese private firms so long as the purchases do not impinge on areas considered sensitive for national security reasons. The one attraction of Chinese spending is speed. When coupled with a large volume of funding, this can be an irresistible combination for political leaders of liberal democracies in the midst of an economic crisis.

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

98

China’s Chance to Lead

Closed autocracies and electoral democracies are predicted to be in the middle. For closed autocracies, the need to divert resources to a narrow clientelist network contributes to compatibility regarding opacity and assigning greater weight to political priorities relative to profit maximization (in turn facilitating government-to-government negotiations). Speed is less important due to the absence of elections and the volume of spending is less important due, in part, to their narrow clientelist network but also because many of these regimes have abundant oil revenue. Criticism of Chinese foreign investment by SOEs is typically even more muted than in electoral autocracies, virtually eliminating concerns regarding socio-environmental impact. The interests of electoral democracies converge with those of China with regard to speed due to election pressures and the desire for a large volume of infrastructure spending to fund a large clientelist network. However, electoral democracies display lower compatibility with Chinese spending because of the former’s robust political competition which offers a partially effective check on corruption and opacity when the ruling party strikes agreements for Chinese infrastructure spending. Insofar as Chinese SOEs are favored for their opacity, this can reduce their appeal. This may be offset by weak legislative and judicial checks on the government’s capacity to divert funds to a clientelist network. Electoral democracies also place greater emphasis on opportunities for private business, and attentiveness to infrastructure investment’s socioenvironmental impact. Finally, it is worth reiterating that China’s policy of noninterference in the domestic affairs of foreign countries enhances the influence of recipient country political leaders in addressing their priorities (Dreher et al. 2018; Ghossein et al. 2021). This is worth highlighting given that the characteristics listed in Table 4.2 vary by host country political regime (i.e., China’s supply-side characteristics are held constant). 4.4 Extensions I consider two additional features that may amplify the compatibility of electoral autocracies with Chinese infrastructure spending, including leaders with an insecure hold on power and a rise in the use of publicprivate partnerships. 4.4.1

Weak Electoral Autocrats

In contrast to the literature which argues higher political instability reduces infrastructure investment (Henisz 2000; Jiang et al. 2019), I

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

Theory

99

argue electoral autocrats with an insecure hold on power will more avidly seek Chinese infrastructure spending. Political rulers of electoral autocracies who retain an insecure hold on power will more avidly seek BRI projects in order to widen their clientelist network through the provision of patronage, and to boost their popular support by improving economic growth more generally. These effects will be stronger in electoral autocracies than in closed autocracies due to the pressures electoral autocrats face from holding multiparty elections, even though they are only semicompetitive. In comparison to electoral democracies, rulers of electoral autocracies have greater control over the media allowing them to more effectively stifle criticism of projects carried out by Chinese SOEs (e.g., use of foreign workers, quality controls, creating opportunities for local business, adhering to environmental or labor standards). And due to their need to appeal to owners of private business by ensuring the state does not crowd out opportunities for them, either by expanding the role of host country SOEs in these projects or by inviting Chinese SOEs, weak leaders of electoral democracies fearing comparable electoral threats will be less avid in attracting BRI projects in comparison to weak leaders of electoral autocracies. 4.4.2

Rising Reliance on PPPs

PPPs are of increasing interest on the part of China and other developing countries, including both electoral autocracies and electoral democracies. Because PPPs allow political incumbents to leverage limited financial resources to attract private financing, they can be used to increase the total number of projects and the total volume of infrastructure financing. Electoral autocrats are less likely to be constrained by government debt limits (and more capable of ignoring them) than their counterparts in electoral democracies. It follows that weak electoral autocrats are the most likely to engage in such risky leveraging activities. China also has an increasing interest in the use of PPPs. While Chinese SOEs will continue to play an important role in foreign infrastructure spending, there is growing interest in projects that include private firms, as with public-private partnerships. Also, as infrastructure financing gaps narrow, the private sector will become more engaged in infrastructure spending, as in advanced economies. In the context of autocracies, where SOEs are the dominant infrastructure players, electoral autocracies are likely to be more attractive to PPP structured infrastructure projects due to the relatively greater presence of private capital in electoral autocracies than in closed autocracies, yielding more opportunities for Chinese SOEs to collaborate with or subcontract to private firms. Additionally,

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

100

China’s Chance to Lead

Chinese private firms are more attracted to investing in electoral autocracies compared to closed autocracies. This is of increasing importance to BRI projects as the Chinese state reduces financial support for SOE investments in order to prevent the depletion of its foreign exchange reserves and to reduce exposure to bad loans by the state’s policy banks. Thus, Chinese SOEs will favor projects located in electoral autocracies over closed autocracies to the extent they depend on collaborating with private firms while preserving the state’s residual control rights. 4.5

Adopting Chinese Standards

There are two alternative ways to establish technical standards. The first regards the establishment of formal standards which are decided via negotiations and compromise in the context of standards development organizations (the International Organization for Standardization, the International Telecommunication Union, the International Electrotechnical Commission, etc.). Private commercial enterprises are the primary actors represented in these organizations (Büthe and Mattli 2011). In the ideal case, these organizations select a standard because it offers the best technical solution to common problems resulting from a lack of interoperability (Brunsson and Jacobsson 2002). However, commercial considerations such as the prevalence of a standard used in the marketplace can also influence standardization negotiations (Werle and Iversen 2006). In the end, standards are the outcome of negotiations and compromise among competing private organizations that can afford participation in the relevant standards development organization (SDO). The second method concerns de facto standards which are the result of the market dominance of a given company’s product which impacts the development of other products. For example, the market dominance of Google Android and the Apple iOS smartphone operating systems makes them the de facto standards for most smartphone software developers. Neither formal nor de facto standards are laws or legally binding state regulations. They are voluntarily negotiated and accepted private selfregulation that are enormously powerful in practice, with far-reaching implications (Brunsson and Jacobsson 2002; Brunsson et al. 2012). Decisions regarding the adoption of formal standards can be characterized as distributional choices along the Pareto frontier (Krasner 1991). Players compete to decide which standard will be selected, as in a Battle of the Sexes game in which both players prefer to abide by a common standard, though they favor their own standard over that of their competitor. The power that actors wield in these negotiations is largely due to their technical expertise, the ability to establish and exploit

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

Theory

101

a first-mover advantage, active participation, and the capacity for multiple actors with shared interests to speak with one voice (Ruhlig and ten Brink, 2021). As mentioned earlier, commercial considerations influence these negotiations – larger firms usually possess greater technical expertise, they are more capable of funding or buying cutting-edge innovations and quickly commercializing them to gain a first-mover advantage, and they have the resources and technical experts to actively participate in multiple SDOs (including chairmanships, committees, and subcommittees). Larger firms also have an advantage in promoting coordination among industry representatives to support a given standard proposal. With regard to each of these dimensions, China’s party-state effectively bolsters the negotiating power of its representatives (Chen and Kang 2018; Seaman 2020; Ruhlig and ten Brink 2021). The particular advantage of the BRI to these negotiations is the ability for Chinese firms to establish and exploit a first-mover advantage. For China, de facto standards are particularly important for their dual role in directly affecting the standards host countries adopt and for their indirect influence on formal standardization decisions by international SDOs. From the perspective of the international regimes literature (e.g., Krasner 1991), cross-national coordination in the adoption of de facto standards is about addressing market failures rather than distributional choices along the Pareto frontier. Such coordination decisions are often characterized by Pareto suboptimal outcomes where defection is the dominant strategy, as represented by the Prisoner’s Dilemma. The primary focus of this literature has been on identifying how actors can overcome mutual defection. For example, the allocation of the electromagnetic spectrum initially posed pure coordination problems where actors wanted to avoid the mutually undesirable outcome of radio interference due to signals transmitted across state boundaries. Broadcasters have an incentive to defect in order to reach a larger audience, but mutual defection creates greater costs than benefits. Cooperation is more likely when there is iteration and no defined number of plays, when discount rates are low, and when payoffs for cooperation and defection are modest (Oye 1985). Institutions can also promote cooperation by increasing the symmetry and amount of information, raising the cost of illegitimate behavior, promoting convergent expectations, and fostering cross-issuearea linkages (Keohane 1984). But such dilemmas are predicated on there being incentives to cheat. In the context of choosing whether to adopt Chinese standards, such incentives typically do not exist. For example, two nations would derive far greater economic benefits from adhering to common railway standards (e.g., track width) than by choosing different standards. The

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

102

China’s Chance to Lead

concern for most developing countries is not so much about ensuring common standards between countries given a choice of different technical standards, but rather about benefiting from the installation of the infrastructure and accompanying technologies regardless of the standard. Whoever can offer the infrastructure and associated technologies fastest to all interested countries will win the de facto standards race, thereby enabling countries to avoid coordination problems altogether. The actor that builds infrastructure and associated technologies first across numerous countries stands to reap tremendous rewards by locking-in foreign countries to their standards and lockingout competitors. Thus, first-mover advantage and widespread adoption become critical to establishing de facto standards both for the enormous direct benefits as well as the indirect benefits from influencing international standards negotiations. The numerous characteristics associated with Chinese SOEs’ capacity for infrastructure spending thus gain added importance due to their impact on de facto and formal standards adoption. And as discussed earlier, these characteristics can be regarded as strengths or weaknesses depending on the political regime. The earlier analysis suggests electoral autocracies will be the most eager recipients of Chinese infrastructure spending by SOEs, and it is expected they will display a corresponding willingness to adopt Chinese technical standards. 4.5.1

Coordination via Digital Technologies

Although the promotion of China’s technologies, products, and standards are driven primarily by high-tech companies that are not stateowned, the state has considerable influence on their entry into foreign markets in at least three important ways. First, the integration of digital infrastructure into conventional infrastructure projects requires private companies to follow the lead of SOEs, and work closely with them. Second, private firms depend on communications networks that are typically established and controlled by SOEs and the state, such as underseas cables (e.g., the PEACE cable) and satellite networks (e.g., the BeiDou Navigation Satellite System). Consequently, there is relatively greater cooperation between the state and private digital technologies firms than with private firms in other industries. Third, private firms and the state share a mutual interest in promoting their own standards that can exclude competitors or at least put them at a competitive disadvantage. This has been most clearly manifested with the Great Firewall of China which created strong incentives for Chinese private firms to promote Chinese standards and work with the state.

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

Theory

103

4.5.1.1 Commercial and Development Policy Demand for Digital Silk Road Projects Although Chinese digital technologies are dominated by private firms, which tend to favor investing in advanced economies, there are several reasons why emerging economies may be attractive to Chinese private technology firms. First, digital technologies are being integrated into the construction of traditional infrastructure projects which are primarily located in emerging economies. Because many of these projects last more than twenty-five years due to their PPP structure, the benefit of incorporating digital technologies into those infrastructure projects may be enhanced for the host country (e.g., to monitor the quality of the infrastructure over time while also providing additional services such as reducing congestion), which elevates the potential returns to private firms. Second, the building of traditional infrastructure projects eases entry for digital technologies firms regardless of whether the technologies are incorporated into infrastructure projects. For example, once Cosco was committed to upgrading the Piraeus Port facilities in Greece, Huawei and ZTE followed with new telecommunications infrastructure plans and services (Tonchev and Davarinou 2017). Third, Port-ParkCity projects amplify the opportunities for the integration of digital technologies into modern infrastructure projects and ease entry for Chinese technology firms not already partnering in infrastructure construction. Fourth, the construction of hard infrastructure projects across countries enhances the benefit to each nation that is part of a regional transportation and communications network if they abide by Chinese standards, thereby raising the economic benefits of working with Chinese technology companies familiar with these standards (Chen and Lin 2018). Fifth, Chinese firms have developed products and business strategies tailored to low- and middle-income consumers, making their products and services portfolio more appealing to consumers in other developing economies. This raises the demand for digital technologies and consumer services proffered by Chinese technology companies relative to Western competitors. Finally, China’s firms offer products that are more attractive to businesses and governments in developing economies because of the similarity in the level and timing of development compared with advanced economies. For example, China implemented an information and communications technology infrastructure that skipped the need for hardlines. The mammoth size of its firms has generated substantial economies of scale and scope, in turn yielding lower costs compared with firms from advanced economies. Developing economies with sizeable infrastructure gaps are understandably attracted by China’s low-cost solutions. In summary, the combination of advanced Chinese digital

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

104

China’s Chance to Lead

technologies supplied by its private firms coupled with the infrastructure capacity of Chinese SOEs matches the growing demand for these technologies and products among developing economies, especially as population growth and urbanization rates swell. 4.5.1.2 Political Incumbents’ Demand for Digital Silk Road Projects Chinese digital technologies not only represent enormous development policy and commercial opportunities for Chinese and host country firms, but also offer a powerful set of tools to aid autocrats’ hold on power (Feldstein 2019; Frantz et al. 2020). Rulers of both closed and electoral autocracies welcome Chinese technologies for their capacity to identify and monitor their rivals and subordinates as well as surveil, censor, and manipulate their citizens (Greitens 2020). With regard to identifying and monitoring rivals, surveillance technologies such as high-resolution cameras, facial recognition, spying malware, automated text analysis, geo-tracking, socio-behavioral analysis, and big data processing (collecting data on financial transactions, tax returns, bank statements, etc.) can all be deployed with increasing ease to detect potentially problematic individuals. In China, digital technologies have enabled information about the performance of lower-level political officials to be gathered more easily thanks to citizens posting videos to social media sites that provide information about local corruption (e.g., Weibo, the Chinese version of Twitter). Chinese digital technologies also aid autocrats in surveilling, censoring, and manipulating citizens. In autocracies, citizens typically refrain from revealing their true beliefs in public. However, digital surveillance technologies enable the identification of individual grievances and the potential for organized protests before they occur, enabling the regime to proactively address sources of discontent. To filter information that could promote discontent or potentially harmful behavior, censorship is a common tool. China’s Great Firewall, the largest censorship system in the world, is achieved via collaboration between the government and technology and telecommunications companies to monitor and filter content considered undesirable by the CCP. Autocrats may also manipulate the information environment by framing public debates with choices that exclude those considered undesirable by the regime, by disseminating pro-regime narratives on the internet or social media, and diverting attention from negative news. Advances in digital technology now enable this to occur via micro-targeting of individuals. Finally, autocrats can compel citizens to comply with forms of behavior considered desirable. For example, smart cities technologies enable the government to precisely levy fines for even minor offenses such as jaywalking.

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

Theory

105

But this increasingly sophisticated system of rewards and punishments can extend to all types of behavior and government services such as bus passes, passports, jobs, and access to education. 4.5.1.3 Political Regimes and the Digital Silk Road Due to the combination of development policy and commercial reasons and political incumbents’ incentives, electoral autocracies are expected to possess stronger demand for Chinese digital technologies than other regimes among developing economies. Commercially, electoral autocracies host larger private sectors than closed autocracies due to lower crowding-out effects. With regard to development policy, electoral autocracies are expected to also host a larger share of BRI projects than closed autocracies which provide the basis for digital technologies as part of smart cities projects, for example. From the perspective of political incumbents, the same reasons theorized with regard to the larger share of BRI projects locating in electoral autocracies would be expected to apply to the DSR as well. That is, electoral autocrats face stronger threats to their hold on power than rulers of closed autocracies, amplifying their interest in digital technologies that can help them censor, monitor, and manipulate political rivals and the citizenry. These political, development policy, and commercial attributes of electoral autocracies are likely to elevate their demand for digital technologies in relation to electoral and liberal democracies too. More specifically, the larger concentration of BRI projects in electoral autocracies will help attract more Chinese digital technologies firms despite more attractive settings for private firms in electoral democracies. Politically, electoral autocracies would certainly have higher demand than electoral democracies. With respect to liberal democracies, the main attraction is commercial opportunities though development policy may also be important if Chinese technologies remain competitive with those offered by Western firms. Given that these countries are predominantly advanced economies without the substantial infrastructure gaps of emerging economies, they will have far greater preparedness and demand for digital technologies and products. SOEs lead these efforts and the view of their characteristics as either strengths or weaknesses as enumerated in the prior section varies by regime type. Overall, Chinese SOEs are beneficial to leaders of electoral autocracies and closed autocracies; these benefits are reduced for electoral democracies, and further diminished for liberal democracies (except during exceptional circumstances such as an economic crisis). In summary, electoral autocracies are predicted to display the highest demand for Chinese digital technologies and to be the most willing to

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

106

China’s Chance to Lead

adopt Chinese technical standards due to a combination of development policy, commercial, and political reasons. Liberal democracies will have high demand for Chinese digital technologies primarily for commercial reasons, though they will be unlikely to adopt Chinese standards. Closed autocracies will be interested primarily in the political benefits of digital technologies and likely to adopt Chinese standards. Electoral democracies will possess weaker commercial demand than liberal democracies and weaker political demand compared to autocracies for Chinese digital technologies. 4.6 Conclusions Why do countries vary in their participation in the BRI? In this chapter, I examine both the demand- and supply-side characteristics of BRI spending to assess which countries are likely to receive the largest (and smallest) share of Chinese spending. To systematically assess demand-side characteristics, I place countries into one of four political regime categories, including liberal democracies, electoral democracies, electoral autocracies, and closed autocracies. Political rulers in electoral autocracies are posited to display the greatest desire for BRI spending due to their elevated demand for clientelist resources with distribution via state-controlled mechanisms. By contrast, closed autocracies do not hold elections, which reduces the scale of clientelist resources their leaders seek. Electoral democracies hold elections, but state control governing the distribution of clientelist resources is weaker, thereby reducing the power of political leaders to attract Chinese BRI funds to large-scale, state-sponsored projects. Liberal democracies minimize the reliance on clientelism and the role of the state in deciding how such foreign funds are distributed, thereby further reducing the demand for Chinese BRI spending. On the supply side, Chinese SOEs possess a number of characteristics that affect the compatibility with political regimes. Key characteristics include SOEs’ capacity for accommodating opacity into the provision of BRI spending, assigning greater weight to political priorities in relation to profit maximization (the main interest of private firms), access to a large volume of funds, the ability to quickly deliver a project but which also leads to inattention to the socio-environmental impact of a project. These characteristics are most compatible with the demand characteristics of electoral autocracies, followed by closed autocracies and electoral democracies, and least compatible with liberal democracies. It should also be highlighted that China’s policy of non-intervention in the domestic affairs of foreign countries elevates the importance of understanding the demand-side factors affecting BRI spending patterns.

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

Theory

107

The prediction that electoral autocracies will display the highest compatibility with Chinese spending is amplified by two factors. First, electoral autocrats with an insecure hold on power are likely to have an elevated demand for Chinese spending in order to increase the volume of clientelist resources. Second, an increasing reliance on public-private partnerships in which the state retains residual control rights matches the public-private orientation of electoral autocracies. The implication for the adoption of Chinese standards is that electoral autocracies will greatly contribute to their de facto adoption. Electoral autocracies will be unlikely to oppose the inclusion of Chinese standards into BRI infrastructure investments since they do not typically host companies competing to influence the adoption of technical standards relating to the next generation of digital technologies. Via this de facto adoption in the context of the BRI, China can influence their formal adoption in the context of international standards development organizations. But even if Chinese standards are not formally adopted, host countries may remain unwilling to change due to being locked-in to China’s standards.

https://doi.org/10.1017/9781009385879.004 Published online by Cambridge University Press

5

Measuring Clientelism and the Corporate Sector across Political Regimes

In Chapter 4, I theorize clientelism and the public-private orientation of the corporate sector across political regimes will affect the prevalence of Chinese infrastructure spending. Specifically, I argue that regular ­elections held in electoral autocracies and electoral democracies ­contribute to greater demand for clientelist resources than in closed autocracies. The administration of these clientelist benefits will be more state controlled in electoral autocracies than in electoral democracies. Consequently, electoral autocracies will display the highest demand for Chinese infrastructure spending. This chapter establishes the empirical facts regarding political regimes and the prevalence of clientelism and the public-private orientation of the corporate sector. I begin by providing an overview of the distribution of political regimes over time, from 1980 to 2019, for low- and middle-income countries. The data reveal that electoral autocracies constitute around half of all developing countries during the 2010s, the most of any regime type, and their prevalence slowly increases during the last decade. The second most common regime type is electoral democracies, which make up about one-third of all developing countries in 2019. I also display the political regime of each country on a global map for 2019. It shows that electoral autocracies are especially prevalent in Africa and Asia, though they are also spread across Latin America, the Middle East, and Europe. I next display the prevalence of infrastructure spending by political regime. I theorize that clientelism and the public-private orientation of the corporate sector in political regimes offer greater explanatory power for the prevalence of Chinese foreign spending across countries than their infrastructure spending needs. Before proceeding to the political regime characteristics, I first show that countries’ domestic infrastructure spending does not display any meaningful pattern with regard to political regime. Clientelism is theorized to play an important role in driving Chinese foreign infrastructure spending. Specifically, electoral autocracies and electoral democracies are posited to rely more heavily on clientelism than 108

https://doi.org/10.1017/9781009385879.005 Published online by Cambridge University Press

Measuring Clientelism and the Corporate Sector

109

closed autocracies. To assess this claim, I examine the prevalence of clientelism according to three widely used indicators including corruption, rule of law, and bureaucracy quality. The specific measures come from the International Country Risk Guide and closely match the theoretical constructs of interest. The findings generally match the expectations, with electoral autocracies followed by electoral democracies displaying the highest levels of corruption and the lowest levels for the rule of law and bureaucracy quality. I use secondary school enrollment as a robustness test, and it also supports this pattern. I then examine the public-private orientation of the corporate sector across political regimes. Several different methods are used to generate confidence in the results and to provide additional details relevant to BRI projects. First, I examine the prevalence of state ownership of publicly listed companies across political regimes with descriptive statistics and regression analyses that control for income level and legal tradition with cross-sectional data. The results support the relationship between political regime and the prevalence of state ownership in the corporate sector. I supplement the regression analyses with a case study of the Philippines across the twentieth century which displays a sharp increase in SOEs during the Marcos dictatorship followed by a corresponding decline in SOEs with the restoration of democracy. Next, I provide the results of a very time-consuming data collection effort to identify SOE industry concentration by political regime using data for both publicly listed and private firms provided by Bureau van Dijk. Industries of greatest relevance to the BRI are selected and offer helpful guidance regarding the prevalence of state control depending on the type of infrastructure project and political regime of the host country. This is important for establishing the potential for state control and administration of a project in the host country, for illuminating how clientelist benefits may be distributed, and the potential for state-to-state deal-making and the accompanying characteristics associated with SOEs and infrastructure projects as enumerated in Table 4.2. To establish the public-private orientation of the corporate sector, it is also necessary to consider the protections for private capital. I therefore also examine the prevalence of expropriation risk and the protections for minority investors across political regimes. The results indicate that closed and electoral autocracies are very similar for both measures. This implies that the greater prevalence of the private sector in electoral autocracies is due more to the reduction of crowding out effects rather than stronger protections for private capital in these regimes. That political rulers in electoral autocracies retain strong political control over private investment offers helpful insights into the benefits of a foreign

https://doi.org/10.1017/9781009385879.005 Published online by Cambridge University Press

110

China’s Chance to Lead

state investor (i.e., Chinese SOE) initiating and leading projects in these regimes and paving the way for private investors to follow. Finally, I examine the prevalence of public-private partnerships (PPP) across political regimes. This involves illustrating the rise in total investment in PPPs across time, between 1990 and 2019, across political regimes. There are three notable findings. First, there is a substantial increase in the use of PPPs starting in 2005, and this primarily occurs among electoral democracies. Second, there is a dramatic increase in PPPs into electoral autocracies in 2015, just after the initiation of the BRI. Third, closed autocracies display a dramatic surge in the use of PPPs starting in 2017, which is primarily due to China. Overall, the trends suggest PPPs are of increasing importance to electoral autocracies and China which matches China’s statements regarding the desire to increase the involvement of the private sector in BRI projects, especially in relation to their financing. Finally, to assess the argument that political regimes are highly salient to attracting private participation in PPPs, I show that a political regime approach offers greater explanatory power for countries’ PPP legal frameworks than income level (according to the World Bank’s PPP Legal Framework database). Overall, this chapter provides robust evidence about the characteristics of political regimes posited to influence Chinese infrastructure spending as developed in Chapter 4. With these characteristics now firmly established, the subsequent chapters analyze and test whether political regimes display a robust relationship to Chinese infrastructure spending. 5.1

The Distribution of Political Regimes

As discussed in the previous chapter, the classification of political regimes includes closed autocracy, electoral autocracy, electoral democracy, and liberal democracy. This four-regime typology is anchored in a widely accepted theoretical categorization of political regimes (e.g., Diamond 2002; Roessler and Howard 2009; Schedler 2013). The coding rules for placing countries into specific regime categories are discussed in Lührmann et al. (2018). The political regimes dataset is comprehensive, classifying virtually all country-years since 1900 based on this typology using data compiled for the Varieties of Democracy (V-Dem) database. Figures 5.1 and 5.2 display the temporal and spatial distribution of these regimes, respectively. Figure 5.1 displays the prevalence of these regimes for low- and middle-income countries, specifically. Looking across time, we can observe that the prevalence of both closed and electoral autocracies declined from 1982 until around 2011. But during

https://doi.org/10.1017/9781009385879.005 Published online by Cambridge University Press

Measuring Clientelism and the Corporate Sector

111

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1

Closed Autocracies

Electoral Autocracies

Electoral Democracies

2018

2016

2014

2012

2010

2008

2006

2004

2002

2000

1998

1996

1994

1992

1990

1988

1986

1984

1982

1980

0

Liberal Democracies

Figure 5.1  Distribution of political regimes among low- and middleincome countries, 1980–2019

the subsequent decade, the period during which the BRI was introduced, autocratic regimes have stabilized and modestly increased their relative prevalence. Notably, electoral autocracies account for the largest share of all developing countries at the end of the sample in 2019 (approximately 50 percent), followed by electoral democracies (about 32 percent), closed autocracies (13 percent), and then liberal democracies (5 percent). Figure 5.2 shows the locations of these regimes across the world in 2019. It reveals there is a concentration of electoral autocracies in Africa and Asia. Closed autocracies are primarily located in North Africa, the Middle East, as well as East Asia near China. Electoral democracies are concentrated in Latin America, Central and Eastern Europe, and spread across Africa and Asia. Liberal democracies are primarily found among the advanced economies of North America, Western Europe, Northeast Asia, and Australia and New Zealand. Among developing countries, there are a handful of liberal democracies scattered across Latin America, Africa, and Asia. In Table 5.1, I also show descriptive statistics for infrastructure spending by political regime. The expectation is that this will not display a clear pattern. The results for the median values offer more reliable

https://doi.org/10.1017/9781009385879.005 Published online by Cambridge University Press

https://doi.org/10.1017/9781009385879.005 Published online by Cambridge University Press

Figure 5.2  World map of political regimes, 2019 Note: No data are available for Greenland.

Measuring Clientelism and the Corporate Sector

113

Table 5.1  Infrastructure spending by political regime in 2011 (% of GDP)

Closed Autocracy Electoral Autocracy Electoral Democracy Liberal Democracy

Mean

Median

Std. Dev.

Min

Max

Obs

3.99 2.15 2.35 2.15

1.76 1.93 1.97 2.39

4.46 1.24 1.66 2.23

1.29 0.25 0.00 0.39

11.86 6.49 8.40 7.94

5 50 44 9

results due to the small number of observations for closed autocracies (n=5) and liberal democracies (n=9). The median results indicate very similar levels of infrastructure spending as a percentage of GDP across political regimes. If a political regimes explanation can account for varying levels of Chinese infrastructure spending, then it is likely driven by factors other than the infrastructure spending gap. Indeed, I argue it is due to clientelism and the public-private orientation of the corporate sector across political regimes. I therefore turn to an examination of these regime characteristics next. 5.2

Political Regimes and Clientelism

The theory developed in Chapter 4 argues elections motivate leaders in electoral autocracies and electoral democracies to engage in a higher level of clientelism and to therefore seek a higher level of Chinese spending than leaders of closed autocracies. Electoral democracies are expected to display less clientelism than electoral autocracies as checks and balances are strengthened and the judiciary, which offers legal protections against corruption, becomes less vulnerable to political influence. To establish whether these expectations are manifested in reality, we can assess the extent of clientelism across political regimes. Commonly used indicators for the prevalence of clientelism include corruption, rule of law, and quality of the bureaucracy (Keefer 2007; Hicken 2011; Bold et al. 2018; Lo Bue et al. 2021). I use data from the International Country Risk Guide (ICRG) which focuses on the characteristics and measurement of variables that closely aligns with the prevalence of clientelism in a country. An additional benefit of the ICRG data is their comprehensive global coverage in 2014, the first full year of the BRI. According to Political Risk Service’s (2012) methodological report for the construction of the corruption variable, the corruption measure focuses on “actual or potential corruption in the form of excessive patronage, nepotism, job reservations, ‘favor-for-favors’, secret party

https://doi.org/10.1017/9781009385879.005 Published online by Cambridge University Press

114

China’s Chance to Lead Table 5.2  Clientelism proxies Mean

Median

Std. Dev.

Min

Max

Obs

Panel A. Corruption Closed Autocracy Electoral Autocracy Electoral Democracy Liberal Democracy

2.16 1.79 2.14 3.96

2.54 1.54 2.00 4.02

0.73 0.67 0.46 1.07

1.00 0.50 1.50 2.04

3.08 4.50 3.50 5.50

11 39 37 34

Panel B. Rule of Law Closed Autocracy Electoral Autocracy Electoral Democracy Liberal Democracy

4.02 2.89 2.96 4.84

4.5 3 5 5

1.37 0.94 1.00 1.07

0.5 1 1.5 2.5

5 5 5 6

11 39 37 34

2.00 1.50 2.00 3.50

0.61 0.77 0.90 0.79

0 0 0 2.00

2.00 4.00 3.00 4.00

11 39 37 34

69.13 33.38 18.20 64.06

120.33 4 106.59 19 107.80 32 162.30 29

Panel C. Bureaucracy Quality Closed Autocracy 1.73 Electoral Autocracy 1.49 Electoral Democracy 1.93 Liberal Democracy 3.28

Panel D. Gross Secondary School Enrollment Closed Autocracy 95.57 96.42 21.03 Electoral Autocracy 70.39 80.80 26.91 Electoral Democracy 77.51 87.31 26.79 Liberal Democracy 111.53 109.33 17.32

funding, and suspiciously close ties between politics and business” (PRS 2012).1 The score ranges from 0 to 6, where 6 denotes the lowest level of corruption. The Rule of Law indicator measures the “strength and impartiality of the legal system in combination with popular observance of the law” (PRS 2012). The measure ranges from 0 to 6. With regard to bureaucracy quality, higher points are given to countries in which the bureaucracy “tends to be somewhat autonomous from political pressure” (PRS 2012). The bureaucracy quality score ranges from 0 to 4. Table 5.2 reports the findings, which are largely consistent with the expectations. Specifically, both the mean and median scores for Corruption in Panel A are the lowest in electoral autocracies followed by electoral democracies and then closed autocracies. Other measures for corruption, such as the Worldwide Governance indicators dataset (Kaufmann et al. 2010), yield similar patterns. The Rule of Law indicator in Panel B mirrors these results for the mean scores. For Bureaucracy Quality in Panel C, electoral autocracy again displays the lowest mean 1

See www.prsgroup.com/wp-content/uploads/2012/11/icrgmethodology.pdf (accessed February 12 2022).

https://doi.org/10.1017/9781009385879.005 Published online by Cambridge University Press

Measuring Clientelism and the Corporate Sector

115

and median scores. Closed autocracy is tied with electoral democracy for the median score and slightly lower for the mean score. Overall, both the corruption and bureaucracy quality scores confirm the expectation that electoral autocracies are likely to engage in higher levels of clientelism than either closed autocracies or electoral democracies. In Panel D, I display the results for a robustness check using gross secondary school enrollment following Keefer (2007). The level of secondary school enrollment is a proxy for the provision of public goods. Thus, a lower level of secondary school enrollment indicates a lower government commitment to programmatic spending and a relatively greater allocation of spending via clientelism. The results are strongly aligned with the corruption findings. That is, electoral autocracy has the lowest level of secondary school enrollment for both mean and median values, followed by electoral democracy, then closed autocracy and finally liberal democracy. 5.3

Political Regimes and the Public-Private Orientation of the Corporate Sector

While the extent of clientelism is expected to contribute to higher demand for infrastructure spending from China by electoral ­autocracies and democracies in comparison to closed autocracies and liberal ­democracies, I expect the public-private orientation of the ­corporate sector to d ­ ­ifferentiate between electoral autocracies and electoral ­democracies. ­Specifically, greater state control of the corporate sector in electoral autocracies is predicted to contribute to a higher level of Chinese foreign spending in these regimes than in electoral democracies. This section examines the public-private orientation of the corporate ­sector across political regimes with data for the following categories: (1) the prevalence of state ­ownership of listed firms; (2) the SOE c­ oncentration of industries; (3) protections for private capital; and (4) institutional ­conditions to support PPPs. To assess the prevalence of state control of the corporate sector, I examine two sets of measures. The first is state ownership of publicly listed companies. The primary benefit of this measure is being able to assess the prevalence of state ownership across many countries. To evaluate the relation between state ownership and political regimes, it is necessary to have accurate information about shareholders of large companies across a large number of countries. Publicly listed firms, by virtue of being public so as to attract private investors, provide information about their major shareholders and their ownership stakes. A key weakness is that many state-owned companies are not publicly listed. Consequently, a measure

https://doi.org/10.1017/9781009385879.005 Published online by Cambridge University Press

116

China’s Chance to Lead

of state ownership of publicly listed companies is likely to underestimate their true significance in countries with underdeveloped, or nonexistent, stock markets. This motivates the need for an additional measure, the extent of industry concentration in the hands of SOEs. SOE industry concentration provides an indicator for the extent to which economic output is concentrated in the hands of a small number of state-owned enterprises. This measure offers useful supplementary information by accounting for both public and private firms. For example, an economy may have a relatively small number of SOEs to enhance the concentration of corporate control and reduce coordination problems (e.g., Brunei). Thus, the first measure could suggest that state-owned firms comprise a relatively small fraction of the country’s corporate sector. The second measure provides a helpful supplementary indicator by providing information about how concentrated an industry is in the hands of SOEs. The primary benefit of these indicators relative to other measures of state intervention in the economy, such as the Heritage Foundation’s Index of Economic Freedom, is that they offer a more precise measure for the state’s control of the corporate sector. Even though an economy may score high with regard to economic freedom, this does not necessarily mean the state lacks control. It could simply mean that economic freedom is compatible with the interests of the political rulers. Singapore, for example, has long scored near the top of the Economic Freedom Index; however, it also has a high level of state ownership. In strategically important industries, Singaporean SOEs retain near-monopoly control while the country also provides a welcoming environment for private investors that does not infringe on the turf of SOEs. This bifurcation of the Singaporean economy does not get accurately captured by economic freedom indices and fails to illustrate the link between Singapore’s ruling party and its corporate sector. It is therefore necessary to turn to more precise measures of state control and to supplement an indicator of state ownership of listed firms with an indicator for industry concentration. A comprehensive assessment of state ownership in relation to the private sector must also consider protections for private capital from expropriation either by the state or from majority shareholders (typically a family owner if not the state). Protections for private capital can offer helpful guidance on the willingness of private business to collaborate with or coinvest with SOEs, as in the context of public-private partnerships (PPPs). PPPs have emerged as an increasingly important option for developing economy governments to achieve their infrastructure development goals, as discussed in Chapter 2. Often, PPPs are used to alleviate the

https://doi.org/10.1017/9781009385879.005 Published online by Cambridge University Press

Measuring Clientelism and the Corporate Sector

117

debt burden of the government and to thereby enable the funding of infrastructure projects that would not otherwise be initiated. A comprehensive assessment of the public-private orientation of a country’s corporate sector with specific relevance to infrastructure development must therefore consider the strength of its PPP framework and its capacity to attract private investment. 5.3.1

State Ownership of Publicly Listed Companies

Bureau Van Dijk’s ORBIS database provides comprehensive coverage of ownership information about publicly traded firms globally.2 Data are available since the late 1990s up to the present, however, coverage improves considerably after 2003. There remain inconsistencies and errors (e.g., double entries), as well as missing information for many companies (see Kalemli-Ozcan et al. 2015). After extensive improvements and “cleaning” of the data, Aminadav and Papaionnou (2020) obtain a sample for 2012 of 26,843 firms in 85 countries, representing approximately 95.2 percent of global GDP, 89 percent of the total value of market equity,3 and 85 percent of global population. Identifying control across publicly traded firms is challenging as corporate law on managerial power, shareholder rights, and civil procedures differ around the world (La Porta et al. 1999, 2008). Additionally, corporate ownership structures are complex, with cross-holdings, pyramids, and shareholders often controlling firms via intermediate firms. Prior work has abstracted from these issues and applied a voting-rights cutoff to identify a controlled corporation. Following the literature, Aminadav and Papaionnou (2020) employ a 20 percent cutoff.4 Because of my interest in comparing state control to political regimes, I use data aggregated at the country level from this sample. 5.3.1.1 Patterns: State Control around the World In aggregate, governments controlled around 4.8 percent of listed firms in 2012; yet these firms amounted to 13.8 percent of the total market capitalization as the state typically controls large companies in strategic industries such as utilities and banks. These aggregate numbers also 2

Aminadav and Papaioannou (2020) provide a thorough discussion of the benefits and drawbacks of the ORBIS database. 3 This is based on the global sample available from Datastream. 4 La Porta et al. (1999), Claessens et al. (2000), and Carney and Child (2013) identify firms as controlled if a shareholder holds more than 20 percent. Lins et al. (2013) employ a 25 percent cutoff, while Laeven and Levine (2008) use a smaller 10 percent cutoff.

https://doi.org/10.1017/9781009385879.005 Published online by Cambridge University Press

118

China’s Chance to Lead

Table 5.3  Publicly listed state-owned enterprises by regime type All Economies

Closed Autocracies Electoral Autocracies Electoral Democracies Liberal Democracies Total

Emerging Economies

N

Proportion of Total

Proportion of listed firms that are state owned N

Proportion of Total

Proportion of listed firms that are state owned

10

11.7

18.3

9

10.5

19.9

14

16.4

14.7

13

15.3

15.4

22

25.8

7.4

21

24.7

7.5

39

45.8

4.7

6

7

85

6

49

Note: firms are classified as state-owned if the government is the dominant shareholder with a 20 percent voting-rights threshold (Aminadav and Papaiounnou 2020).

reflect the overrepresentation of firms in liberal democracies where a large share of the world’s listed companies is located. When examining the extent of government control by regime type, as shown in Table 5.3, there is considerable heterogeneity. The results indicate clear differences in the prevalence of state ownership between autocracies and democracies. Among all economies and among the subset of emerging economies, the distribution of listed SOEs displays a linear relation with regime type, though they tend to be more prevalent in emerging economies. Can these patterns be accounted for by countries’ legal traditions or levels of economic development? Table 5.4 assesses these possibilities with regression tests that follow the specifications of Aminadov and Papaionnou (2020). Model 1 displays results with common law as a control variable. The results indicate that closed autocracies display a significantly larger share of state-owned publicly listed firms than any other regime type; electoral autocracies are close behind. Notably, the results for electoral democracy do not display a significant difference with respect to liberal democracies (the omitted comparison category). Model 2 includes a control for the natural log of GDP per capita to account for the level of economic development; additionally, regional fixed effects are included. The results are marginally weaker, but still significant and consistent with model 1: state ownership is more prevalent among listed firms in closed autocracies followed by electoral autocracies; state ownership of listed firms in electoral democracies is not significantly different from liberal democracies.

https://doi.org/10.1017/9781009385879.005 Published online by Cambridge University Press

Measuring Clientelism and the Corporate Sector

119

Table 5.4  Listed SOEs and regime type regressions

Closed Autocracy Electoral Autocracy Electoral Democracy Common Law

State

State

(1)

(2)

13.742*** (2.86) 11.077** (2.64) 2.355 (1.43) −1.134 (−0.47)

14.018** (2.51) 10.678* (1.93) 1.620 (0.41) 0.780 (0.19) −0.189 (−0.07) 4.566 (0.18) Y 82 0.143

Ln(GDP per capita) constant Regional Fixed effects N adj. R2

5.051*** (3.88) N 83 0.201

Note: Regions are identified according to the World Bank’s classification, including East Asia and Pacific, Europe and Central Asia, Latin America and the Caribbean, Middle East and North Africa, North America, South Asia, subSaharan Africa.5 *, **, and *** indicate significance at the 0.10, 0.05, and 0.01 level.

The data do not allow for conclusive evidence in favor of political determinants relative to economic development, but it is highly suggestive of the relative importance of political determinants to autocratic regimes in comparison to democracies. Panel data on state ownership that includes autocratic regimes are necessary but difficult to compile annually and across countries. I therefore conduct a focused analysis of temporal change in the context of The Philippines to identify whether a change in political regime leads to a change in the prevalence of SOEs. The Philippines is well-suited to study this relationship because its political regime changed from electoral democracy to closed autocracy and back to electoral democracy over the course of the twentieth century. Meanwhile, its economic development slowly increased while its legal tradition remained unchanged. Additionally, it did not adopt a centrally 5

See: https://datahelpdesk.worldbank.org/knowledgebase/articles/906519-world-bank-countryand-lending-groups (accessed January 2021).

https://doi.org/10.1017/9781009385879.005 Published online by Cambridge University Press

120

China’s Chance to Lead

planned economy as in communist countries, making it suitable to compare with contemporary dynamics and my argument. 5.3.1.2 Temporal Case Analysis: The Philippines In the contemporary Philippines, an electoral democracy, powerful local families form the building blocks of electoral competition and politics. Such familial coalitions have “a unique capacity to create an informal political team that assigns specialized roles to its members, thereby maximizing condition and influence” (McCoy 2009). Kinship networks act to ensure that elected representatives deliver patronage and clientelist goods from the government, enhancing the consolidation of their wealth and influence. Congress acts as the nexus for these national-local clientelist exchanges. Because representatives are tied to local family networks rather than a national political party, individual representatives commonly engage in party-switching in order to maximize the delivery of resources to their local support coalition, contributing to the formation of short-lived dominant parties. Public opinion surveys have consistently shown that citizens do not vote for representatives on the basis of party reputation; for example, a survey conducted by Pulse Asia in March 2010 found that 91 percent of respondents did not identify with any political party. Teehankee also reports that since 1987, an average of 33.5 percent of all lower house representatives elected to Congress switched parties (Teehankee 2012), with 60.2 percent of these party switchers jumping into the party of the sitting president in order to magnify their access to national resources. The political clan also makes it possible to exercise influence beyond the term limits of a single politician through inter-generational clientelist bonds (Muno 2010). The construction of “political dynasties” has occurred by members of the same clan occupying numerous local positions who continuously succeed each other in these positions. To maximize access to government resources, families will also seek to capture the most potent combination of local political offices, such as holding the congressional district seat together with the gubernatorial seat or a big city mayoralty seat (Dios 2007). As of 2010, for example, the Philippine Congress had the highest percentage of elected dynastic legislators in the world at 68 percent, followed by Mexico at 40 percent, Japan at 33 percent, and Argentina at 10 percent. The US Congress had only 6 percent elected dynasts (Mendoza et al. 2012). Around 160 of these political clans have had two or more members who have served in Congress, and they account for more than 400 of the 2,407 men and women who have been elected to the Philippines national legislature between 1907 and 2010 (Teehankee 2012).

https://doi.org/10.1017/9781009385879.005 Published online by Cambridge University Press

Measuring Clientelism and the Corporate Sector

121

Many therefore regard the legislature as being controlled by elite families with considerable influence over almost all aspects of state functions (McCoy 2009; Kondo 2014). The public service, for example, is widely known to provide opportunities for plunder to families in power since the Spanish period (World Bank 2001), and corruption is regarded as the most problematic factor for doing business in the country (the Philippines was ranked 141 out of 144 countries for the ease of starting a business according to the Global Competitiveness Report; Schwab and Sala-i-Martin 2014). The capture of national government by local interests is also cited as a major reason for the failure of land reform (Abinales and Amoroso 2005; You 2015). The Philippines therefore exhibits the characteristics of an electoral democracy in which national resources are captured by local political clans whose power commonly derives from the ownership of large, diversified conglomerates. These arrangements trace their origins to 1907, when the first congressional elections were held. However, they were interrupted during the period of the Marcos dictatorship (1972–1986). When Marcos declared martial law in 1972, he abolished the legislature, weakened potential political rivals, centralized the bureaucracy, and brought local government under the control of the state. He further attempted to “[replace] established families with a coterie of his own” (McCoy 2009). The boundaries between the state treasury and the private purse of the president and his cronies became blurred, leading the Philippines to look remarkably similar to its neighbor, Brunei. Scholars have referred to the Philippines during the Marcos era as a “Sultanistic” regime.6 Table 5.5 displays the rise and decline of SOEs in the Philippines from 1935 to 2000. Corresponding to the dominance of local families on national politics, SOEs remained relatively few in number until Marcos introduced martial law in 1972. The expansion of SOEs during Marcos’s reign occurred for two primary reasons. First, Marcos’s cronies expanded their businesses with preferential loans guaranteed by the state. When their firms incurred financial losses or went bankrupt, private debts became the state’s obligation and the state had to bail them out, which involved taking ownership of them. Second, in order to weaken the power of elite rivals, Marcos expropriated numerous private assets, transforming them into state-owned corporations. As a result, the government owned almost 29 6

Thompson (1995, 51), for example, argues that the imposition of martial law under Marcos enabled him to transform the Philippines from a political system based on clientelism to a form of sultanism, under which “a sultanistic dictator exercises power not for a particular class but for the benefit of family and friends.”

https://doi.org/10.1017/9781009385879.005 Published online by Cambridge University Press

122

China’s Chance to Lead Table 5.5  Rise and decline of SOEs, the Philippines (1935–2000) Year

Parents

Subsidiaries

Total

1935 1945 1950 1955 1960 1965 1970 1975 1981 1986 1990 1995 1999 2000

12 18 21 27 30 44 47 71 92 105 71 74 77 84

2 3 8 8 10 18 18 49 120 162 101 104 90 79

14 21 29 35 40 62 65 120 212 327 172 178 167 163

Data Source: CPED (2002, 6). Original data come from Presidential Commission on Good Governance (19135–1986) and the Commission on Audit Corporate (1990–2000).

percent of the total assets of the largest 1,000 corporations in the Philippines in 1978 (Tabbada 1985, 434). When Marcos left office in 1986, the total number of SOEs reached 327. During the Marcos period, the fiscal burden of SOEs became excessive. When the economic situation deteriorated after the second oil crisis in 1979, Marcos requested international assistance. This was followed by a domestic financial crisis in 1981 which contributed to the bankruptcy of many large firms. Two years later, the economic situation further deteriorated, leading to assistance from the World Bank and the IMF. As part of the aid package that was offered, rationalization of the state sector was required (Haggard 1988). Consequently, Marcos agreed to initiate privatizing reforms, although these plans were not implemented due to the regime’s collapse in 1986. After the fall of Marcos, the Aquino and Ramos administrations implemented privatization programs. By 2000, there were 163 SOEs, including 84 parent companies and 79 subsidiaries – a 50 percent reduction from the peak in 1986. Privatization occurred primarily via bidding or negotiation, enabling wealthy private conglomerates to gain ownership. Compared to public offerings, bidding is less likely to broaden public ownership, enabling large business groups to gain control of lucrative

https://doi.org/10.1017/9781009385879.005 Published online by Cambridge University Press

Measuring Clientelism and the Corporate Sector

123

state enterprises – often involving the transformation of public monopolies into private monopolies (Briones 1989, 36; Patalinghug 1996, 16–19). The lack of transparency and inadequate oversight involved in the privatization process enabled many leading business groups to acquire some of the largest and most valuable SOEs, including Philippine Airlines, major utilities, and banks.7 In the wake of the AFC, few reforms were initiated to improve corporate governance standards, and the state did not take a leading role in the recovery. By contrast, it was the family-owned conglomerates that tended to purchase state-owned firms. Companies in key business sectors often owned by the state in other economies due to having a natural monopoly, such as those in transportation, energy, telecommunications, and financial services, are controlled by a limited number of corporations attached to family-owned conglomerates (World Bank 2008). Although there were numerous SOEs according to official figures, they comprised a very small fraction of the country’s largest listed firms.8 The country’s political dynamics have impacted government initiatives for public goods such as infrastructure projects. Prior to a 2002 Government Procurement Reform Act, there was a maze of laws at both the federal and local levels, supplemented by an array of executive and administrative orders detailing various rules on public procurement procedures. This complexity made the process vulnerable to manipulation, enabling kickbacks to rise up to 30 percent of the value of a contract and allowing for the construction of substandard projects (Furmas 2013). The Reform Act9 was passed in order to address what a World Bank Country Report characterized as a “dysfunctional” public procurement system (Furmas 2013). The Reform Act codified the general principle of transparency in two specific ways. First, the law created an e-procurement portal, PhilGEPS, which would serve as the definitive source of information on government procurement. All public tenders at both the national and local levels must pass through the PhilGEPS system. Second, at all stages of the 7

See Wang (2005) for a list. For example, Philippines Airlines was sold to the Lucio Tan Group; the Manila Electric Company (Meralco) was sold to the Lopez Group; the Metropolitan Waterworks and Sewage System (MWSS), West Zone was sold to the Lopez Group; the Metropolitan Waterworks and Sewage System (MWSS), East Zone was sold to the Ayala Corporation; the Union Bank of the Philippines was sold to the Aboitiz Group. 8 See the Carney and Child (2013) dataset for details. This is also partly due to the relative health of the Philippines banking sector (in 1996 the ratio of bad loans to total loans was only 3 percent) (Financial Times, 8 October 1997). In contrast to elsewhere in the region, bank closures were not a part of the Philippine bailout plan offered by the IMF. 9 Republic Act No. 9184

https://doi.org/10.1017/9781009385879.005 Published online by Cambridge University Press

124

China’s Chance to Lead

procurement process, a representative of the Commission on Audit and at least two outside observers must be present – one from a private sector organization and another from a nongovernment organization. However, the promise of these principles has been unrealized in practice because compliance and implementation are lacking. A critical part of the problem is that outside oversight is nearly impossible because of restrictions placed on access to PhilGEPS data. Consequently, public procurement has remained riddled with corrupt practices. For example, according to a 2006/2007 survey of 705 business managers, 61 percent of respondents declared that there was “a lot” and 33 percent “some” corruption in the Philippine public sector.10 Nearly 50 percent of respondents also stated that “almost all companies” or “most companies” in their line of business paid bribes to secure a public contract. In addition, 32 percent of the respondents indicated the size of the bribe to be the equivalent of 21 percent or more of the contract sum, with a further 25 percent indicating it was between 11 percent and 20 percent (Social Weather Station 2007). According to Jones, Many of the bribes are paid to senators in the Philippine Congress who can influence contract awards, especially in large infrastructure projects. The money received is used by senators for their own personal benefit, or to fund their election campaigns illicitly. The sources of the payments are wealthy elite families in the business community who can ensure that the preponderance of public procurement contracts go to their businesses.11

In summary, the Philippines illustrate state sector changes that accompany regime changes between electoral democracy and closed autocracy. The Marcos period witnessed a dramatic rise in the size and power of the state sector which the political rulers used to centralize their control and weaken political opponents, including the expropriation of assets from family-owned business groups. Its resemblance to Brunei was so striking that it has even been referred to as a “sultanistic” regime. Following the regime’s collapse, local clans regained their influence, producing an electoral democracy characterized by the diversion of state resources to political leaders’ local support coalitions. Consequently, the state sector remains small; even when natural monopolies would normally call for state ownership, often they wind up in the hands of family-owned business groups. The incentives of political representatives to deliver patronage and engage in clientelist exchanges contribute to opacity governing public-private exchanges, such as public procurement and infrastructure programs. 10

The survey was conducted by the Philippine survey organization Social Weather Station (2007). 11 Jones (2009). Also see Global Advice Network (GAN), (2007) and Quimson (2006).

https://doi.org/10.1017/9781009385879.005 Published online by Cambridge University Press

Measuring Clientelism and the Corporate Sector

5.3.2

125

State-Owned Enterprises Market Concentration

In addition to thorough coverage of publicly listed firms globally, Bureau van Dijk’s Orbis database also provides comprehensive information about private firms globally. That said, private firm information tends not to be updated as regularly as publicly listed firms. In some countries, private firms are among the largest enterprises. But because data about them may be available only every few years, I construct industry concentration measures for the ten-year period, 2010–2019. Thus, the strength of the publicly listed data about state ownership is its reliable coverage across countries for a particular year. The key weakness is the lack of coverage of private firms. The SOE concentration sample compensates for this by providing information about private firms, but this information must be compiled across multiple years in order to make cross-country comparisons. In this way, the two datasets complement each other and provide a more complete picture of state ownership across countries around the world. Altogether, the SOE concentration sample includes over 21 million firm-year observations from over 4,700,000 firms from 160 countries. Needless to say, this entailed a considerable data collection effort. To calculate SOE market concentration, I use the concentration ratio which is calculated by taking the operating revenue of the top three SOEs in an industry for a specific country in relation to the sum of the operating revenues of all firms in that industry. This is the same calculation as that used by researchers at the World Bank and IMF to calculate cross-country measures of bank concentration (except using firm revenue rather than bank assets).12 Industries selected for reporting include those important to the Belt and Road Initiative as well as manufacturing industries due to their importance to the economic structure of emerging economies. Firms are selected based on their affiliation to an industry according to their two-digit industry code. The industries include agriculture, forestry, fishing, and hunting, construction, finance and ˇ ihák, Feyen, See the Financial Development and Structure Dataset by Demirgüç-Kunt, C Beck, and Levine, available at: www.worldbank.org/en/publication/gfdr/data/financialstructure-database (Beck et al. 2000, 2009; Cˇihák et al. 2012). The main alternative measure of market concentration is the Herfindahl-Hirschman Index (HHI) which provides a more complete picture of industry concentration by using the market shares of all firms in an industry. While the HHI is straightforward to calculate, it depends on having accurate information about all firms in an industry which can make its construction problematic when information about private firms is unavailable, as in the context of developing countries. Thus, the concentration ratio offers a useful estimate of SOE concentration for drawing comparisons across countries without the need for information about all firms so long as one can calculate a reasonably accurate approximation of total market size (typically possible with information from very large, large, and medium-sized firms). 12

https://doi.org/10.1017/9781009385879.005 Published online by Cambridge University Press

126

China’s Chance to Lead

insurance, mining, quarrying, and oil and gas extraction, manufacturing: consumer goods, manufacturing: basic chemicals, manufacturing: metals and machinery, real estate, transportation, and utilities. Firms used to calculate the total size of an industry based on firms’ operating revenue are those classified by Bureau van Dijk as very large, large, and medium-sized companies.13 Due to incomplete annual information about many firms in emerging economies, it is necessary to use aggregate information provided across multiple years. For example, firm A may provide revenue information for 2016 only while firm B may provide revenue information for 2013 only. To enhance the accuracy and comparability of the concentration measure across industries, countries, and political regimes, I aggregate data across the 2010–2019 time period to maximize the number of observations available for calculating market concentration and select the largest value for a firm’s revenue during this time period. This comes with the drawback of not accounting for changes to market concentration that may occur within this time span. Due to data limitations for firms from many of the countries of interest for this study, as well as the general stability of state-owned firms in their respective markets, I opt for maximizing information used to construct the measure and to thereby maximize the accuracy of the market concentration measure over identifying change. A country’s political regime is assigned according to the prevailing political regime during the ten-year time period. Table 5.6 presents the results of this analysis. The results reported in the SOE Market Concentration table corroborate the relatively greater role played by SOEs in closed autocracies as reported in the publicly listed results (Table 5.3), which declines as one proceeds toward liberal democracies.14 Also note the greater role of SOEs in Closed Autocracies is observed both when China is included and excluded from this category. The results also document that the extent of SOE dominance across political regimes varies by industry. The two industries with the highest levels of SOE concentration include Mining, Quarrying, and Oil and Gas Extraction as well as Utilities. In the latter industry, the decline in industry concentration is relatively mild compared to the former industry. This 13 Given the large number of firms available from these categories, coupled with restrictions on the number of firms that can be downloaded at a time, the marginal benefit of including small firms is very low. 14 These results likely understate the state’s concentration in the context of closed autocracies in particular because many private firms are effectively owned by the ruling elite who draw on public funds to finance them, as in Brunei (see Carney 2018) and the UAE (see Chapter 8).

https://doi.org/10.1017/9781009385879.005 Published online by Cambridge University Press

https://doi.org/10.1017/9781009385879.005 Published online by Cambridge University Press

Firm-year observations Number of firms Countries SOE Conc. All Industries SOE Conc. Construction SOE Conc. Finance and Insurance SOE Conc. Mining, Quarrying, and Oil and Gas Extraction SOE Conc. Real Estate SOE Conc. Transportation SOE Conc. Utilities SOE Conc. Information SOE Conc. Agric, Forestry, Fishing and Hunting SOE Conc. Manufacturing: Consumer Goods SOE Conc. Manufacturing: Basic Chemicals SOE Conc. Manufacturing: Metals and Machinery

Table 5.6  SOE industry concentration, 2010–2019

1,025,096 337,570 21 0.25 0.06 0.26 0.56 0.08 0.31 0.34 0.37 0.04 0.05 0.38 0.07

Closed Autocracies 575,138 119,962 20 0.26 0.06 0.28 0.62 0.09 0.35 0.37 0.43 0.08 0.05 0.43 0.09

Closed Autocracies (China Excluded) 1,023,233 181,226 58 0.19 0.08 0.28 0.37 0.09 0.25 0.40 0.15 0.03 0.02 0.12 0.06

Electoral Autocracies

3,269,556 733,722 47 0.11 0.06 0.16 0.34 0.02 0.07 0.28 0.13 0.04 0.01 0.15 0.02

Electoral Democracies

16,512,708 3,522,538 34 0.06 0.01 0.09 0.21 0.02 0.07 0.20 0.08 0.02 0.01 0.04 0.01

Liberal Democracies

128

China’s Chance to Lead

may suggest the need for SOEs in utilities due to the ­common occurrence of a natural monopoly as well as the potential incapacity of private firms to access the volume and long-term commitment of capital necessary to participate in this industry, regardless of the political regime (though the results suggest the regime retains some influence). In the Mining industry, the argument for a natural monopoly or lack of capital to private players is weaker, which may account for the more dramatic decline in SOE concentration between electoral democracies and liberal democracies. Industries with the lowest levels of SOE concentration include Construction, Real Estate, Agriculture, Forestry, Fishing and Hunting, Manufacturing of Consumer Goods, and Manufacturing of Metals and Machinery. Transportation and Utilities are the most common industries in which projects are initiated in the context of the Belt and Road Initiative, to be discussed in Chapter 6. It is therefore useful to observe that the level of SOE concentration in Transportation is only modestly lower than that in Utilities among closed and electoral autocracies; there is a steep decline in SOE concentration for the Transportation industry when going from electoral autocracies (0.25) to electoral democracies (0.07). The main takeaway from these market concentration results, when considered together with the listed firms’ results, is the clearly lower level of SOE market concentration in electoral autocracies in comparison to closed autocracies, and the relatively greater prevalence of large private firms in electoral democracies in comparison to electoral autocracies. These patterns are consistent with more detailed country studies, such as those discussed in Carney (2018) and Naughton and Tsai (2015). The upper echelons of China’s corporate sector, a closed autocracy, are populated by numerous mammoth SOEs; private firms tend to be relatively small. In Malaysia and Singapore, electoral autocracies, industries tend to have fewer mammoth SOEs dominating the corporate sector, though there are many large SOEs across industries. In Indonesia and the Philippines, electoral democracies, large family-owned firms, and affiliated business groups have displaced many once powerful SOEs, as mentioned in the Philippines case. 5.3.3

Protections for Private Capital

There are two alternative explanations for the varying presence of private capital across regime types. The first regards crowding out effects. Through their dominance, SOEs can crowd out opportunities for private capital, thereby reducing their presence in particular industries, as shown in the prior section. The second explanation is weak protections for private investment.

https://doi.org/10.1017/9781009385879.005 Published online by Cambridge University Press

Measuring Clientelism and the Corporate Sector

129

Table 5.7  Protections for private capital by regime type Expropriation Protections for GDP per Risk Minority Investors capita Closed Autocracy Electoral Autocracy Electoral Democracy Liberal Democracy

61 60 68 85

47 45 54 66

15279 4497 8459 39477

Note: Data for Expropriation Risk are on a 0 to 1 scale; I multiply them by 100 to place them on a 0 to 100 scale. Data for each variable are for 2019. GDP Per capita is for 2010 USD from the World Bank.

To assess which explanation accounts for differences in the presence of private capital across political regimes, and between closed and electoral autocracies in particular, I examine the strength of property rights and investor protections.15 I use two alternate variables to measure the degree to which private property is protected. The first is based on a country’s expropriation risk. Expropriation risk is extracted from the International Country Risk Guide (ICRG). The variable used is “ICRG Investment Profile,” which is described as the assessment of risk of investment due to contract viability/expropriation and profits repatriation. The index ranges from 0 (high expropriation risk) to 1 (low expropriation risk). The data are for 2019. Second, to measure investor protections, I use the protecting minority investors indicator from the World Bank Doing Business Index. The measure is also for 2019. The results are displayed in Table 5.7. The results resemble those from the publicly listed SOEs data insofar as there is a clear difference between autocracies and democracies. The data also display a sizeable difference between electoral democracies and liberal democracies, which is consistent with differences in institutional quality between emerging and advanced economies. But given the negligible difference between closed autocracies and electoral autocracies, the evidence suggests the main reason for private capital to vary in its willingness to invest across autocratic regimes is not due so much to varying protections for private capital, but instead due to variation in crowding out effects. With regard to electoral autocracies and electoral democracies, private capital’s willingness to invest is due to a combination of crowding out effects and protections for private investors. The results suggest this joint effect also applies to differences between electoral and liberal democracies. 15

This follows the same approach as that used by other scholars such as Durven, Ernunza, and Molchanov (2009).

https://doi.org/10.1017/9781009385879.005 Published online by Cambridge University Press

130

China’s Chance to Lead

5.3.4

Public-Private Partnerships

Public-private partnerships offer great promise for addressing the infrastructure needs of emerging economies. They are commonly used for the construction and maintenance of highways, electricity, water supply, and many other dimensions and typically last twenty-five years or more. In PPPs, the public and private sector each leverages their own strengths. The public sector possesses the legal authority to promote infrastructure projects while the private sector commonly has money, technical expertise, and greater efficiency. Through collaboration the public sector mitigates political and regulatory risks such as currency inconvertibility and transfer restrictions, expropriation, breach of contract, political violence, as well as legal, regulatory, and bureaucratic risks. The private sector manages construction, supply-chain, and operational risks. The precise division of risk and responsibilities varies across countries and projects. For example, advanced economies may desire private participation more for their technical expertise or greater efficiency; emerging economies may focus more on the financing benefits that private partners can bring. The ultimate goal is for each side to draw on their particular strengths. Consider the construction of a subway. Large cities benefit from rapid railway systems. They are faster than alternative forms of transport, take up less land, are energy efficient, have low carbon emissions, and help promote economic growth. But they also require a huge initial capital investment and the ability to pay for ongoing maintenance costs. However, the annual return of investing in a railway system is only about 1.5 percent which is too low to attract investors. The key to financing these kinds of projects is real estate. The price of an apartment above a subway station is often significantly higher than other locations. This enables a subway developer to include property development into the returns available to private investors. In this way, a rail plus property model adds value to the property and benefits the society and economy. For example, in Hong Kong the government grants the state-owned enterprise MTR Corporation with concessions that allow the company with the operation right of the new rail line as well as the land development right of the site. When building a new rail line, MTR partners with private developers to complete new residential and commercial properties atop the MTR station site. By capturing part of the value of the land property around railway lines, MTR generates funds for new projects as well as for operations and maintenance. This model has proven so successful that MTR has expanded its rail operations to Shenzhen, Beijing, New Zealand, Australia, and the UK.

https://doi.org/10.1017/9781009385879.005 Published online by Cambridge University Press

Measuring Clientelism and the Corporate Sector

131

180,000

Millions of USD

160,000 140,000 120,000 100,000 80,000 60,000 40,000 20,000

Electoral Autocracy

Electoral Democracy

Liberal Democracy

18

16

20

14

20

12

Closed Autocracy

20

10

20

08

20

06

20

20

04

02

20

00

20

98

20

96

19

94

19

92

19

19

19

90

0

Figure 5.3  Total investment in PPPs by regime type, 1990–2019 Note: The Private Participation in Infrastructure (PPI) Project database records contractual arrangements for public infrastructure projects in low- and middle-income countries (as classified by the World Bank) that have reached financial closure, in which private parties assume operating risks. The database has data on over 6,400 infrastructure projects in 137 low- and middle-income countries, but only projects classified as PPP are used to construct the figure. PPPs account for approximately 85 percent of total projects. Data Source: World Bank: https://ppi.worldbank.org/en/ppi

PPPs have emerged as an effective way for emerging economy governments to collaborate with private partners to achieve their infrastructure development goals. Often, however, the “private” partner is not actually a private company. It may be a state-owned enterprise, typically of a foreign country. But the PPP structure applies regardless of the identity of the “private” partner. It is worth emphasizing that PPPs are often used in the context of emerging economies to alleviate the debt burden of the government, and to thereby enable the initiation of a larger number of infrastructure projects. According to the World Bank, private sector investments in PPP projects (including foreign SOEs as the “private” partner) have increased from under USD5 billion in 1990 to over USD90 billion in 2019, as displayed in Figure 5.3. One can observe that following the start of the BRI in late 2013, there is a spike in PPPs among electoral autocracies, which declines in 2016 but rises again from 2017 to 2019. There is also a dramatic increase in PPPs in closed autocracies from 2017 to 2019 which is primarily attributable to China. Of the 409 projects recorded in 2019, about 40 percent (160 projects) had a

https://doi.org/10.1017/9781009385879.005 Published online by Cambridge University Press

132

China’s Chance to Lead

majority of their stakes sponsored by international e­ ntities, which includes both private firms and foreign SOEs.16 ­International ­sponsors primarily focused on energy, with 115 projects out of 160 in the energy sector. Private investment commitments occurred in 62 ­countries in 2014, which is the highest number in the last decade, illustrating the increasingly widespread use of PPPs among low- and middle-income countries. The rise in PPPs by electoral autocracies since the initiation of the BRI is consistent with the argument made in Chapter 4 (and the evidence about the substitution effect from international financing for infrastructure development in Chapter 3) that political rulers of these semicompetitive regimes are more likely to use PPPs as a mechanism for leveraging their own limited financial resources to attract more foreign funding. PPPs in which the “private” partner is actually a foreign public entity (i.e., state-owned enterprise) may present an especially attractive opportunity to leaders of electoral autocracies to expand their clientelist resources and state-administered benefits since there is a stronger match with regard to the capacity for opacity and the willingness to assign greater weight to political priorities relative to profit maximization. More detailed analysis of this issue is conducted in Chapter 7. Although PPPs have become increasingly common, a major challenge for emerging economies is attracting private sector participation (excluding foreign SOEs as a ‘private’ entity). Based on a review of PPP studies between 2000 and 2015, Osei-Kyei and Chan (2017) identify sixteen factors considered important to attracting private investments to PPPs in emerging economies, as detailed in Table 5.8. To determine which factors are of greatest importance, the authors conducted a global survey of dozens of PPP academic and practitioner experts. The top four factors identified include (in descending order) political support and acceptability for PPPs, government positive attitude toward private sector investments, political stability, and a favorable existing legal framework and policy. Political support and acceptability for PPPs by a country’s political incumbents will have a substantive impact on attracting investors and reducing political risk. The credible threat of political opponents gaining power and withdrawing support creates insecurity for private sector investments. This tends to be more of a concern among electoral democracies where allegations of corruption and lack of transparency may cause the project to lose support if political opponents rise to power 16

Sponsors are private entities that have an equity participation of at least 10 percent in the project contract for greenfields, brownfields, and management and lease contracts, and 15 percent of ownership in the project company for divestitures. A foreign state-owned enterprise is considered a private entity, although a domestic state-owned enterprise is not.

https://doi.org/10.1017/9781009385879.005 Published online by Cambridge University Press

Measuring Clientelism and the Corporate Sector

133

Table 5.8  Factors attracting private investments in PPPs in emerging economies Factors

Sources

Adequate financial capacity of contracting authority Adequate public sector experience in PPP Available and mature financial market

Adhadzi and Bowles (2004) World Bank (2009) Ozdoganm and Birgonul (2000); Li et al. (2005) Ahadzi and Bowles (2004); Shendy et al. (2011) Li et al. (2005) World Bank (2009) World Bank (2009) Hwang et al. (2013)

Competent PPP unit Favorable existing economic policies Favorable existing legal framework and policy Financial viability of projects Government positive attitude toward private sector investments Government providing guarantees Government providing tax rebate on imported equipment Investors’ knowledge of host country environment Political stability Political support and acceptability for PPPs Positive public ideology toward private sector ­participation in public infrastructure delivery Stable macro-economic indicators Well organized and committed contracting authority

Ng et al. (2012) Osei-Kyei et al. (2014) Zhang (2005) Zhang (2005) World Bank (2009) Jefferies et al. (2002), Zhang (2005) Osei-Kyei and Chan (2015) Hwang et al. (2013)

Source: adapted from Osei-Kyei and Chan (2017)

(e.g., India as compared to China or Singapore). These dynamics will be examined in Chapter 7. The second ranked factor is government positive attitude toward private sector investments. This is a particular concern for those industries dominated by SOEs. As discussed in the prior section, autocratic regimes typically lack strong protections for private capital from state expropriation and for minority investors. In China, for example, weak protections for private business have kept their participation in PPPs at a persistently low level (Tan and Zhao 2019). Political stability is ranked third. Political stability denotes the “absence of political violence, governmental longevity and existence of a legitimate constitutional regime” (Hurwitz, 1973). Political instability is most commonly observed when a country fails to conduct an orderly and peaceful transfer of political power from one political party to another. Private investors are wary of countries with a record of unstable political regimes. For example, unstable political conditions in Thailand contributed to the failure of the Bangkok Elevated Transport System (Tam 1999). This factor will be examined in detail in the context of electoral autocracies in Chapters 6 and 7.

https://doi.org/10.1017/9781009385879.005 Published online by Cambridge University Press

134

China’s Chance to Lead

Fourth, a favorable existing legal framework and policy is required. A well-defined legal structure for PPP arrangements makes investors feel secure and protected from both public and private sector expropriation (Pongsiri 2002). A well-structured legal framework includes, among other things, a detailed tendering process, fair, transparent, and competitive contract awarding and construction management procedures, and clear dispute resolution mechanisms (IMF 2010). However, in autocracies and electoral democracies, the judiciary typically lacks strong independence from political interference and exerts weak checks on executive power. To assess whether the strength of the legal framework of PPPs is related to the political regime, I examine a World Bank dataset on Procuring Infrastructure PPPs that measures the characteristics for 135 economies with a comprehensive set of survey questions. The survey is organized into four thematic areas: preparation, procurement, contract management, and unsolicited proposals (USP). The Preparation of PPPs covers activities that inform the decision of whether to launch a PPP procurement process, including whether the identification of a prospective PPP is consistent with other public investment priorities. It also accounts for activities undertaken prior to publishing the public tender notice, such as acquiring the necessary land and permits, and preparing the draft contract and tender documents. The Procurement of PPPs concentrates on whether distinct legal and regulatory arrangements adhere to recognized good practices in terms of choosing the private partner. These include fairness, neutrality, transparency of the process and a competitive bidding process. PPP Contract Management assesses the framework in place to implement PPP projects, including the existing monitoring and evaluation systems. This category also considers related regulatory provisions such as contract modification and renegotiation and dispute resolution. Finally, Unsolicited Proposals (USPs) for PPPs first identifies whether the submission of USPs is permitted by a country’s legal and regulatory framework. If allowed, it then assesses whether a specific process is in place to evaluate the USP, its alignment with other government priorities, compensation design procedures, as well as whether private partners are selected via a competitive process. Table 5.9 presents the average value for each of these four PPP dimensions as well as the average score by regime type (Panel A) and income level (Panel B). Data are for the 2020 dataset, which offers larger country coverage than the 2018 dataset and numerous methodological improvements.17 17

See the Benchmarking 2020 Infrastructure Development Report, available here: https:// bpp.worldbank.org/ (accessed February 5 2021).

https://doi.org/10.1017/9781009385879.005 Published online by Cambridge University Press

Measuring Clientelism and the Corporate Sector

135

Table 5.9  PPP legal framework dimensions by regime type and income level

Panel A. Regime Type Closed Autocracy Electoral Autocracy Electoral Democracy Liberal Democracy Panel B. Income Level Low income Lower middle income Upper middle income High Income

Preparation

Contract GDP per Procurement Management USP Average capita18

35 38

55 56

55 63

56 51

48 52

15279 4497

51

67

69

62

63

8459

50

71

63

66

61

39477

36 40

54 56

61 63

55 51

51 52

44

62

64

60

57

50

73

64

63

62

Note: The data for each variable are for 2019. GDP per capita data are for constant 2010 US$ from the World Bank.

The table illustrates that PPP legal system preparedness, on average, improves when moving from closed autocracy to electoral autocracy and especially from electoral autocracy to electoral democracy. Notably, regime type offers marginally greater explanatory power than income level, the category typically used by the World Bank to classify countries. The problem with grouping countries by income level is that many closed autocracies fall into the high-income category. However, their legal and regulatory arrangements do not resemble those of other high-income economies, which are typically liberal democracies. It is more appropriate to group countries by regime type. The improvement can be seen by the larger range of values spanning the average score category, extending from 48 to 63 (15 points), compared with 51 to 62 points (11 points) for the income-level grouping as in the World Bank’s (2020) report. The smaller range (11 points) is due to moving closed autocracies out of the low-income category into the high-income category, which both raises the low-income PPP score and lowers the high-income PPP score. 18

The World Bank report on PPPs uses GNI per capita, which shows the same pattern as GDP per capita.

https://doi.org/10.1017/9781009385879.005 Published online by Cambridge University Press

136

China’s Chance to Lead

The additional reason to consider political institutions is the lack of independence of legal systems from political interference in autocracies. Given that this is a score regarding PPP legal systems, in which the state is a key participant, politics will be relatively more important than in purely private commercial endeavors. Consequently, it is necessary to account for patterns associated with political regimes. Indeed, in the survey of PPP experts discussed above, the number one factor driving PPP success is government support (Osei-Kyei and Chan 2017). Legal systems function effectively to the extent the political system allows them to operate independently of political influence; but ultimately the power they possess is delegated by political rulers who retain the power to rescind that delegated authority. 5.4 Conclusions In Chapter 4, I develop a two-part argument regarding the impact of political regimes on Chinese infrastructure spending. First, regular elections held in electoral autocracies and electoral democracies contribute to greater demand for clientelist resources than in closed autocracies. Second, the administration of these clientelist benefits will be more state-controlled in electoral autocracies than in electoral democracies. Consequently, China will share a larger number of shared interests for BRI infrastructure spending with electoral autocracies relative to other political regimes, as enumerated in Table 4.2. This chapter establishes the empirical facts regarding political regimes and the prevalence of clientelism and the public-private orientation of the corporate sector. First, I established that clientelism is higher in electoral autocracies followed by electoral democracies and closed autocracies. Three of the most commonly used indicators for clientelism support this contention, including corruption, bureaucracy quality, and the rule of law. Secondary school enrollment offers a robustness test that further supports this relationship. Second, the relative balance of public (state) versus private ownership of the corporate sector varies by political regimes, even after controlling for income level and legal family. As the proportion of autocracies increases, it is important to understand their influence on state control of the corporate sector if we are to explain the prevalence of Chinese infrastructure spending across countries. In addition to establishing the empirics regarding these key characteristics of political regimes, there are three additional key takeaways. First, the evidence presented in this chapter clearly establishes that electoral autocracies are more conducive to private ownership than closed autocracies. Nevertheless, the state retains residual control rights as indicated

https://doi.org/10.1017/9781009385879.005 Published online by Cambridge University Press

Measuring Clientelism and the Corporate Sector

137

by the equivalent levels of risk of expropriation by the two types of autocratic regimes. The key difference is the relatively lower crowding out effects in electoral autocracies. This observation is consistent with a clientelist explanation of electoral autocracies which argues that political rulers of these regimes must expand their patronage network via a more institutionalized party apparatus to defend against political opponents during elections. A larger private sector generates more opportunities (carrots) for elites to share in the spoils of the regime (via public procurement contracts, favorable business regulations, lending from state-owned banks, etc.) and expands the support network on which incumbent rulers depend during elections while retaining the stick of expropriation. Second, electoral autocracies are especially prevalent in Africa and Asia. This is important as these locations overlap with the locations where the largest volume of BRI spending is concentrated, and their geographical concentration facilitates the ease with which China can promote coordination among these economies and governments and reap gains from network effects. The third key takeaway concerns the rise of public-private partnerships. The evidence indicates these are growing rapidly among closed and electoral autocracies since the introduction of the BRI in late 2013. The evidence raises the question of whether Chinese SOEs are playing an increasingly important role, by acting as the “private” partner, that is contributing to the rise of PPPs in electoral autocracies. The evidence also raises the question concerning the relative importance of the private sector in the rise of these PPPs. In summary, the evidence presented in this chapter empirically confirms the characteristics of political regimes regarding clientelism and the public-private orientation of the corporate sector that are posited to influence the prevalence of Chinese infrastructure spending across countries. Having empirically confirmed these political regime characteristics, I now turn to the analysis of Chinese foreign spending across political regimes in the context of the Belt and Road Initiative.

https://doi.org/10.1017/9781009385879.005 Published online by Cambridge University Press

6

Political Regimes and BRI Country-Level Patterns

The previous chapter empirically established the prevalence of political regimes both temporally and spatially. Crucially, the findings illustrated that electoral autocracies are the most prevalent type of political regime among emerging economies, accounting for about half of all countries. Moreover, the decline in autocracy that started in the early 1980s was reversed around 2011. Over the past decade, autocracies have witnessed a resurgence. Chapter 5 also examined evidence for the prevalence of clientelism across political regimes according to commonly used proxies (corruption, rule of law, bureaucracy quality, and secondary school enrollment), which displayed patterns consistent with the theory. Specifically, electoral autocracies followed by electoral democracies display the highest levels of clientelism, followed by closed autocracies and then liberal democracies. Additionally, a variety of indicators for the public-private orientation of the corporate sector were analyzed. They consistently showed that public control of the corporate sector is highest in closed autocracies and declines as one moves toward liberal democracies. As part of this analysis, the SOE concentration of different industries was examined. The evidence showed that SOE concentration varies across industries but is generally quite high in the infrastructure sectors that are primarily involved with BRI projects. The key question of this chapter is: which countries are the most receptive to Chinese foreign spending? I theorize that electoral autocracies will be the most avid recipients of Chinese foreign construction spending due to their convergence of interests with Chinese SOEs. Specifically, electoral autocrats have a high demand for clientelist resources that can be delivered relatively quickly due to electoral pressures, and in an opaque manner so as to minimize accusations of corruption by political opponents. This matches China’s desire to quickly deliver infrastructure projects both to address excess capacity issues for its SOEs and to quickly introduce its technical standards that promote de facto standardization before non-Chinese rivals can introduce their technologies and 138

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

Political Regimes and BRI Country-Level Patterns

139

standards. Closed autocracies and electoral democracies will have fewer interests aligned with China, and less avidly seek Chinese infrastructure spending. Liberal democracies will have the smallest demand for Chinese infrastructure spending. To assess these posited outcomes, this chapter analyzes Chinese foreign spending since the introduction of the BRI at the end of 2013. To assess whether Chinese foreign spending matches my expectations, it is important distinguish between construction spending and investment. As mentioned earlier, construction spending is typically associated with infrastructure projects by SOEs that end after a finite period of time (the construction phase often lasting up to five years with maintenance of the asset lasting for two or more decades in the context of a PPP). Investments confer indefinite ownership rights to the foreign entity, as with mergers and acquisitions and greenfield projects. I test my argument with data from the China Global Investment Tracker (CGIT) dataset first introduced in Chapter 3 which distinguishes between construction and investment spending. I conduct multivariate tests for the posited relationships during the BRI timeframe, and then extend the analysis to the 2005–2019 period to identify whether the patterns identified during the BRI are simply part of a longer-term pattern. I then also examine evidence for changes to countries’ logistics performance to determine whether electoral autocracies have improved at a faster rate than other regimes. Finally, I examine the share of Chinese exports that has gone to different political regimes since 2000 to identify whether there is a marked change since the launch of the BRI. The evidence indicates that electoral autocracies are the major recipients of Chinese foreign construction spending and foreign investment during the BRI timeframe, from 2014 to 2019. The relationship between electoral autocracies and Chinese construction spending is particularly strong and robust. Additionally, there is a strong and significant relationship between the level of corruption in host countries relative to China and Chinese construction spending. Together these findings suggest clientelism in recipient countries, and especially in electoral autocracies, may play an important role in driving the level of construction spending. Extending the timeframe to 2005 to 2019, the findings indicate there is a substantive difference in the relationship between Chinese foreign construction spending and electoral autocracies (and electoral democracies) that occurs with the initiation of the BRI. This effect is clearest in the context of regimes with a low level of durability which proxies for political rulers with an insecure hold on power. The logistics performance indicators (infrastructure and tracking and tracing) also show that electoral autocracies have displayed the

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

140

China’s Chance to Lead

greatest improvement from before to after the introduction of the BRI. Finally, the share of Chinese exports flowing to electoral autocracies does increase during the BRI time period, but it is not possible to conclude this is a deviation from previous trends. It will take more time to establish confidence for these effects. The main takeaway from the analysis in this chapter is the exceptional role electoral autocracies play in attracting Chinese foreign spending in the context of the BRI, especially when they are not durable. The chapter proceeds with a brief background discussion of China’s foreign spending activity before examining general trends over time with descriptive and then more rigorous multivariate regression analyses for both the BRI time period and then the 2005 to 2019 timeframe. I then examine trade effects, first by analyzing changes in political regimes’ logistics performance indicators and then the share of Chinese exports to political regimes. The appendix provides additional details and analysis relating to the tests and findings presented in the main body of the chapter. Overall, the strong relationship identified between electoral autocracies and Chinese construction spending during the BRI calls for more detailed analysis of BRI projects to understand how their characteristics vary across regime types, to be examined in Chapter 7. 6.1

Background on China’s Foreign Spending Activity

Prior to the initiation of China’s Going Out policy in 2000, China’s foreign development financing focused on supporting the country’s economic development via export credits and subsidized loans, especially to SOEs (Dreher et al. 2018). The Going Out policy coincided with China becoming a member of the World Trade Organization in 2001. This new international positioning of the Chinese economy invigorated support for industries and activities that would enhance economic restructuring (e.g., value-chain upgrading and integration) and resource allocation optimization (e.g., reducing the cost of raw materials and labor). For example, some firms (mostly SOEs) were granted preferential access to foreign exchange. Subsidies to qualifying projects, such as building foreign R&D centers, equipment manufacturing facilities, economic and trade zones, and investments in natural resources, were initiated. Although both private firms and SOEs could tap this government assistance in principle, SOEs typically maintained privileged access. Following a rapid increase in foreign investment starting in 2005, China became the world’s third largest outward investor in 2012, following the United States and Japan. When the BRI was launched at the end

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

Political Regimes and BRI Country-Level Patterns

141

of 2013, many of the pre-existing “Going Out” policies were retained (Sauvant and Chen 2014). But one major difference was the enhanced state support to SOEs involved in infrastructure projects. Many SOEs received jolts of subsidized financing to prevent a decline in economic growth following the Global Financial Crisis of 2008. Excess debt and the need to find new business activities to pay off their debts and prevent massive layoffs contributed to heightened involvement in foreign construction projects.1 The BRI added new sources of financing, such as access to a large pool of loan capital from central policy banks (like the Export-Import Bank) and local financing, including targeted Belt and Road Initiative lending from the “Silk Road Fund” and other centraland local-government external trade and investment initiatives to promote the project.2 This financial assistance was often supplemented by the central government through diplomatic channels and government programs such as the Forum on China-Africa Cooperation (FOCAC). But a crucial determinant of this involvement concerns the process for identifying and participating in new projects. Although SOEs can propose projects to foreign governments, it is ultimately a recipient-led process that typically begins with requests from foreign governments (Jones and Hameiri 2020). As discussed in prior chapters, there are several important features of Chinese foreign economic activity that are different from traditional foreign investment originating from the United States, Japan, and other advanced economies. First, Chinese firms are involved in two types of foreign spending activity. There is traditional foreign investment which involves the purchase of a foreign asset such as land or a company, and which confers ownership rights that can last indefinitely. There is also foreign construction activity, such as the building of highways or power plants, but only lasts for a predefined period of time. Traditional foreign investment measures focus primarily on the first type of investment activity. The second notable difference is the importance of state-owned enterprises to Chinese foreign spending. Their motivations and capabilities can differ markedly from their private sector counterparts because of the 1

Overcapacity was most problematic in the construction industry, but overproduction also existed for aluminum, cement, chemicals, coal, coal-fired power, paper, plated glass, shipbuilding, solar power, and steel (International Monetary Fund Country Report, 2017: no. 17.247, p. 18). The government’s objective to undertake foreign infrastructure and industrial projects to alleviate overcapacity is stated in a 2013 directive, “Guiding Opinions of the State Council on Resolving Serious Production Overcapacity Conflicts,” by The Central People’s Government of the PRC, 6 October 2013, www.gov.cn/zwgk/201310/15/content_2507143.htm (accessed March 24 2022). 2 See Dutton, Kardon, and Kennedy (2020) for additional details.

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

142

China’s Chance to Lead

need to balance commercial and political priorities and their privileged access to state-backed lending and access to domestic stock markets (Shen et al. 2009; Lai 2011; Liang et al. 2017). The potential supply of such lending is enormous due to China’s huge store of foreign exchange reserves. Third, the objectives that SOEs pursue and the types of markets Chinese private firms prefer to enter also depend on the various characteristics of host countries. Aside from the availability of natural resources, scholars have begun paying greater attention to host country factors such as the level of economic development and the desire for infrastructure projects, the legal system, and property rights protections (Liu et al. 2017; Tuman and Shirali 2017; Chen and Lin 2018). However, I argue the structure of host country political regimes is a major factor influencing the varying levels of Chinese BRI spending across countries. 6.2

Trends and Patterns of the Belt and Road Initiative

To investigate Chinese foreign investment and construction spending, I use the China Global Investment Tracker (CGIT) dataset. To my knowledge, it offers the most comprehensive and up-to-date coverage regarding Chinese firms’ foreign spending that distinguishes between investment and construction activities. The CGIT dataset includes all verifiable investment and construction transactions worth $100 million or more since 2005.3 Data come from primary sources provided by MOFCOM, Chinese corporate sources and partners, and independently derives numbers for Chinese foreign spending. The dataset is revised biannually to account for revisions by these sources. Construction activity is valuable but does not involve ownership. It is typically performed by massive state-owned enterprises, such as Sinomach, China Communications Construction, State Construction Engineering, and others. Investment includes transactions that involve the purchase of a foreign asset, such as a corporate acquisition or a greenfield investment. The dataset used for this study is from Spring 2020 and includes 1706 investments worth more than $1.2 trillion. It also includes 1,749 construction projects worth over $830 billion from 2005 up to the end of 2019, which marks a natural end point due to the onset of the COVID-19 pandemic in January 2020. 3 The CGIT dataset does not include trade, bonds, loans, aid, bank deposits, and other forms of storing foreign exchange overseas that do not involve ownership or services in the host country.

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

Political Regimes and BRI Country-Level Patterns

6.2.1

143

The Belt and Road Initiative

Figures 6.1 to 6.3 illustrate the annual investment and construction activities for BRI projects by political regime. Figure 6.1 illustrates that electoral autocracies have consistently outpaced other political regimes with regard to the total volume of investment and construction activity they have received; more than twice the levels of electoral democracies, typically the second highest. Figure 6.2 presents investment levels, which again show electoral autocracies as the largest recipient, though at a level that is less pronounced than Figure 6.1. Because developing countries are the primary recipients of BRI spending, the volumes going to liberal 80,000

Millions of USD

70,000 60,000 50,000 40,000 30,000 20,000 10,000 0 2013

2014

2015

2016

Closed Autocracy Electoral Autocracy

2017

2018

2019

Electoral Democracy Liberal Democracy

Figure 6.1 BRI foreign investment and construction spending combined, 2013–2019 35,000

Millions of USD

30,000 25,000 20,000 15,000 10,000 5,000 0 2013

2014

2015

Closed Autocracy Electoral Autocracy

2016

2017

2018

Electoral Democracy Liberal Democracy

Figure 6.2  BRI foreign investment, 2013–2019

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

2019

144

China’s Chance to Lead

60,000

Millions of USD

50,000 40,000 30,000 20,000 10,000 0 2013

2014

2015

2016

Closed Autocracy Electoral Autocracy

2017

2018

2019

Electoral Democracy Liberal Democracy

Figure 6.3  BRI foreign construction spending, 2013–2019

democracies (typically advanced economies) are at a far lower level than the total foreign investment data show. Figure 6.3 displays construction activity, with electoral autocracies far exceeding other regime types, typically around four to five times higher. To provide more detail for the general trends shown in Figures 6.1 to 6.3, Table 6.1 displays total investment and construction values, country-year values, and the first and second most heavily invested sectors. The total numbers reveal electoral autocracies receiving nearly three times more BRI-affiliated spending than electoral democracies, the second highest regime category as displayed in Panel A. Disaggregating these numbers by investment and construction (Panels B and C) reveals similar patterns. Electoral autocracies receive twice as much investment spending as electoral democracies and liberal democracies (tied for second). Electoral autocracies receive nearly three times as much construction activity as electoral democracies, the second highest category. The biggest difference between the total sample and the BRI subsample is for liberal democracies, followed by electoral democracies. This difference is primarily due to investment spending, rather than construction, which is consistent with BRI lending being led by SOEs. With regard to country-year values, the highest mean and median scores both go to electoral autocracies for the sum of investment and construction activity. Mean investment values follow the expected distribution for private business, descending from liberal democracies to closed autocracies, though the median values are less consistent. With regard to construction activity, the distribution of both mean and median values matches the expectations of SOE spending, descending

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

276,380 92,020 184,360 23,800 124,480 63,720 60,090

756,250 514,880 241,370 23,800 125,130 112,770 490,010

470,510 56,880 413,630 79,150 266,440 98,920 25,210

Panel B. Investment All Economies Advanced Economies Emerging Economies Closed Autocracies Electoral Autocracies Electoral Democracies Liberal Democracies

Panel C. Construction All Economies Advanced Economies Emerging Economies Closed Autocracies Electoral Autocracies Electoral Democracies Liberal Democracies 425,400 36,980 388,420 78,650 248,230 84,210 13,520

701,780 129,000 572,780 107,540 376,890 132,450 79,820

35 3 32 6 20 7 1

23 8 15 2 10 5 5

57 11 47 9 31 11 7

341 49 292 53 173 90 23

212 55 157 29 91 44 45

413 83 330 59 184 109 57

1,248 755 1,330 1,479 1,500 831 522

1,304 1,673 1,174 877 1,274 1,453 1,484

1,699 1,554 1,736 1,759 2,040 1,273 1,382

660 520 715 920 890 425 360

600 680 590 610 600 640 570

870 700 930 1,110 1,165 600 570

1,460 756 1,532 1,473 1,674 944 576

1,826 2,553 1,479 920 1,859 1,772 2,234

2,141 2,304 2,100 1,932 2,402 1,738 2,047

BRI BRI by Country-year, 2014–2019 2014–2019 Percent (mill. USD) of Total Obs Mean Median Std Dev

Panel A. Investment + Construction All Economies 1,226,760 Advanced Economies 571,760 Emerging Economies 655,000 Closed Autocracies 108,040 Electoral Autocracies 379,830 Electoral Democracies 211,600 Liberal Democracies 521,960

Total 2014–2019 (mill. USD)

Table 6.1  BRI foreign investment and construction spending by political regime

Energy and Transport Transport and Real Estate Energy and Transport Energy and Transport Energy and Transport Energy and Transport Transport and Real Estate

Energy and Real Estate Real Estate and Transport Energy and Metals Energy and Transport Energy and Real Estate Energy and Metals Real Estate and Transport

Energy and Transport Transport and Real Estate Energy and Transport Energy and Transport Energy and Transport Energy and Transport Transport and Real Estate

1st and 2nd sectors, 2014–19

146

China’s Chance to Lead

in size from closed autocracies to liberal democracies. The disaggregated investment and construction values thus illustrate that electoral autocracies are the largest recipients of Chinese foreign spending because of the convergence of investment (private business) and construction (SOEs) spending. Finally, the top two sectors for investment and construction spending are consistently energy and transport; real estate is third and is primarily in the investment category; metals is a distant fourth. Overall, the BRI results demonstrate the importance of electoral autocracies as the main recipient of Chinese foreign investment and construction spending. 6.2.1.1 Multivariate Regressions: Methods and Variables To what extent do political regimes affect Chinese foreign investment and construction spending across countries? And do electoral autocrats with a weak or insecure hold on power attract more Chinese spending than leaders with secure rule? Chinese foreign investment and construction data (CGIT dataset) include continuous variables with right-skewed distributions (see Table 6A.2 in the chapter appendix). The accepted approach to estimating effects for this type of data is to use the fixed effects Poisson estimator (Wooldridge 1999, 2010).4 In accordance with empirical work on foreign investment, I also follow a gravity-type empirical specification commonly used in previous studies (Feenstra et  al. 2001; Yeaple 2003; World Bank 2018).5 The dependent variables include Chinese foreign investment and Chinese foreign construction spending by country-year. Because observations are only included if the amount is USD100 million or more, I examine a second dataset (Refinitiv) on BRI projects which does not include such value restrictions in Chapter 7. However, its major limitation is its omission of non-BRI spending. Together, the two datasets generate confidence in the overall findings. All tests below include year fixed effects to reduce the likelihood of attributing changes to Chinese foreign spending due to some simultaneously experienced shock. Country fixed effects are also tested in some models in order to assess whether the results are robust to the inclusion of constant features of the countries being analyzed but which are not being theorized, such as legal family or region. Additionally, country 4

Specifically, the fixed effects Poisson estimator is fully robust for estimating the conditional mean parameters. See Wooldridge (1999, 2010). 5 Following a study on Chinese foreign investment by the World Bank (2018) and work by Santos and Tenreyro (2006), I repeat the analysis using a Poisson Pseduo Maximum Likelihood (PPML) Estimator. The results are qualitatively similar.

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

Political Regimes and BRI Country-Level Patterns

147

fixed effects illuminate whether a change to a country’s political regime is statistically related to a change in Chinese foreign spending. Many tests do not include country fixed effects in order to explicitly measure the effect of country-specific attributes of interest such as infrastructure spending and distance from China. In these cases, tests using country fixed effects are displayed alongside those without them to illustrate whether the results change in a substantive way. The results are based on robust standard errors with a correction for clustering on countries. Results for three sets of explanatory variables are reported. The first set of explanatory variables includes the political regime categories: Closed Autocracy, Electoral Autocracy, and Electoral Democracy. Liberal Democracy is the excluded category, thus the results for each regime category should be interpreted in relation to it. Liberal Democracy is the excluded category because I am interested in identifying the influence of autocratic regimes on Chinese foreign spending in addition to their relation to electoral democracy which comprises a large fraction of emerging economies. Liberal democracies are almost entirely located among advanced economies, thus making it an appropriate category to exclude for the examination of effects among emerging economies. The second set of explanatory variables includes Corruption Difference and Durable. Corruption Difference measures the level of corruption in China in relation to that in the recipient country based on data provided by the World Bank’s Governance Indicators; higher values indicate a higher level of corruption in the recipient country in comparison to China and suggest a greater propensity for clientelism. It is necessary to consider the recipient country’s level of corruption in relation to China and not simply the absolute value or allowing the political regime to proxy for it due to the initiation of Xi Jinping’s anti-corruption policies starting in 2013. Durable is a proxy for the strength and stability of a political regime and its ruler’s hold on power. I use a measure for the durability of a political regime which measures the number of years since the most recent regime change or the end of a transition period defined by the lack of stable political institutions.6 The coding of the variable comes from the Polity V dataset which defines regime change as a threepoint change in the POLITY score over a period of three years or less. The third set of explanatory variables include Infrastructure spending, Distance, and Log(GDP per capita). These are widely used variables for tests on FDI involving western firms, as discussed in Chapter 3. Infrastructure spending measures the percentage of national GDP a country 6

The lack of stable political institutions is identified by a standardized authority score according to the Polity V description of the regime durability variable.

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

148

China’s Chance to Lead

spends on infrastructure development in 2011. The Distance variable measures the distance by sea between a country and China.7 Log(GDP per capita) in 2010 US dollars is used as well. For studies on Chinese foreign spending, this variable is commonly used as a proxy for real wages due to the difficulties of getting average wage data for low-income economies (Liu et al. 2017; Tuman and Shirali 2017; Kang et al. 2018). It measures the advantage of a country in attracting foreign manufacturing firms. A number of other control variables are also included to reduce the risk of drawing inappropriate inferences about the effects of political regimes. The control variables included are those commonly used to explain foreign investment patterns regardless of country of origin (Jensen 2003; Busse and Hefeker 2007; Daude and Stein 2007; Lucke and Eichler 2016) and those which have been identified as particularly relevant to Chinese foreign spending (Liu et al. 2017; Tuman and Shirali 2017; Kang et al. 2018). I use the variable Market size, measured by the natural log of GDP (in 2010 US dollars) of the host country, and Population growth in the host country in order to account for a country’s current and future market potential. Trade openness is calculated by the sum of exports and imports divided by GDP in the host country. As a proxy for macroeconomic stability, the variable Inflation is measured by the annual percentage change in the consumer price index in the host country. The variable Natural resources is used to represent the present value of future rents from subsoil natural resources (fuels, metals and ores) as a ratio of GDP in the host country. To account for the attractiveness of a national environment to private business and investors, I use the variables Business freedom and Investment freedom which come from the Economic Freedom in the World Index provided by the Heritage Foundation. I also include the variable Rule of law in order to account for the varying strength of legal systems across emerging economies in protecting private investment. For example, Singapore has a strong commitment to protecting private investment although it is an electoral autocracy.8 For the tests relating to Investment as the dependent variable, I also use a one-year lag of the Contract variable which measures construction spending for the prior year to assess whether this impacts investment in the subsequent year. 7

In some cases, distance by air or geographic distance between the two capital cities is used due to collinearity issues. 8 The Political Constraints III and V (polconiii and polconv) variables were tested but are not included in the results due to collinearity problems with the regime variables.

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

Political Regimes and BRI Country-Level Patterns

149

6.2.1.2 Multivariate Regressions: Findings Table 6.2 displays the results for BRI Construction spending. Note that models (1) and (4) include country fixed effects to provide baseline results for comparison when country fixed effects are not included. All models included year fixed effects and control variables. The first notable finding is that Electoral Autocracy is significant (typically at the 1% or 5% level) across all models. Model (1) indicates that electoral autocracies attract more than USD1.8 million in Chinese construction spending on average per year in comparison to liberal democracies. These results support the main argument of this book that electoral autocracies will receive a larger share of Chinese foreign construction spending in the context of the BRI. Another notable point related to this finding is that Closed Autocracy is not significant in any of the models, underscoring the need to distinguish between types of autocratic regimes. Further, Electoral Democracy does not display consistently significant results, and for those models in which the results are significant, the magnitudes of the coefficients are markedly smaller than those for Electoral Autocracy. Second, the magnitudes of the coefficients for Electoral Autocracy in the tests with and without fixed effects (Models 1 and 2) are notably different (1.878 and 0.623) with greater significance for the test with fixed effects (at the 5% level as compared to 10% for the test without fixed effects). The results for the test with fixed effects suggest regime change to or away from electoral autocracy correlates with a significantly and substantively larger effect than the pooled sample without country fixed effects.9 Third, Corruption Difference is strongly significant across all models, indicating that countries with a higher level of corruption in relation to China receive a higher level of Chinese foreign spending. Specifically, a one-point positive difference in the level of corruption in the recipient country in relation to China corresponds to an increase in construction spending of USD0.8 to 0.9 million per year, on average, in Model (1). This finding is notable because it is contrary to existing literature regarding FDI from MNCs which typically identifies a negative relationship. Additionally, it is consistent with the argument that clientelism increases the demand for Chinese foreign construction spending. Fourth, Model (2) indicates that Infrastructure spending in the recipient country displays a significant positive relationship to Chinese foreign construction spending. However, Model (5) shows there is not a significant interaction effect with respect to political regime. 9

Inspection of the sample reveals there are twelve instances of regime change, including five instances involving electoral autocracies.

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

150

China’s Chance to Lead

Table 6.2  BRI construction spending DV: Contract

Closed Autocracy Electoral Autocracy Electoral Democracy

1

2

3

4

5

0.492 (0.46) 1.878** (2.56) 0.618 (1.28)

−0.082 (−0.17) 0.623* (1.86) 0.246 (0.77) 0.142*** (2.86)

−12.255 (−0.82) 4.576*** (4.42) 2.751*** (2.68)

0.207 (0.35) 1.560*** (3.60) 1.120** (2.14) 0.129*** (2.72)

0.152** (2.41) 0.041 (0.17) −0.201*** (−3.11) −0.137** (−2.12)

0.033 (1.58) −0.019 (−0.74) −0.048** (−2.21) −0.038 (−1.64)

0.929*** (4.16)

2.069 (0.39)

0.225** (2.17) −0.006 (−0.05) −0.229* (−1.71)

0.023 (0.01) 1.778** (2.57) 1.238 (1.29) 0.099 (0.41) 0.106 (0.07) −0.064 (−0.22) −0.017 (−0.03) 0.037 (1.62) −0.032 (−0.54) −0.048 (−1.60) −0.030 (−1.04) 0.000 (0.00) −0.001 (−0.15) −0.005 (−0.38) 0.261** (2.39) 0.054 (0.51) −0.262* (−1.93)

Y Y Y 195

Y N Y 204

Y N Y 204

Infrastructure spending Closed Autocracy*Infra. spending Electoral Autocracy*Infra. spending Electoral Democracy*Infra. spending Durable Closed Autocracy* Durable Electoral Autocracy* Durable Electoral Democracy* Durable Closed Autocracy*Infra. spending*Durable Electoral Autocracy*Infra. spending*Durable Electoral Democracy*Infra. spending*Durable Corruption Difference 0.851*** (4.13) Distance Log(GDP per capita)

2.699 (0.54)

0.193** (2.04) 0.026 (0.24) −0.284** (−2.37)

Controls Country fixed effect Year fixed effect N

Y Y Y 199

Y N Y 209

Note: *, **, and *** indicate significance at the 0.10, 0.05, and 0.01 level, respectively. Standard errors are clustered at the country level. Controls include Market Size, Population Growth, Trade Openness, Inflation, Natural Resources, Business Freedom, Investment Freedom, and Rule of Law.

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

Political Regimes and BRI Country-Level Patterns

151

Model (3) introduces the Durable variable with country fixed effects. The results indicate that electoral autocracies that are less durable attract a larger volume of Chinese construction spending than electoral autocracies that are more durable (see Figure 6A.1 in the chapter appendix for a plot of these interaction effects). Specifically, Electoral Autocracies that have lasted for five years will attract USD3 million more Chinese construction spending than equivalent liberal democracies, and about USD2 million more than Electoral Autocracies that have lasted for twelve years, on average. Similar results are obtained for Electoral Democracies, but the effect is smaller and less significant. Model (5) shows these results hold when country fixed effects are not included though the magnitude of the effect is much smaller (−0.201 and −0.048 for Models 3 and 4, respectively), mirroring the results for Models 1 and 2. Finally, Model (5) tests whether there are significant results for the triple interaction of regime type, durability, and infrastructure spending. The results for Electoral Autocracy do not yield significant results, nor do those for Electoral Democracy or Closed Autocracy. Overall, the findings support the argument that electoral autocracies attract a larger share of Chinese construction spending than other regimes, and this is amplified when the durability of the regime is low.10 A higher level of corruption in the recipient country relative to China is also a significant predictor of a larger volume of Chinese foreign construction spending flowing to the host country. Table 6.3 displays the results for BRI Foreign Investment. Like the BRI Construction tests, I include country fixed effects (models 1, 3, and 5) to provide baseline results for comparison when country fixed effects are not included. All models included year fixed effects and control variables. The first notable finding is the consistently negative and statistically significant results (at either the 5% or the 1% level) for Electoral Autocracy country fixed effects models. Model (1) indicates that electoral autocracies attract USD1.6 million less in investment per year than liberal democracies, on average. This result holds when country fixed effects are not included in Model (2). Notably, the magnitude of the coefficient for Electoral Autocracy in the fixed effects test in Model 1 (−1.555) displays a larger difference with the test without fixed effects in Model 2 (−0.901). This suggests Chinese private firms are also responsive to regime change involving Electoral Autocracy, mirroring the results of Table 6.2 regarding Chinese firms involved in construction spending (principally SOEs). 10

Repeating the tests using a Poisson Pseudo Maximum Likelihood (PPML) Estimator model yields qualitatively equivalent results.

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

18.943 (1.16) Y Y Y 96

−1.009*** (−6.03)

−0.990 (−1.01) −1.555** (−2.47) 2.182*** (3.97)

1

−0.359** (−2.06) 0.136*** (3.04) −0.366** (−2.30) −0.294 (−1.33) Y N Y 113

−1.814*** (−3.36) −0.901* (−1.92) −0.321 (−0.79)

2

−8.593 (−0.36) Y Y Y 51

0.000*** (3.15) −1.992*** (−7.80)

3.418 (1.03) −3.690*** (−2.88) 4.797* (1.80)

3

0.000 (1.61) −0.380 (−1.12) 0.219*** (3.29) 0.510** (2.05) −0.468 (−1.43) Y N Y 66

−1.266 (−1.40) −0.541 (−0.98) 0.264 (0.49)

4

DV: Investment

14.944 (0.90) Y Y Y 96

−1.048*** (−4.11)

−36.696*** (−3.06) −12.320*** (−3.85) −8.346*** (−2.70) −0.419* (−1.86) 0.829** (2.54) 0.430** (2.07) 0.770*** (3.42)

5

−0.352* (−1.93) 0.109* (1.91) −0.301* (−1.65) −0.205 (−0.73) Y N Y 112

−2.681* (−1.81) −1.438 (−1.00) −0.223 (−0.16) −0.021 (−0.42) 0.029 (0.57) 0.013 (0.26) −0.018 (−0.36)

6

−2.065* (−1.75) −1.103 (−1.00) 0.423 (0.36) −0.002 (−0.05) −0.010 (−0.23) −0.021 (−0.47) −0.051 (−1.27) 0.000 (1.28) −0.318 (−0.81) 0.262*** (3.46) −0.034 (−0.09) −0.406 (−1.22) Y N Y 75

7

Note: *, **, and *** indicate significance at the 0.10, 0.05, and 0.01 levels, respectively. Standard errors are clustered at the country level. Controls include Market Size, Population Growth, Trade Openness, Inflation, Natural Resources, Business Freedom, Investment Freedom, and Rule of Law.

Controls Country fixed effect Year fixed effect N

Log(GDP per capita)

Distance

Infrastructure Spending

Corruption Difference

1 year lag contract

Electoral Democracy* Durable

Electoral Autocracy* Durable

Closed Autocracy* Durable

Durable

Electoral Democracy

Electoral Autocracy

Closed Autocracy

Table 6.3  BRI investment

Political Regimes and BRI Country-Level Patterns

153

The second notable result relates to Corruption Difference. The results indicate a significant negative relationship (at the 1% level) for all country fixed effects tests. On average, a one-point increase in the level of corruption between China and the recipient country corresponds to a decline of around USD1 million in investment according to the results in Model (1). This relationship mirrors findings for private investment in the context of Western FDI and is opposite to the findings reported in Table 6.2 regarding construction spending. Third, I test whether a 1-year lag in construction spending corresponds to an increase in investment in models (3) and (4). The results indicate a significant positive effect (at the 1% level) in model (3) with country fixed effects only, though the magnitude of the effect is very small. Fourth, I test whether the durability of the regime interacted with regime type impacts investment flows in models (5) and (6). The results for country fixed effects in Model (5) indicate electoral autocracies with low durability will attract less investment than equivalent liberal democracies (see Figure 6A.2 in the chapter appendix for a plot of these interaction effects). However, electoral autocracies with high durability do not display any noticeable difference in attracting Chinese investment than liberal democracies. Electoral democracies and closed autocracies display similar results. These results are what one would expect to see in relation to private investors from Western countries. Finally, model (7) tests whether the one-year lag in construction spending affects the level of Chinese investment observed with regard to the interactions between Durable and political regimes. The results do not indicate a significant relationship. Overall, the important takeaways include the negative relation between electoral autocracies and Chinese foreign investment according to the country fixed effects tests. Additionally, countries with higher levels of corruption in relation to China display a negative relationship to Chinese investment. These results are consistent with what one would expect in relation to private foreign investment and opposite to the findings regarding Chinese foreign construction spending. 6.3

Chinese Foreign Investment and Construction Activity, 2005–2019

As mentioned in the background section, the BRI builds on policies that promoted foreign spending prior to its introduction in 2013. This section therefore assesses to what extent the findings are specific to the BRI versus the period preceding it.

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

154

China’s Chance to Lead

160,000

Millions of USD

140,000 120,000 100,000 80,000 60,000 40,000 20,000

20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 20 14 20 15 20 16 20 17 20 18 20 19

0

Closed Autocracy

Electoral Democracy

Electoral Autocracy

Liberal Democracy

Figure 6.4 Chinese foreign investment and construction spending, 2005–2019 140,000

Millions of USD

120,000 100,000 80,000 60,000 40,000 20,000

20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 20 14 20 15 20 16 20 17 20 18 20 19

0

Closed Autocracy

Electoral Democracy

Electoral Autocracy

Liberal Democracy

Figure 6.5  Chinese foreign investment, 2005–2019

Figures 6.4 to 6.6 report the annual totals for investment plus construction activity, investment activity, and construction activity by regime type, respectively. Figure 6.4 illustrates that from 2005 to 2015, liberal democracies and electoral autocracies were the top recipients for Chinese foreign spending. In 2016 and 2017, liberal democracies suddenly spike upwards before falling back down in 2018 and sinking below electoral autocracies in 2019 largely due to restrictions imposed on Chinese investments, led by the United States. Throughout this entire period, electoral democracies and closed autocracies ranked third

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

Political Regimes and BRI Country-Level Patterns

155

60,000

Millions of USD

50,000 40,000 30,000 20,000 10,000

20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 20 14 20 15 20 16 20 17 20 18 20 19

0

Closed Autocracy

Electoral Democracy

Electoral Autocracy

Liberal Democracy

Figure 6.6  Chinese foreign construction spending, 2005–2019

and fourth, respectively. Figure 6.5 confirms that the spike in 2016 and 2017 for liberal democracies is due to investments, primarily by private firms. Electoral autocracies and electoral democracies are tied for second throughout the sample while closed autocracies are consistently last. Figure 6.6 displays construction activity. Throughout the entire time period, electoral autocracies consistently attract the largest share of construction activity. From 2010 onward electoral democracies are second followed by closed autocracies and then liberal democracies. 6.3.1

Multivariate Regressions

Tables 6.4 and 6.5 display the results for tests relating to Chinese foreign construction spending and Chinese foreign investment using data for the entire sample period, 2005–2019. Dummy variables are included to indicate whether foreign spending occurred in the time period for 2014 and later. In Table 6.4, Models (1) to (4) display results for Chinese foreign construction spending with interactions included for political regimes with the 2014 and later variable as well as interactions for political regimes with infrastructure spending. Models (1) and (2) use the full sample whereas Models (3) and (4) use the SOE subsample by coding whether firms are a SOE or not. The two SOE tests have slightly fewer observations (n=434 and n=444) than the full sample tests (n=451 and n=460), confirming the importance of SOEs to Chinese foreign construction spending. There are two notable results. First, Electoral Democracy displays a positive significant relation to Chinese foreign construction spending in three of the four tests

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

Electoral Democracy*2014 and later

Electoral Autocracy*2014 and later

Closed Autocracy*2014 and later

2014 and later

Electoral Democracy*Infra. spending

Electoral Autocracy*Infra. spending

Closed Autocracy*Infra. spending

Infra. spending

Electoral Democracy

Electoral Autocracy

Closed Autocracy

1.162 (1.45) −0.069 (−0.14) 0.149 (0.37) −0.304 (−0.77)

1.398 (1.08) 1.156 (1.18) 1.912*** (2.74)

1 0.331 (0.69) 0.679* (1.68) 0.947** (2.18) 0.103 (0.56) 0.045 (0.25) −0.114 (−0.61) −0.326 (−1.61) 1.084** (2.01) −0.514 (−1.02) 0.155 (0.48) −0.285 (−0.83)

2

Full Sample

Table 6.4  Chinese foreign construction spending, 2005–2019

1.458* (1.73) 0.046 (0.11) 0.149 (0.38) −0.268 (−0.71)

0.327 (0.22) 0.285 (0.23) 1.267 (1.34)

3 0.334 (0.76) 0.794** (2.09) 1.111*** (2.65) 0.167 (0.95) −0.009 (−0.05) −0.177 (−0.99) −0.418** (−2.13) 1.239** (2.16) −0.405 (−0.88) 0.149 (0.47) −0.251 (−0.73)

4

SOE Subsample

0.416 (0.41) 1.911* (1.65) 1.331** (2.09) 0.636 (0.96)

1.011 (1.37) 0.158 (0.36) −0.154 (−0.38)

5

0.496 (0.71) −0.692 (−0.88) 0.920* (1.78) 0.610 (1.21)

0.890 (1.61) 0.457* (1.67) 0.427 (1.42) 0.078** (2.40)

6

Full Sample

DV: Contract

0.665 (0.66) 1.571 (1.53) 1.390** (2.49) 0.783 (1.37)

0.985 (1.46) −0.121 (−0.34) −0.426 (−1.26)

7

0.676 (0.90) −0.937 (−1.24) 0.865 (1.62) 0.628 (1.22)

1.038** (2.03) 0.408 (1.47) 0.364 (1.18) 0.083** (2.45)

8

SOE Subsample

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

0.183 (1.42) −2.260 (−0.97) 0.614 (0.66) Y Y Y 451

0.097 (1.28) −0.267*** (−3.27) 0.061 (0.81) Y N Y 460

0.160 (1.15) −2.129 (−0.89) 2.397*** (2.75) Y Y Y 434

0.089 (1.21) −0.262*** (−3.15) 0.080 (1.11) Y N Y 444

0.004 (0.17) −0.023 (−0.79) −0.006 (−0.22) 0.048*** (2.76) 0.027 (1.26) −0.061* (−1.80) −0.089*** (−3.33) −0.052** (−2.36) 0.268** (2.31) −3.320 (−1.62) 0.738 (0.86) Y Y Y 445

0.003 (0.79) −0.007 (−0.48) 0.001 (0.10) −0.007 (−1.06) 0.015 (1.17) 0.001 (0.05) −0.042** (−2.32) −0.036** (−2.53) 0.116 (1.56) −0.193*** (−2.75) −0.006 (−0.08) Y N Y 454

0.009 (0.43) −0.040 (−1.50) −0.012 (−0.52) 0.045*** (3.74) 0.026 (1.31) −0.045 (−1.45) −0.087*** (−3.38) −0.054*** (−2.63) 0.239* (1.88) −3.360 (−1.59) 2.352*** (2.78) Y Y Y 428

0.001 (0.26) −0.013 (−0.92) 0.003 (0.40) −0.004 (−0.63) 0.014 (1.03) 0.011 (0.45) −0.041** (−2.22) −0.037** (−2.49) 0.109 (1.49) −0.178** (−2.56) 0.004 (0.05) Y N Y 438

Note: *, **, and *** indicate significance at the 0.10, 0.05, and 0.01 level, respectively. Standard errors are clustered at the country level. Controls include Market Size, Population Growth, Trade Openness, Inflation, Natural Resources, Business Freedom, Investment Freedom, and Rule of Law.

Controls Country fixed effect Year fixed effect N

Distance

GDP per capita

Closed Autocracy*2014 and later*Durable Electoral Autocracy*2014 and later*Durable Electoral Democracy*2014 and later*Durable Corruption Difference

Durable*2014 and later

Electoral Democracy*Durable

Electoral Autocracy*Durable

Closed Autocracy*Durable

Durable

158

China’s Chance to Lead

Table 6.5  Chinese foreign investment, 2005–2019 DV: Investment Full Sample

Closed Autocracy Electoral Autocracy Electoral Democracy

1

2

3

4

0.622 (0.36) −1.491 (−1.02) −1.556 (−1.49)

−1.289 (−1.48) −1.276* (−1.88) −0.905 (−1.17) −0.999* (−1.85) 1.031* (1.90) 1.112** (2.09) 0.811 (1.43) −0.914 (−0.91) −0.654 (−0.69) −0.878 (−0.93) −0.284 (−0.30)

−0.076 (−0.13) −0.115 (−0.21) −0.086 (−0.14)

−0.973 (−0.97) −0.422 (−0.48) 0.224 (0.34) −0.530 (−0.35) −0.604 (−0.87) −0.053 (−0.12) 0.764* (1.67)

−1.452 (−1.56) −1.432* (−1.71) −1.348* (−1.65) −1.197** (−2.29) 1.238** (2.35) 1.250** (2.28) 1.080** (1.99) −0.394 (−0.63) −0.899 (−1.55) −0.807 (−1.58) −0.198 (−0.37)

−0.216 (−1.06) 0.196 (0.06)

−0.113 (−0.89) −0.250 (−1.39)

0.009 (0.06) −0.121 (−0.50)

Infrastructure spending Closed Autocracy*Infra. spending Electoral Autocracy*Infra. spending Electoral Democracy*Infra. spending 2014 and later Closed Autocracy*2014 and later Electoral Autocracy*2014 and later Electoral Democracy*2014 and later

SOE Subsample

Durable Closed Autocracy*Durable Electoral Autocracy*Durable Electoral Democracy*Durable Durable*2014 and later Closed Autocracy*2014 and later*Durable Electoral Autocracy*2014 and later*Durable Electoral Democracy*2014 and later*Durable Corruption Difference GDP per capita

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

−1.269 (−0.67) −0.034 (−0.02) −0.257 (−0.14) 0.610 (0.34) −0.020 (−0.69) 0.023 (0.72) 0.040 (1.07) 0.008 (0.26) 0.011 (0.22) −0.004 (−0.08) −0.006 (−0.10) −0.013 (−0.28) 0.023 (0.15) −0.027 (−0.11)

Political Regimes and BRI Country-Level Patterns

159

Table 6.5  (cont.) DV: Investment Full Sample 1

2

3

4

Y Y Y 240

0.072 (0.30) Y N Y 256

0.026 (0.15) Y N Y 207

0.155 (0.75) Y N Y 207

Distance Controls Country fixed effect Year fixed effect N

SOE Subsample

Note: *, **, and *** indicate significance at the 0.10, 0.05, and 0.01 levels, respectively. Standard errors are clustered at the country level. Controls include Market Size, Population Growth, Trade Openness, Inflation, Natural Resources, Business Freedom, Investment Freedom, and Rule of Law.

compared to Electoral Autocracy which displays a positive but weaker relation for two of the tests. Second, the 2014 and later variable is positive and significant for three of the four tests, however the interactions with the regime variables are not significant. Further analysis in Models (5) through (8) offer insight into what may account for these results. Models (5) through (8) add the Durable variable as well as double interactions with the political regime and 2014 and later variables, as well as triple interactions among all three. The results confirm the findings of the BRI Construction tests that electoral autocracies with less durable regimes attract a larger share of Chinese construction spending. Figure 6A.3 in the appendix graphically displays the interpretation of the triple interaction which clearly illustrates that electoral autocracies with low durability (regime durability equals five years) attract more than two times the volume of Chinese foreign construction spending than more durable counterparts (regime durability equals twenty years). The results are evident in tests with and without country fixed effects. Results for the triple interactions involving Electoral Democracy are similar to those of Electoral Autocracy, though the magnitudes of the coefficients are smaller for the former. Additionally, the findings for this longer sample do not confirm the positive significant finding for the Corruption Difference variable in the BRI Construction Spending tests from Table 6.2. Overall, the findings indicate there is a substantive and significant relationship between Chinese foreign construction spending and electoral autocracies with a low level of durability that occurs with the initiation of the BRI.

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

160

China’s Chance to Lead

Table 6.5 displays results for Chinese foreign investment spending for the 2005–2019 time period. Models (1) and (2) examine effects for the full sample with and without country fixed effects,. The results suggest a significant relation between infrastructure spending and each of the regime variables for the test without country fixed effects in Model (2). This finding also appears for Electoral Autocracy when restricting the sample to SOE investments as reported in Models (3) and (4). Notably, infrastructure spending displays a significant negative relationship with Chinese foreign investment in Model (3). A 1 percent of GDP increase in infrastructure spending on the part of the recipient country corresponds to a decline of nearly USD1 million in Chinese foreign investment, on average. Interactions between Infrastructure spending and Electoral Autocracy in Model (2) indicate that these regimes are less likely to attract Chinese investment in relation to liberal democracies if they have a low level of infrastructure spending. If they have a high level of infrastructure spending, then they are more likely to attract investment. Figure 6A.4 in the appendix illustrates that electoral autocracies that spend 10 percent of GDP on infrastructure will attract about three times more Chinese investment than electoral autocracies that spend 1 percent of GDP on infrastructure, on average. This result suggests Chinese private investors are more willing to invest in electoral autocracies when they see a substantial commitment on the part of Chinese SOEs to infrastructure development. Model (4) examines the effects for the 2014 and later period with respect to the Durability of the regime for the SOE subsample. The results do not yield any significant relationships. Overall, the findings offer tentative support for Chinese foreign investment increasing (typically by Chinese private firms) when there is a higher level of Chinese foreign construction spending (typically by SOEs). 6.4

Trade Effects

This section examines whether trade improvements are manifested for countries participating in the BRI. The argument and findings so far suggest electoral autocracies are the most likely to receive Chinese spending that would in turn promote trade. I begin with an examination of logistics performance measures across countries that span the 2007 to 2018 time period. The measures come from the World Bank Logistics Performance Index.11 The data are for 160 countries and are 11

See International LPI | Logistics Performance Index (worldbank.org) (accessed May 24 2022).

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

Political Regimes and BRI Country-Level Patterns

161

based on surveys conducted from 2007 to 2018, allowing for a useful comparison for pre- and post-2013 changes associated with the initiation of the BRI. Two measures are examined, including the infrastructure score and the tracking and tracing score. The infrastructure score offers a measure for the physical infrastructure associated with trade and transport. It is the most directly affected by BRI infrastructure development projects. The second measure is a country’s ability to track and trace consignments. This is more closely related to the Digital Silk Road, to be discussed in Chapter 9, but it offers helpful insight into whether countries’ trade capabilities improved as a result of the BRI. Tables 6.6 and 6.7 report the findings for infrastructure and tracking and tracing, respectively. I begin by taking the average score for each regime type for each year the survey was conducted. I then calculate the average score for each regime type for the 2007 to 2012 time period and the 2014 to 2018 time period. I then calculate the percentage change between these two periods. The results indicate electoral autocracies display the largest percentage improvement for both categories. Closed autocracies displayed the smallest improvement. Overall, the results suggest electoral autocracies have displayed the biggest improvement in the logistics indicators most closely associated with the BRI. To assess whether these improvements, coupled with the other infrastructure developments that countries have implemented as part of the BRI, yield trade gains, I examine changes in the share of Chinese exports going to BRI-participating economies, OECD countries, and the group of all other economies in Figure 6.7. Trade data come from the IMF

Table 6.6  Logistics performance index: Infrastructure score

Year

Closed Autocracy

Electoral Autocracy

Electoral Democracy

Liberal Democracy

2007 2010 2012 Pre-BRI: 2007–2012 avg 2014 2016 2018 BRI: 2014–2018 avg % Change Pre- and BRI

2.6 2.61 2.88 2.7 2.74 2.65 2.76 2.72 0.77

2.18 2.25 2.3 2.24 2.35 2.35 2.33 2.34 4.48

2.35 2.43 2.67 2.5 2.67 2.55 2.5 2.57 3.54

3.29 3.35 3.41 3.35 3.46 3.47 3.45 3.46 3.33

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

162

China’s Chance to Lead

Table 6.7  Logistics performance index: Tracking and tracing score

Year

Closed Autocracy

Electoral Autocracy

Electoral Democracy

Liberal Democracy

2007 2010 2012 Pre-BRI: 2007–2012 avg 2014 2016 2018 BRI: 2014–2018 avg % Change Pre- and BRI

2.73 2.64 2.89 2.75 2.78 2.72 2.87 2.79 1.34

2.35 2.6 2.49 2.48 2.56 2.48 2.57 2.54 2.29

2.55 2.8 2.78 2.71 2.85 2.7 2.7 2.75 1.41

3.39 3.55 3.46 3.47 3.46 3.56 3.54 3.52 1.61

100% 90% 22% 22% 23% 24% 24% 25% 26% 29% 32% 31% 31% 32% 32% 33% 35% 35% 35% 35% 35% 37% 80% 70% 60% 50%

63% 63% 62% 62% 61% 61% 60%

40%

58%

51% 49% 49% 50% 51% 53% 53% 57% 56% 56% 54% 51%

30% 20%

Other

OECD

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

16% 18% 16% 15% 14% 15% 14% 15% 14% 14% 14% 15% 13% 13% 12% 12% 11% 13% 14% 14% 2001

0%

2000

10%

BRI

Figure 6.7 Chinese exports to BRI-participating economies versus OECD countries, 2000–2019

Direction of Trade Statistics Database. Between 2014 and 2019, BRIparticipating countries’ share of Chinese exports increased by 4% (from 33% to 37%), OECD countries increased by 2% (from 39% to 41%), and other countries declined by 6% (from 18% to 12%). Although the share of exports to BRI countries increased, these data do not allow us to conclude the BRI contributed to these changes given that similar increases occurred during the period prior to the BRI. More time is

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

Political Regimes and BRI Country-Level Patterns

163

100% 90% 80%

36.1% 36.5% 35.5% 34.4% 34.2% 33.4%

32.8% 31.1% 31.7%

29.8% 29.8% 28.6%

26.1% 24.3% 24.1% 25.2% 25.8% 26.1% 26.5% 25.3%

70% 60% 50%

17.1% 17.4% 16.7%

18.2% 17.3% 16.8%

17.7%

36.6% 37.0% 38.2%

38.1% 39.0% 40.5% 40.7%

16.8% 17.0% 17.9% 19.8% 20.5%

20.1%

19.3% 18.3% 17.9% 17.7% 18.2%

18.6% 18.8%

46.1% 47.1% 46.1% 46.5% 46.7%

46.2% 46.6%

40% 30%

42.1% 40.7% 41.4% 40.8% 40.9% 43.3%

20% 10% 9.1% 9.2% 9.5% 9.7% 9.3% 9.2% 9.3% 9.9% 10.5% 10.7% 9.3% 9.7% 10.2% 10.1% 10.3% 10.7% 10.0% 8.8% 8.5% 9.2%

19

18

20

17

20

16

20

15

20

14

20

13

20

12

20

11

20

10

Closed Autocracy Electoral Democracy

20

09

20

08

20

07

20

06

20

05

20

04

20

03

Other

20

02

20

20

20

20

01

0.0% 0.0% 0.0% 0.0% 0.1% 0.1% 0.1% 0.2% 0.2% 0.2% 0.3% 0.2% 0.3% 0.3% 0.3% 0.2% 0.1% 0.1% 0.2% 0.1%

00

0%

Electoral Autocracy Liberal Democracy

Figure 6.8  Chinese exports to BRI-participating economies by political regime, 2000–2019

needed to identify whether the BRI has contributed to stable and substantive increases in Chinese trade exports. Figure 6.8 displays changes in Chinese exports to BRI-participating economies by political regime. Between 2014 and 2019, there is little change; however, the longer-term trend since 2010 clearly shows a notable increase of 5.8% (40.8% to 46.6%) to electoral autocracies, in comparison to electoral democracies which declined by 1% (19.8% to 18.8%) and liberal democracies which declined by 4.5% (29.8% to 25.3%), while closed autocracies remained nearly static (9.3% to 9.2%). Overall, the share of Chinese exports to electoral autocracies has increased substantially during the 2010–2019 time period, coinciding with policy initiatives that predated the formal launch of the BRI. But these conclusions must remain tentative since the BRI trends may simply be continuations of pre-BRI patterns.12 More time and data are needed to draw confident conclusions. 12 But even if the trends predate the BRI, the argument made in this book could still apply (i.e., the heightened role of clientelism and public control of the corporate sector in electoral autocracies elevates their demand for Chinese spending).

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

164

China’s Chance to Lead

6.5 Conclusions This chapter presents an analysis of the relationship between political regimes and Chinese foreign spending as part of the BRI and for total Chinese foreign spending between 2005 and 2019 (with a minimum foreign investment of USD100 million). This has involved grouping the data into several distinct categories. The first categorization regards distinguishing between foreign investment and construction spending. It is important to distinguish between these two types of foreign spending because they involve different commitments on the part of Chinese firms. Investment involves the purchase of an asset via mergers and acquisitions (M&As) or greenfield investments that can last for an indefinite period of time. Construction spending is for a predefined time period for work on a specific project. The second categorization concerns differences between non-BRI spending and BRI spending. BRI spending is more focused on infrastructure development and is expected to entail a larger share of construction spending. The findings are consistent with this expectation, with a far larger share of construction spending directed toward emerging economies in the context of BRI-affiliated projects; investments are more focused on advanced economies. The third categorization regards political regimes. The findings reveal clear differences, especially with regard to non-BRI versus BRI spending. With regard to non-BRI spending, liberal democracies are the main recipients, especially for Chinese foreign investments (m&a’s and greenfield investments). But with regard to BRI spending, electoral autocracies are, by far, the major recipients of Chinese foreign construction and investment spending. Electoral autocracies account for more than half of all BRI spending. They also receive a larger share of BRI spending on a country-year level relative to other political regimes. The clear importance of electoral autocracies as recipients of BRI spending represents the most important takeaway point of the chapter. After establishing the importance of electoral autocracies to Chinese foreign spending in the context of the BRI, I then examine the trade effects. While the logistics performance indicators suggest electoral autocracies display larger improvements than other regimes, it is too early to confidently conclude that this has yielded a larger share of Chinese exports. The data suggest this could be occurring, but more time is needed to draw confident inferences. The evidence presented in this chapter calls for more detailed examination of BRI projects in order to identify whether project characteristics display patterns consistent with recipient countries’ political regimes.

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

Political Regimes and BRI Country-Level Patterns

165

This requires analysis of a new dataset with more detailed data about these projects, which is the task of the next chapter.

Appendix

Descriptive Statistics for the Full BRI Sample (2005–2019)

Table 6A.1 reports Chinese foreign economic activity disaggregated at five-year intervals for investment and construction activity. Panel A reports the total of these two, in millions of US dollars. Column (1)

BRI Construction Spending (millions of USD)

300 250 200

Durable = 5 years Durable = 20 years

150 100 50

0 Liberal Democracy

Electoral Autocracy

Figure 6A.1  Two-way interaction for electoral autocracy and durability from Model (4) in Table 6.2

BRI Investment (millions of USD)

0.5 0.45 0.4 0.35

Durable = 5 years

0.3

Durable = 20 years

0.25 0.2 0.15 0.1 0.05 0 Liberal Democracy

Electoral Autocracy

Figure 6A.2  Two-way interaction for electoral autocracy and durability from Model (5) in Table 6.3

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

166

China’s Chance to Lead

SOE Subsample: Construction Spending

16 14 12

Durable = 20 years, 2014 and later

10

Durable = 20 years, before 2014

8

Durable = 5 years, 2014 and later

6

Durable = 5 years, before 2014

4 2

0 Liberal Democracy

Electoral Autocracy

Figure 6A.3 Triple interaction for electoral autocracy*durable*2014 and later from Model (7) in Table 6.4

SOE Subsample: Investment

0.9 0.8 0.7 0.6

Infra Spending = 1% of GDP

0.5 0.4

Infra Spending = 10% of GDP

0.3 0.2 0.1

0 Liberal Democracy

Electoral Autocracy

Figure 6A.4 Two-way interaction for electoral autocracy and infrastructure spending from Model (3) in Table 6.5

displays the total for the entire sample, spanning 2005 to 2019. Nearly 58 percent has gone to emerging economies, with over half of the total occurring between 2015 and 2019. During the entire sample period, electoral autocracies and liberal democracies have received a nearly equivalent amount despite the temporary spike in investment toward liberal democracies in 2016–2017; electoral democracies are a distant third and closed autocracies a very distant fourth.

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

60.1 37.6 22.4 2.0 12.4 8.8 36.6

1,225,610 767,480 458,130 41,570 253,930 179,460 746,010

815,330 92,380 722,950 140,170 434,800 185,740 53,690

Panel B. Investment All Economies Advanced Economies Emerging Economies Closed Autocracies Electoral Autocracies Electoral Democracies Liberal Democracies

Panel C. Construction All Economies Advanced Economies Emerging Economies Closed Autocracies Electoral Autocracies Electoral Democracies Liberal Democracies 39.9 4.5 35.4 6.9 21.3 9.1 2.6

100 42.1 57.9 7.9 36.7 16.8 38.2

2005–19 (%)

Panel A. Investment + Construction All Economies 2,040,940 Advanced Economies 859,860 Emerging Economies 1,181,080 Closed Autocracies 161,520 Electoral Autocracies 749,670 Electoral Democracies 343,850 Liberal Democracies 780,330

2005–19 (mill. USD)

121,920 10,460 111,460 38,950 53,560 23,690 5,720

174,600 95,440 79,160 5,210 56,160 17,940 95,190

296,520 105,900 190,620 43,800 113,080 38,630 100,910

2005–09 (mill. USD)

6.0 0.5 5.5 1.9 2.6 1.2 0.3

8.6 4.7 3.9 0.3 2.8 0.9 4.7

14.5 5.2 9.3 2.1 5.5 1.9 4.9

291,090 30,980 260,110 43,940 159,830 68,290 14,320

395,670 227,560 168,110 12,200 76,420 81,590 221,280

686,760 258,540 428,220 54,890 253,450 132,670 236,860

2005–09 2010–14 (mill. (%) USD)

Total, millions of USD

Table 6A.1  Chinese foreign investment and construction spending by political regime, 2005–2019

14.3 1.5 12.7 2.2 7.8 3.3 0.7

19.4 11.1 8.2 0.6 3.7 4.0 10.8

33.6 12.7 21.0 2.7 12.4 6.5 11.6

402,320 50,940 351,380 69,310 215,240 86,070 31,170

655,340 444,480 210,860 27,670 105,300 95,870 426,140

1,057,660 495,420 562,240 89,280 344,140 178,760 444,590

2010–14 2015–19 (mill. (%) USD)

19.7 2.5 17.2 3.4 10.5 4.2 1.5

32.1 21.8 10.3 1.4 5.2 4.7 20.9

51.8 24.3 27.5 4.4 16.9 8.8 21.8

2015–19 (%)

168

China’s Chance to Lead

Panels B and C disaggregate the data into investment and construction-related spending. Panel B displays investment, which involves the purchase and ownership of foreign assets such as a company or investment in a greenfield project. Advanced economies have been the primary recipients, accounting for nearly twice the volume going to emerging economies over the 2005–2019 time period. Examination of political regimes affirms the patterns displayed in Figure 6.5, with liberal democracies as the primary recipients of Chinese investments and electoral autocracies accounting for the largest share among the remaining political regime categories. Column (1) of Panel C shows that construction-related spending in emerging economies has been about seven times greater than in advanced economies over the 2005–2019 time period. Analysis by political regime reveals that electoral autocracies have received more than two times the amount of construction-related spending than the next highest political regime, electoral democracies, over the entire sample period. Closed autocracies are close behind but liberal democracies are far behind. The five-year time intervals illustrate the long-term trends in investment and construction spending. They confirm that increases in construction activity to emerging economies rose at a consistent pace, pre-dating the BRI. That also indicates that the volume of constructionrelated spending flowing to electoral autocracies has increased far more quickly than other political regimes. To assess the volume of Chinese foreign spending to individual countries, categorized by political regime, I examine country-year mean and median values. Because the country-year values do not display illuminating short-term fluctuations, I report only the long-term patterns in Table 6A.2. Over the entire 2005–2019 time period, the distribution of the data is clearly right skewed, as reflected by the min, max, mean, median, and standard deviation values. The five-year intervals consistently display similar right-skewed distributions. Examination of mean and median values by political regime reveals some illuminating patterns. For the entire sample period, electoral autocracies display the second highest mean value and the highest median value for investment plus construction values. With regard to investment, the regime patterns follow the expected distribution based on the preferences of private capital; that is, mean and median values are highest for liberal democracies and lowest for closed autocracies. For construction activity, the distribution runs in the opposite direction, which is consistent with expectations regarding SOEs’ predicted behavior. Thus, the high value for electoral autocracies with regard to the combined sum of investment and construction activity matches the theoretical

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

100 100 100 100 100 100 100

593 237 356 45 197 123 223

758 114 644 115 359 202 79

Panel B. Investment All Economies Advanced Economies Emerging Economies Closed Autocracies Electoral Autocracies Electoral Democracies Liberal Democracies

Panel C. Construction All Economies Advanced Economies Emerging Economies Closed Autocracies Electoral Autocracies Electoral Democracies Liberal Democracies 100 100 100 100 100 100 100

100 100 100 100 100 100 100

min

Panel A. Investment + Construction All Economies 1,057 Advanced Economies 289 Emerging Economies 768 Closed Autocracies 128 Electoral Autocracies 405 Electoral Democracies 260 Liberal Democracies 257

Obs

9,960 6,690 9,960 5,780 9,960 6,720 5,380

53,020 53,020 13,490 4,070 13,260 13,490 53,020

53,120 53,120 14,080 8,010 14,210 14,080 53,120

max

1,076 810 1,123 1,267 1,230 880 624

2,067 3,238 1,287 965 1,229 1,578 3,324

1,931 2,975 1,538 1,478 1,689 1,430 3,076

mean

2005–19

565 460 600 650 660 480 320

770 1,130 600 650 560 750 1,170

850 1,080 770 855 950 665 900

1,289 1,009 1,328 1,299 1,460 1,038 786

4,124 5,970 1,774 990 1,638 2,310 6,100

3,413 5,608 1,913 1,603 1,985 2,051 5,870

median Std Dev

847 805 851 1,034 806 798 572

1,695 2,512 1,218 669 1,697 776 2,505

1,366 2,302 1,115 1,047 1,276 960 2,293

mean

470 500 440 640 460 415 210

740 985 550 740 970 360 770

650 750 610 670 610 640 720

1,018 756 1,062 1,098 1,171 943 465

1,807 2,709 1,245 1,303 931 1,793 2,833

1,743 2,510 1,472 1,388 1,515 1,523 2,514

560 320 600 600 660 575 300

740 1035 500 880 460 970 1,110

830 1,100 770 865 850 720 880

1,227 849 1,311 1,606 1,475 861 785

2,418 3,865 1,352 904 1,355 1,757 3,973

2,371 3,539 1,837 1,907 2,099 1,538 3,810

650 520 675 1,140 875 445 370

840 1,330 610 610 610 785 1,340

1,015 1,170 985 1,180 1,170 650 1,250

median

2015–19

median mean

2010–14

median mean

2005–09

By Country-year, millions of USD

Table 6A.2  Chinese foreign investment and construction spending by country-year and political regime, 2005–2019

170

China’s Chance to Lead

Table 6A.3  Top two sectors for Chinese foreign investment and construction spending, 2005–2019 1st and 2nd sectors, 2005–19

1st and 2nd sectors, 2015–19

Panel A. Investment + Construction All Economies Energy and Transport Advanced Economies Transport and Real Estate Emerging Economies Energy and Transport

Energy and Transport Transport and Real Estate Energy and Transport

Closed Autocracies Electoral Autocracies Electoral Democracies Liberal Democracies

Energy and Transport Energy and Transport Energy and Transport Energy and Transport

Energy and Real Estate Energy and Transport Energy and Transport Transport and Real Estate

Panel B. Investment All Economies Advanced Economies Emerging Economies

Energy and Real Estate Real Estate Energy Energy and Metals

Energy and Real Estate Real Estate and Transport Energy and Metals

Closed Autocracies Electoral Autocracies Electoral Democracies Liberal Democracies

Energy and Other* Energy and Metals Energy and Metals Real Estate and Energy

Energy and Other* Energy and Real Estate Energy and Transport Real Estate and Transport

Panel C. Construction All Economies Advanced Economies Emerging Economies

Energy and Transport Transport and Energy Energy and Transport

Energy and Transport Transport and Real Estate Energy and Transport

Closed Autocracies Electoral Autocracies Electoral Democracies Liberal Democracies

Energy and Transport Energy and Transport Energy and Transport Transport and Energy

Energy and Real Estate Energy and Transport Energy and Transport Transport and Real Estate

* The leading other sector for investment is appliances (Scissors 2017)

expectations that spending by Chinese private firms and SOEs will converge on this regime type. Table 6A.3 displays the sectors with the highest and second-highest levels of investment and construction activity following the classification used by the CGIT dataset. Energy is the most common sector across all categories, followed by transport for the entire sample period and for the 2015–19 time period. Both energy and transport are dominated by SOEs, as reported in Table 5.6 in Chapter 5. Real estate is a distant third, followed by metals (for investment only). Energyrelated construction is topped by hydropower with coal also substantial

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

Political Regimes and BRI Country-Level Patterns

171

(Scissors 2020). Energy investments involve energy extraction assets, topped by oil. Transport construction is focused mainly on roads and rail, while transport investment features the acquisition and building of new auto plants. Real estate investment is understated due to home purchases of less than $100 million; property construction includes commercial buildings as well as low-income housing. Finally, metals investments are focused on bauxite, copper, and iron mining and refining, reinforcing the continued centrality of commodities serving Chinese industry.

Troubled Transactions and Greenfield Investments

The CGIT database provides data for two additional categories of Chinese foreign investment and construction spending of interest that can be analyzed by regime type, including troubled transactions and greenfield investments. Troubled transactions refer to projects in which investment or construction is impaired or failed after a commercial agreement was finalized (Scissors 2018). Greenfield investments involve the creation of a subsidiary in a foreign country, with operations built from the ground up. Greenfield investments provide the highest degree of control for the sponsoring company, but also entail high risk due to the substantial investment. Insofar as political support is a key, if not the primary, determinant for a project’s success in emerging economies (Osei-Kyei and Chan 2017), Electoral Autocracies are likely to impact the prevalence of troubled transactions in ways consistent with the argument developed in this chapter. Specifically, the same electoral pressures that push electoral autocrats to attract Chinese foreign spending will also push them to successfully complete projects so as to promote economic growth and woo more voters to support them relative to other political regimes. With regard to greenfield investments, the substantial risk that firms bear when engaging in these types of investments suggests that Chinese SOEs will be more capable of using this form of investment in emerging economies, and especially in the context of autocracies, due to their privileged access to financing and because government-to-government relations are more likely to influence SOE investments. Further, emerging economies possess fewer attractive M&A targets for private firms, leading to a higher share of greenfield investments in these economies by default. Results reported in Table 6A.4 offer evidence generally consistent with both of these conjectures. The table reports total numbers rather than total values since these offer a more representative picture of each

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

172

China’s Chance to Lead

Table 6A.4  Troubled transactions and greenfield investments Troubled Transactions Panel A. General Overview Total Value Troubled Transactions (mill. USD) Total Value Investments and Construction, 2005–19 (mill. USD) Total Value Troubled Transactions/ Total Value Investments and Construction Projects Total Number Troubled Transactions Projects Total number investments and Construction Projects, 2005–19 Total Number Troubled Transactions Projects/Total Number Investments and Construction Projects Total Number SOE Troubled Transactions Projects/Total Number Investments and Construction Projects Total Number SOE Troubled Transactions Projects/Total Number SOE Investments and Construction Projects Total number BRI Troubled Transactions Projects/Total Number Investments and Construction Projects Total Number BRI Troubled Transactions Projects/Total Number BRI Investments and Construction Projects Panel B. By Political Regime Total Number Troubled Transactions in Closed Autocracies/Total Number Investments and Construction Projects in Closed Autocracies Total Number Troubled Transactions in Electoral Autocracies/Total Number Investments and Construction Projects in Electoral Autocracies

Greenfield Investments 391,810 2,040,400

Total Value Greenfield (mill. USD) Total Value Investments, 2005–19 (mill. USD)

409,640 1,225,610

0.192

Total Value Greenfield/ 0.334 Total Value Investments

266

Total Number Greenfield 464 Projects Total Number invest1166 ments, 2005–19 Total Number Greenfield 0.398 Projects/Total Number Investments

1980 0.134

0.10

0.140

0.05

0.080

0.121

0.096

Total Number SOE Greenfield Projects/ Total Number Investments Total Number SOE Greenfield Projects/ Total Number SOE Investments Total Number BRI Greenfield Projects/ Total Number Investments Total Number BRI Greenfield Projects/ Total Number BRI Investments

0.25

0.438

0.26

0.570

Total Number Greenfield 0.747 in Closed Autocracies/ Total Number Investments in Closed Autocracies Total Number Greenfield 0.659 in Electoral Autocracies/ Total Number Investments in Electoral Autocracies

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

Political Regimes and BRI Country-Level Patterns

173

Table 6A.4  (cont.) Troubled Transactions

Greenfield Investments

Total Number Troubled 0.103 Transactions in Electoral Democracies/Total Number Investments and Construction Projects in Electoral Democracies Total Number Troubled 0.198 Transactions in Liberal Democracies/Total Number Investments and Construction Projects in Liberal Democracies

Total Number 0.534 Greenfield in Electoral Democracies/Total Number Investments in Electoral Democracies Total Number Greenfield 0.185 in Liberal Democracies/ Total Number Investments in Liberal Democracies

category.13 With regard to troubled transactions, the results reported in Panel A indicate they constitute 13.4 percent of all investments and construction projects. Panel B reports their prevalence across political regimes, which indicates they are least common among electoral autocracies at 9.6 percent, followed by electoral democracies at 10.3 percent and then closed autocracies at 12.1 percent, and they are most common among liberal democracies at 19.8 percent. With regard to greenfield investments, Panel A reports that they represent 43.8 percent of all investments. As a proportion of all investments by each regime category, they are most common in closed autocracies at 74.7 percent, followed by electoral autocracies at 65.9 percent, then electoral democracies at 53.4 percent, and followed by liberal democracies at 18.5 percent. The prevalence of greenfield investments in emerging economies is largely due to their use by SOEs, which are in turn widely used in the context of BRI projects (untabulated results for the distribution of greenfield investments by SOEs and in the context of the BRI across political regimes mirror the general political regime results reported in Panel B).

13

For example, a small number of highly valued projects will bias the results in favor of those particular projects in the same way that means create biases, which motivates the reporting of median values. The equivalent to median values in this context is the number of projects.

https://doi.org/10.1017/9781009385879.006 Published online by Cambridge University Press

7

Political Regimes and BRI Project Characteristics

The previous chapter examined Chinese foreign investment and ­construction activities to both advanced and emerging economies, both before and after the BRI was launched in the autumn of 2013. The evidence reveals significant new BRI-related investments during the 2014 and later period. The effects are observed to varying degrees across political regimes, but the most intriguing patterns are evident among electoral autocracies. This chapter turns to a detailed examination of BRI projects. By delivering new, large and multiyear commitments (often lasting twenty-five years or more when structured as a public-private partnership or PPPs), these projects can be a substantial boon to host country leaders by creating new patronage opportunities and economic growth. Consequently, these projects can help host country leaders retain power. The question of this chapter is whether characteristics of BRI projects display patterns corresponding to a host country’s political regime. The prevailing counterargument is that BRI projects reflect primarily what China wants. Recipient or host countries are portrayed as the backdrop against which large projects are implemented, with national elites uncritically attracted by the availability of financing and the developmental potential of these initiatives. Consequently, much of the extant literature has focused on the goals of PRC-related actors at the expense of those in host countries, as discussed in Chapter 1. This framing ignores the interests and agency of local actors, who may support, co-opt, or subvert large-scale projects for their own ends. The theory developed in Chapter 4 argues that electoral autocracies will be the main recipients of BRI spending due to leaders’ incentives to attract funds for their clientelist networks combined with their control of state mechanisms (e.g., SOEs) to administer the targeted distribution of these resources. Electoral autocrats with an insecure hold on power are posited to be the most avid recipients of Chinese spending. In this chapter, I delve into the implications of this argument at 174

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

Political Regimes and BRI Project Characteristics

175

the project level in order to identify whether BRI projects in electoral autocracies display distinguishable characteristics from those located in other regimes. The evidence presented in this chapter is based on a different ­dataset from the prior chapter and includes over 2,100 projects up through 2019 (including those started before the BRI but later classified as BRI projects) across 127 countries. Whereas the prior chapter focused on differences between foreign direct investment and construction spending across countries with a value of USD100 million or more, this chapter narrows the focus to BRI spending exclusively for projects with large and small price tags. For each project, ten characteristics were manually coded, including the date of the project’s initiation, whether the project used a PPP structure, whether the owner is a host country SOE and/ or a Chinese SOE, whether a Chinese SOE provides financing and/or whether a Chinese SOE is a shareholder, contractor, developer, and/or consultant, and whether the project has been successfully completed. The findings both corroborate the importance of electoral autocracies and provide more detailed evidence for systematic differences in the structure of BRI projects depending on the political regime in which they are located. The chapter begins with a brief example, the Kenya Standard Gauge Railway Project, to illustrate how project characteristics can be affected by political priorities in the context of an electoral autocracy. The subsequent section generalizes the insights from this example and draws implications for BRI project characteristics based on the theory developed in Chapter 4, including those that differentiate between types of political regimes as well as differences between strong and weak rulers of electoral autocracies. I then turn to a rigorous assessment of BRI project characteristics. The analysis demonstrates clear and significant p ­ atterns between BRI project characteristics and political regimes as well as for strong versus weak electoral autocrats that are consistent with and extend the findings of the previous chapter. 7.1

An Illustrative Example: The Kenya Standard Gauge Railway Project

During the 2013 campaign for president of Kenya, an electoral autocracy, the development of a modern rail network linking the port city of Mombasa to the nation’s largest city Nairobi and other inland destinations became a key campaign message of presidential candidate Kenyatta, the ultimate victor. Originally conceived as a public-private project in collaboration with a Kenyan oligarch, the Kenya Standard Gauge Railway

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

176

China’s Chance to Lead

(SGR) project became a central government project once Kenyatta won the election on March 4. In August 2013, the World Bank published its evaluation of four alternatives for building the railway, including: (1) refurbish the existing meter-gauge network (at a cost of USD0.18 million/km); (2) upgrade the existing network to a higher standard using the same gauge (USD0.49); (3) upgrade the existing network to a standard gauge system on the same network (USD1.50); (4) construct a SGR on a new line (USD3.25). The World Bank recommended the second option as the most cost-effective and one which could easily accommodate forecasted freight traffic up to 2030.1 Despite this recommendation, the Kenyatta government chose the most expensive option, making the SGR the highest priced infrastructure project in the country since independence in 1963. The SGR financing agreement of USD3.8 billion was signed in May 2014 during a visit to Kenya of China’s Premier Li Keqiang. The project would be jointly funded by the Kenyan government (15%) and the Export-Import Bank of China (85%). Although Kenyan law required all government-funded investments to be awarded by public bid, the Public Procurement Act allowed awards without competitive bidding for government-to-government agreements. Chinese Ex-Im Bank financing was offered only if the key construction contracts were given to Chinese firms. As a result, the state-owned China Road and Bridges Corporation (CRBC) was selected as the EPC (engineer, procurer, and contractor). To assure reliability of operations, the SGR would be operated by the China Communications Construction Company (a SOE) for a minimum of five years. The Kenyan sponsor of the project would be the Kenya Railway Corporation (KRC) (a SOE). Based on over 100 interviews with Kenyan and Chinese participants in the SGR, Yuan and Uwe (2019) show that clientelism was a critical feature affecting the project’s construction. Although railway equipment, steel for rails, construction machinery, and locomotives were imported from China, 40% local input was required by the Kenyatta government. More than 45,000 Kenyans were employed both temporarily and permanently in the SGR’s first phase of construction, from the coastal city of Mombasa to the country’s capital and largest city Nairobi. Hundreds of locally owned and operated Kenyan businesses supplied materials and basic services to construction; Kenyan cement makers, logistics companies (clearing companies, container freight stations, and transport companies), and manufacturers of rail cars were 1

“The Economics of Rail Gauge in the East African Community,” The World Bank – Africa Transport Unit, August 8, 2013.

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

Political Regimes and BRI Project Characteristics

177

the biggest winners. Typically, politically connected oligarchs were the main beneficiaries, but subcontracting opportunities also flowed to smaller businesses, especially in the logistics industry. Individuals and businesses who were not aligned with the ruling party were excluded from the opportunities and benefits that flowed from the project. Accusations of corruption and ­kickbacks were rampant, driven in part by the opacity of the project’s costs. Both Kenyan and Chinese interviewees repeatedly emphasized that the project is “a Kenyan railway” with technical and financial support from China (Yuan and Uwe 2019). In other words, KRC was the employer responsible for allocating contracts to local firms while CRBC was the contractor. Controversies during construction were typically caused by Kenyan domestic politics and businesses, with the resolution of such disputes handled by the relevant bodies of the Kenyan government. The CRBC’s role was confined to the technical aspects of railway construction. President Kenyatta was eager to finish the project before the 2017 election. Although SGR construction was scheduled to take five years, it was completed within half that time. On May 30, 2017, two months before Kenya’s presidential election, President Kenyatta inaugurated the new railway as the “Madaraka Express.” The railway’s first passenger service commenced on 1 June, “Madaraka Day,” Kenya’s national holiday commemorating independence from the British in 1963. I contend this case is not unique. Rather, it illustrates how project characteristics are affected by the broader political system in which they are embedded. The form and magnitude of how these characteristics are manifested varies by political regime, which I discuss in the next section. 7.2

Extending the Political Regimes Theory to BRI Project Characteristics

As argued in Chapter 4, electoral autocracies are posited to display the highest compatibility with Chinese BRI spending. In contrast to rulers of closed autocracies, electoral autocrats must contend with political opponents in semi-competitive elections, forcing them to expand their clientelist network. And in contrast to electoral democracies, clientelist networks in electoral autocracies are state-led meaning that state-owned banks control access to financing and SOEs typically control the allocation of subcontracts for large construction projects. Consequently, electoral autocracies and China have multiple shared interests in how BRI projects are implemented, including speed, opacity, a high volume of spending, greater weight to political priorities in relation to firm

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

178

China’s Chance to Lead

profitability, and relative inattention to socio-environmental impact (as displayed in Table 4.2). The power of recipient governments to influence BRI projects is amplified by China’s official policy of noninterference in the domestic affairs of foreign countries. This argument generates a series of expectations regarding the characteristics of BRI projects. First, I expect rulers of electoral autocracies to attract more BRI projects than other regimes, both in total and on a per country basis due to the convergence of interests. In other words, I expect more BRI projects to be initiated in electoral autocracies than in other regimes. This prediction was confirmed in the prior chapter with the CGIT database. This chapter will use a new database regarding individual BRI projects to reassess this expectation. Second, I expect a larger share of projects to be completed by electoral autocracies than other regimes. Because BRI projects are primarily SOEled, state-to-state cooperation is critical as electoral autocrats want these projects for both patronage and growth reasons but need to maintain opacity and retain residual control rights. The political incentives of electoral autocrats to initiate projects also lead them to favor their completion to fulfill the output legitimacy on which their rule in power depends. Their control over SOEs grants them the power to allocate resources and overcome obstacles that could slow or imperil the speedy completion of the project, typically in time for a national election as seen in the Kenya SGR case. Third, I expect SOEs from electoral autocracies to be more frequently involved as owners than SOEs in electoral or liberal democracies due to the incentive of electoral autocrats to initiate projects for electoral/ patronage needs and their greater control of state resources to divert resources to their clientelist networks. The Kenyan government, for example, took control of the SGR project away from an oligarch shortly following Kenyatta’s election win. Additionally, the higher concentration of SOEs in industries related to BRI projects for closed and electoral autocracies as shown in Table 5.6 suggests a higher level of collaboration between SOEs from both China and electoral autocracies. In relation to closed autocracies, it is uncertain whether the reliance on SOEs would be higher or lower (as a proportion of all projects in a country). On the one hand, electoral autocrats may rely more on SOEs because of election pressures to boost patronage opportunities. On the other hand, the greater dominance of SOEs in closed autocracies may produce a relatively higher reliance on them than in electoral autocracies. This is an empirical question to resolve. Fourth, I expect Chinese SOEs to be more frequently involved in BRI projects as owners and/or shareholders for projects located in electoral

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

Political Regimes and BRI Project Characteristics

179

autocracies than in electoral or liberal democracies due to: (1) their ability to accommodate patronage distribution needs on the part of host country rulers without the need to disclose information about project financing or other related details; (2) the pressure from host country private business workers to reduce crowding out effects associated with Chinese SOEled projects being relatively more muted in electoral autocracies due, in part, to media controls and other mechanisms to stifle criticism of government activities; and (3) electoral autocrats being more likely to engage in government-to-government negotiation and cooperation to initiate and see projects to completion due to electoral pressures. The participation of Chinese SOEs is valued in this regard as they will more faithfully uphold and implement agreements between government leaders. In closed autocracies, Chinese SOEs as owners/shareholders will be lower than electoral autocracies for two main reasons. First, rulers of closed autocracies do not have the same election-motivated urgency to seek Chinese state collaboration in implementing projects. Second, the attraction of closed autocracies to Chinese private firms is lower than electoral autocracies, which will in turn reduce the participation of Chinese SOEs in BRI projects located in closed autocracies to the extent such projects depend on Chinese private participation (either concurrently as part of the project’s construction or after the completion of the construction phase, as with industrial parks). Thus, I expect China SOE ownership will be more prevalent in electoral autocracies than in other regimes. Fifth, the same incentives that push rulers of electoral autocracies to seek BRI projects and include China SOEs as partial owners will also motivate them to rely on Chinese SOEs for their greater capabilities to complete projects. The economies of scale and scope of Chinese SOEs, the depth of funding available, and greater experience implementing and managing large infrastructure projects in relation to domestic firms coupled with electoral autocrats’ incentive to complete projects in a timely manner to bolster their political support will enhance the desire to rely on Chinese SOEs for related project responsibilities, such as financier, contractor, developer, and consultant as occurred with the Kenya SGR project. Private firms in electoral and liberal democracies will be more resistant to relying on Chinese SOEs for crowding out reasons; rulers of closed autocracies do not face the same election and patronage incentives to ensure successful completion of the project, and so will not display the same reliance on Chinese SOEs for affiliated services. Sixth, I expect electoral autocracies to host more public-private partnerships (PPPs) with Chinese SOEs than other regimes for two reasons. The first reason relates to the attractiveness of the economy for private firms. Foreign private firms will be more reluctant to invest in closed

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

180

China’s Chance to Lead

autocracies than electoral autocracies, and domestic private firms will be relatively fewer in number in closed autocracies due to crowding out effects, as shown in Chapter 5. This reduces the attractiveness of closed autocracies for Chinese SOEs that require collaboration with private firms. While foreign private firms will favor electoral democracies, and domestic private firms will be more plentiful, Chinese SOEs will disfavor projects in this type of regime due to the decline in their residual control rights. Thus, electoral autocracies offer the most favorable setting for PPPs for SOE-led projects that include private sector involvement. The second reason concerns the heightened political incentives of electoral autocrats to attract Chinese SOEs as the “private” partner in a PPP financing arrangement. Chinese SOEs are attractive “private” partners in PPPs because of their access to financing and the absence of conditionalities placed on the host government. Indeed, electoral autocrats can use PPPs as a mechanism to leverage their own limited financial resources to fund more projects, and autocratic rulers can overcome political opposition more easily (e.g., to excessive government debt) than their counterparts in electoral democracies. 7.2.1

Weak Electoral Autocrats and BRI Project Characteristics

The weakness of political rulers in electoral autocracies will also impact various characteristics of BRI projects. The most immediate effect regards the use of PPPs. These offer a mechanism for the government to increase the number of infrastructure projects by shifting a portion of the financing burden onto another party. As the name PPP suggests, that party is conventionally a private firm, but it may also be a foreign SOE. While weak rulers will face pressures to increase their use of PPPs, partnering with Chinese SOEs is particularly attractive because of their relatively greater opacity, which enables host country rulers to use infrastructure projects as a patronage opportunity. To ensure control over the distribution of patronage, weak rulers will assert greater control over BRI projects through ownership by host country state-owned enterprises. This will facilitate how subcontracts and related business opportunities are allocated so as to reward loyal supporters and also to attract new ones. While majority control of a BRI project by host country SOEs is expected, Chinese SOE minority ownership as a shareholder in BRI projects is expected to increase as the total number of BRI projects increases. Weak rulers will be the most avid recipients of Chinese foreign spending, corresponding to higher levels of Chinese SOE shareholding. At the same time, SOEs will be more capable of successfully pushing large, capital-intensive infrastructure projects to completion by calling

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

Political Regimes and BRI Project Characteristics

181

upon state resources and forms of state assistance (expedited regulatory approvals, land acquisition, etc.). The relatively high concentration of SOEs in the industries commonly involved in infrastructure projects, as displayed in Table 5.6, reflects their compatibility with the political purposes that BRI projects may be used for. In an effort to maximize patronage resources, weak rulers will also turn to Chinese SOEs for financing of BRI projects. This has generated controversy in the literature regarding the emergence of “debt traps” (Singh 2020). Because BRI projects are typically initiated by the host country, leaders of these regimes are more likely to sacrifice the longterm economic health of the country in order to preserve their own hold on power. This trade-off is most likely to occur when political incumbents are weak. To boost support both among their clientelist network and to attract new supporters, weak political incumbents have strong incentives to ensure the successful completion of BRI projects. Indeed, a major criticism that opponents levy at political incumbents is corruption and the diversion of resources away from productive uses, contributing in turn to projects that fail to be completed. Weak rulers can benefit by countering these criticisms with projects that are successfully completed. Finally, many infrastructure projects with Chinese SOE participation are very large, and beyond the capabilities of host country firms. For this reason, Chinese SOEs may be tapped as consultants, developers, and/or contractors. This is especially likely for weak rulers who are looking to successfully complete infrastructure projects. 7.3

Patterns and Trends

7.3.1

Construction of the Dataset

The Refinitiv dataset offers numerous details about individual projects and independent corroboration for some of the patterns identified with the CGIT dataset in Chapter 6. Refinitiv’s BRI database includes projects that have been identified as BRI-affiliated by the Chinese government or Chinese state departments; where project contractors have signed agreements with the government departments of the relevant country along BRI corridors and have obtained the relevant regulators approvals required for Belt and Road initiatives or which are published on the BRI official website,2 or projects that have direct Chinese participation as a consultant, owner, contractor and financier, or are of strategic interest 2

See https://eng.yidaiyilu.gov.cn/zchj/qwfb/86739.htm (accessed July 6, 2020).

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

182

China’s Chance to Lead

located along a BRI economic corridor. Projects in mainland China, Hong Kong, and Macau are excluded, producing a list of 2,478 projects.3 About ten characteristics were manually coded for each project. After cleaning of the dataset, a total of 2,122 projects were included in the final sample spanning 127 countries. The first project included in the BRI sample was initiated in 1972; the most recent is December 2019, just prior to the onset of COVID-19 which contributed to a temporary alteration in historical patterns and thus provides an appropriate time point for the end of the sample period. The comprehensive coverage of the Refinitiv dataset includes both large and small projects as well as numerous characteristics associated with each project. The values of the projects are not independently corroborated as with the CGIT dataset from Chapter 6, and may therefore be less accurate. I therefore choose to focus on the individual characteristics of each project. These include the use of a PPP structure, whether the owner is a host country SOE and/or a Chinese SOE, whether a Chinese SOE provides financing and/ or whether a Chinese SOE is a shareholder, contractor, developer, and/ or consultant, and whether the project has been successfully completed. 7.3.2

Findings across Political Regimes

Table 7.1 displays the distribution of projects across time and across political regimes. Advanced economies account for 7.8% of total p ­ rojects; emerging  economies for 92.2%. This indicates a higher proportion ­ for emerging economies than the CGIT dataset, which reported 80% for emerging economies (330 out of 413 observations). This may reflect the more comprehensive coverage of the Refinitiv database which includes many smaller value projects not included in the CGIT data. Notably, a large proportion (48%) of projects affiliated with the BRI are from the 2005–2014 time period which is consistent with other accounts that include projects begun before the initiative was launched, presumably as part of an effort to include them within broader multilateral coordination initiatives introduced by the NDRC’s Vision and Actions Statement in 2015. The distribution of projects across political regimes indicates that over half of all BRI projects are located in electoral autocracies, 26% in electoral democracies, followed by closed autocracies at 14%, with liberal democracies accounting for only 5%. The proportion of projects initiated with respect to each regime type increases over time, with electoral autocracies displaying the largest increase from before 2010 to after (3.9% to 22% of all projects), and 27% of all projects are located in 3

This is based on the sample provided by Refinitiv on July 6, 2020.

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

2,122 166 1,956

305 1,120 564 116

All Economies Advanced Economies Emerging Economies

Closed Autocracies Electoral Autocracies Electoral Democracies Liberal Democracies

Number of Projects Initiated

14 52 26 5

100 7.8 92.2

Percentage of Total Projects

Table 7.1  BRI projects initiated and completed by political regime

1.9 3.9 1.7 0.4

8 0.5 7.5

Percentage of Total Projects before 2010

6.6 22 9.5 1.6

40 3.3 36.7

Percentage of Total Projects, 2010–2014

5.8 27 15 3

52 4 48

Percentage of Total Projects, 2015–2019

15.2 22.2 15.6 0.8

19.3 14.6 20.5

Percentage of Projects Completed of those Initiated by category

184

China’s Chance to Lead

these regimes between 2015 and 2019. A comparison of the number of projects in electoral autocracies (203) versus electoral democracies (85) indicates BRI projects for the Refinitiv dataset is roughly comparable to the 173 observations for electoral autocracies and 90 observations for electoral democracies from the CGIT dataset used in Chapter 6. These data indicate a slightly higher proportion in electoral democracies for the CGIT dataset, which may be due to the size restrictions on projects included in that database (at least USD100 million). The argument posits that project completion rates will be higher for electoral autocracies than other regimes due, in part, to the incentives of political incumbents facing political opponents in semi-competitive elections. The data indicate a far higher completion rate of projects initiated in electoral autocracies as a proportion of total projects initiated (22.4%) in relation to closed autocracies and electoral democracies (15–16%). Projects in liberal democracies display a very low completion rate (0.8%). This is consistent with the findings from the troubled transactions sample of the CGIT dataset, as reported in Table 6A.1, which indicates electoral autocracies have the lowest proportion of troubled transactions (9.6%); closed autocracies and electoral democracies are in the middle (12.1% and 10.2%, respectively), and liberal democracies have the highest proportion (19.8%). Additionally, untabulated results for the Refinitiv dataset indicate the first and second most common sectors for projects to be initiated are consistent with those identified by the GCIT sample: transportation and energy, with real estate being relatively more common among liberal democracies. The argument also posits that host country SOEs from autocratic regimes will be more involved in BRI projects than those located in democracies. Between closed and electoral autocracies, arguments could be made for why SOE involvement is greater for either regime type, thus making this an empirical question to answer. Table 7.2 indicates nearly 70% of all projects have a host country SOE as an owner; those without a SOE owner may either be private owners or a government agency. This is a remarkably high level of SOE ownership; also remarkable is the importance of electoral autocracies to this phenomenon in relation to other regimes. Electoral autocracies in which a host country SOE is an owner of a BRI project accounts for nearly 39% of all projects; electoral democracies are a distant second at 19%. The data further reveal that this phenomenon has primarily occurred in the 2010–2019 time period, with greater weight on the 2015–2019 timeframe. A sizeable increase also occurs for electoral democracies from before to after 2015. Overall, these results are consistent with the expectation that host country SOEs from electoral autocracies will host a larger share of BRI projects than

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

69.7 5.8 28.9 35.0 7.6 0.9 3.7 3.1 25.0 2.4 10.6 12.1 27.9 1.7 13.3 13.0 42.9 2.3 16.5 24.1 5.8 0.3

Owner China SOE, all years to 2019 Owner China SOE, before 2010 Owner China SOE, 2010–2014 Owner China SOE, 2015–2019

Shareholder China SOE, all years to 2019 Shareholder China SOE, before 2010 Shareholder China SOE, 2010–2014 Shareholder China SOE, 2015–2019

Financing Source China SOE, all years to 2019 Financing Source China SOE, before 2010 Financing Source China SOE, 2010–2014 Financing Source China SOE, 2015–2019

Contractor China SOE, all years to 2019 Contractor China SOE, before 2010 Contractor China SOE, 2010–2014 Contractor China SOE, 2015–2019

Developer China SOE, all years to 2019 Developer China SOE, before 2010

All Econs

Owner Host Country SOE, all years to 2019 Owner Host Country SOE, before 2010 Owner Host Country SOE, 2010–2014 Owner Host Country SOE, 2015–2019

Table 7.2  BRI project characteristics by political regime

0.1 0.1

3.4 0.1 1.4 1.9

1.4 0.1 0.6 0.7

1.6 0.2 0.3 1.0

0.3 0.0 0.1 0.2

6.2 0.4 2.8 3.0

Adv. Econs.

5.6 0.3

39.5 2.3 15.0 22.2

26.5 1.5 12.7 12.3

23.5 2.2 10.2 11.0

7.3 0.9 3.6 2.9

63.5 5.5 26.1 32.0

Emerg. Econs

0.7*** 0.1

5.3*** 0.3* 1.7*** 3.3***

2.7*** 0.4* 1.4*** 1.0***

4.1*** 1.0*** 1.7*** 1.4**

0.8* 0.4*** 0.2 0.2

7.6*** 1.2*** 3.8*** 2.6

Closed Auts

4.2*** 0.2**

24.2*** 1.4*** 10.2*** 12.5***

16.4*** 0.9*** 8.2*** 7.3***

14.0*** 1.2*** 6.3*** 6.6***

4.3*** 0.5*** 2.3*** 1.6***

38.7*** 3.2*** 16.7*** 18.8***

Electoral Auts

(% of all projects)

0.5** 0.1

10.9*** 0.6*** 3.3*** 7.1***

8.0*** 0.2 3.5*** 4.3***

5.8*** 0.1 2.1*** 3.5***

1.9*** 0.0 1.0*** 0.9***

19.0*** 1.1*** 6.5*** 11.4***

Electoral Dems

0.1 0.0

2.0 0.1 0.8 1.2

0.6 0.1 0.2 0.3

1.2 0.1 0.4 0.6

0.4 0.0 0.1 0.3

3.8 0.3 1.4 2.1

Liberal Dems

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

7.0 0.3 2.9 3.8 28.0 1.9 16.3 9.8

Consultant China SOE, all years to 2019 Consultant China SOE, before 2010 Consultant China SOE, 2010–2014 Consultant China SOE, 2015–2019

PPPs, all years up to 2019 PPPs before 2010 PPPs, 2010–2014 PPPs, 2015–2019

2.5 0.1 1.6 0.9

0.5 0.0 0.2 0.3

0.1 0.1

Adv. Econs.

25.5 1.8 14.8 8.9

6.5 0.3 2.7 3.5

2.9 2.5

Emerg. Econs

3.6*** 0.9*** 2.0*** 0.7

0.9*** 0.1* 0.2 0.5**

0.4** 0.2

Closed Auts

15.2*** 0.8*** 9.1*** 5.3***

4.2*** 0.1* 2.1*** 1.9***

2.2*** 1.8***

Electoral Auts

(% of all projects)

7.2*** 0.2 3.9*** 3.1***

1.7*** 0.0 0.4** 1.2***

0.3* 0.2

Electoral Dems

1.4 0.1 0.9 0.5

0.2 0.0 0.1 0.1

0.1 0.1

Liberal Dems

Note: statistical significance is reported for the difference between the proportion of projects (out of the total number of observations) for a particular regime in relation to that reported for liberal democracies, with *** (z-statistic < 1%), ** (z-statistic < 5%), and * (z-statistic < 10%).

2.9 2.5

All Econs

Developer China SOE, 2010–2014 Developer China SOE, 2015–2019

Table 7.2  (cont.)

Political Regimes and BRI Project Characteristics

187

host country SOEs from other regimes, in relation to total BRI projects across all regimes (as opposed to the proportion of BRI projects by country-regime type). The argument further predicts Chinese SOEs will display a higher level of ownership and shareholding in BRI projects than other regimes. Table  7.2 reports Chinese SOEs as owners account for a small proportion of BRI projects, but they are most prevalent among electoral autocracies (4.3%). Yet, the magnitude is so small as to hardly matter in relation to the total volume of BRI projects in which host country SOEs play the dominant ownership role. Of greater interest is the prevalence of Chinese SOEs as shareholders in BRI projects. This is significant insofar as it may not trigger alerts with regard to Chinese SOEs as project owners, but if often signals they have an important stake in the project’s success. Over 25% of all BRI projects have Chinese SOEs as a shareholder. The results indicate a high prevalence of China SOE shareholding in electoral autocracies relative to other regimes (15% for electoral autocracies relative to electoral democracies at 5.75%, closed autocracies are at 4.1% and liberal democracies at 1.18%). A variety of factors enhance the attractiveness of host countries to tap Chinese SOEs for additional responsibilities in BRI projects. The economies of scale and scope of Chinese SOEs, the depth of funding available, and greater experience implementing and managing large infrastructure projects in relation to domestic firms coupled with electoral autocracy rulers’ incentive to complete projects in a timely manner to bolster their political support will enhance the desire to rely on Chinese SOEs for related project responsibilities, such as financing, contractor, developer, and consultant. Table 7.2 indicates about 28% of all projects have Chinese SOE financing, with the largest fraction again for projects in electoral autocracies (16.4%) followed by electoral democracies (8%). The attribute with the highest prevalence of Chinese SOE participation is the contractor, with nearly 43% of all projects having this feature. It is far more common to projects located in electoral autocracies (24.18%), followed by projects in electoral democracies at a distant second (10.89%). Chinese SOEs play a relatively small role as developer (5.75%) or consultant (7%), but this role is more prevalent for projects in electoral autocracies than other regimes. Finally, I expect electoral autocracies to host more public-private partnerships (PPPs) with Chinese SOEs than other regimes. Typically, this project structure occurs in the context of long-term infrastructure development that lasts twenty-five years or more. Thus, it represents a significant and long-term commitment. The data reveal that 28% of all projects have this structure, with most located in electoral autocracies (15.71%) followed by electoral democracies (7.21%).

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

188

China’s Chance to Lead

The clear pattern from these results is that BRI projects typically have state involvement, most frequently by a host country SOE followed by China SOEs as contractors. Additionally, most BRI projects and the accompaniment of Chinese SOEs in various capacities are most frequently located in electoral autocracies. Based on these patterns, a question emerges as to whether projects owned by a host country SOE overlap with Chinese SOE ownership, PPP structure, Chinese SOE shareholding, Chinese SOE financing, or do they tend to occur separately. Analysis below will provide some answers. 7.3.2.1 Pairwise Correlations and Differences in Proportions To identify whether the project characteristics identified above commonly occur together (i.e., as part of a single project), or whether they occur separately, we can examine pairwise correlations between the variables. Tables 7A.1 to 7A.4 report pairwise correlations between these variables for each regime type. So as to minimize the interruption to the discussion, I present the tables in the appendix. For each table, the two columns on the right indicate the number of statistically significant positive and negative correlations with other variables. The bottom row indicates the total number of significant correlations with a variable that is a China SOE (owner is a China SOE, financing source is from a China SOE, etc.). Also reported is the proportion of projects for each correlate out of the total number of projects for that regime (reported at the top of each table). I begin with closed autocracies in Table 7A.1. The results for host country SOE ownership is positively correlated with project completion, with Chinese SOE ownership, with Chinese SOEs as shareholders, and with financing from Chinese SOEs. These results suggest host country and Chinese state co-ownership and collaboration is quite common in BRI projects located in closed autocracies. This is perhaps unsurprising given the dominance of the state sector in these regimes. At the same time, host country SOE ownership also displays a positive correlation with PPPs, indicating the interest in collaborating with Chinese SOEs as the “private” partner, presumably to access their financing capabilities so as to minimize their own debt load since the pairwise correlation for PPPs and financing source China SOE is also significant. Results for electoral autocracies are reported Table 7A.2. The results for Owner host country SOE display a significant negative correlation with Chinese SOEs as an owner, but more interesting is the lack of significant correlations with other variables such as Chinese SOEs being a contractor, financing from Chinese SOEs, and Chinese SOEs as shareholders. Instead, the one strongly positive correlation is with PPPs,

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

Political Regimes and BRI Project Characteristics

189

indicating a strong presence of Chinese SOEs taking the “private” role in these arrangements. Project completion is most strongly correlated with PPPs, with Chinese SOEs as shareholders, and with financing from Chinese SOEs. Results from the correlation tables for other regimes reveal that project completion and PPPs are positively correlated only in the context of electoral autocracies. Table 7A.3 reports pairwise correlations in the context of electoral democracies, which are quite different to those of the two autocratic regimes. While host country SOEs display a positive correlation with PPPs as in the autocratic countries, there is a negative correlation with Chinese SOEs as a financing source and a strongly negative correlation to Chinese SOEs as owners. These results suggest private actors are more important to BRI projects. Consistent with this pattern is the lack of a positive correlation for project completion in relation to Chinese SOEs as shareholders and Chinese SOE financing in contrast to electoral autocracies. The total number of significant China SOE correlations, reported in the bottom row, is 19 compared with 24 for both autocracies, corroborating the reduced role for Chinese SOEs in electoral democracies. Finally, results for liberal democracies in Table 7A.4 reinforce and extend the orientation away from state involvement that is apparent in electoral democracies. There is a strong positive correlation between host country SOE ownership and the use of PPPs, indicating cooperation with private firms since PPPs do not display significant results for any other China SOE variables. Further, there are no other positive correlations between the host country SOE as an owner and other indicators of Chinese SOE involvement, resembling the pattern in electoral democracies. Chinese SOE involvement is most likely when a Chinese SOE is an owner or shareholder, but otherwise there are virtually no positive correlations. The total number of significant China SOE correlations reported in the bottom row drops to 15, compared with 19 for electoral democracies and 24 for both electoral autocracies. Overall, there is a clear pattern starting with heavy host country SOE involvement, including collaboration with Chinese SOEs, in closed autocracies. Project completion is more likely when these state actors are involved. Moving to electoral autocracies, the collaboration between host country and Chinese SOEs dissipates, but project completion is more likely when a Chinese SOE is a shareholder, when Chinese SOEs provide financing, and when PPPs are used, likely corresponding to host country SOEs partnering with a Chinese SOE as the “private” partner. In electoral democracies, PPPs remain common when host country SOEs are involved, but Chinese SOE financing is shunned. Project

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

190

China’s Chance to Lead

completion is relatively less likely with state involvement, though there is a positive correlation with Chinese SOEs as contractors. Finally, BRI projects in liberal democracies display few positive correlations, though Chinese SOEs as shareholders do correlate with Chinese SOE financing, but this does not contribute to better rates of project completion. Pairwise correlation tables offer insightful evidence about the propensity for BRI project characteristics to covary for individual political regimes. But to assess the relative prevalence of these characteristics and covariates across political regimes, it is necessary to investigate differences in proportions. Table 7.3 reports the differences in proportions for individual project characteristics and the correlates with other characteristics between electoral autocracies and other regimes. This table offers useful information about the relative frequency of certain project characteristics by regime type in comparison to other regime types. I compare the relative frequency of these characteristics for projects located in electoral autocracies in relation to those located in other regimes. The PPP results indicate electoral autocracies consistently display a heavier reliance on this project structure in comparison to other regimes. This finding holds when also considering the correlates with other project characteristics. These results are consistent with the expectation that electoral autocracies will more heavily rely on partners to bear some of the costs of infrastructure projects so the government can spread its budget across more projects. With regard to projects completed, the results indicate a higher proportion of projects are completed in electoral autocracies in comparison to any other regime type (6–7%), and these differences are highly significant. The differences with regard to correlates (e.g., projects completed and owner host country SOE) offer interesting additional information. For example, project completion and owner China SOE reduces the difference between electoral autocracies and other regimes (0–2%), indicating much depends on the role of the host country for seeing projects through to completion. For owner host country SOE, electoral autocracies display the greatest proportion of projects with this characteristic, though it is only slightly larger than those in electoral democracies. Surprisingly, the difference is vast in comparison to those located in closed autocracies. This is likely due to the propensity for political rulers and close allies in closed autocracies to own private companies that, while not strictly classified as SOEs, draw upon state finances for their funding, as in the UAE or Brunei, for example. This allows rulers to keep finances hidden from scrutiny via government budget reports and to deliver patronage without

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

24.02*** 3.39*** 9.70*** 15.53*** 5.64*** 6.24*** 12.27*** 7.27***

0.21 −0.01 −0.01 0.12 0.01 0.02 0.06 0.03

Owner Host Country SOE Owner Host Country SOE & Owner China SOE Owner Host Country SOE & Shareholder China SOE Owner Host Country SOE & Contractor China SOE Owner Host Country SOE & Developer China SOE Owner Host Country SOE & Consultant China SOE Owner Host Country SOE & Financing Source China SOE Owner China SOE

12.17*** 10.75*** 3.28*** 5.79*** 8.76*** 3.03*** 3.44*** 8.50***

12.95*** 7.48*** 13.68*** 4.98*** 8.10*** 7.40*** 5.54*** 4.13*** 7.43***

0.07 0.06 0.00 −0.01 0.05 0.00 0.01 0.06

0.04 0.04 0.10 0.02 0.02 0.02 0.02 0.01 0.02

z-statistic

Projects Completed Projects Completed & Owner Host Country SOE Projects Completed & Owner China SOE Projects Completed & Shareholder China SOE Projects Completed & Contractor China SOE Projects Completed & Developer China SOE Projects Completed & Consultant China SOE Projects Completed & Financing Source China SOE

PPPs PPPs & Projects Completed PPPs & Owner Host Country SOE PPPs & Owner China SOE PPPs & Shareholder China SOE PPPs & Contractor China SOE PPPs & Developer China SOE PPPs & Consultant China SOE PPPs & Financing Source China SOE

Difference in proportions

Electoral Autocracies – Closed Autocracies

Table 7.3  Differences in proportions for BRI project characteristics across political regimes

0.01

0.02 0.00 0.05 0.03 0.04 0.02 0.03

0.06 0.06 0.01 0.04 0.02 0.02 0.00 0.03

0.02 0.04 0.02 0.00 0.04 −0.01 0.03 0.01 0.04

Difference in proportions

4.49***

14.16*** 2.30** 8.07*** 9.25*** 6.40*** 4.65*** 7.92***

9.02*** 8.44*** 2.79*** 6.29*** 5.73*** 4.09*** 1.35 6.04***

8.22*** 6.05*** 8.02*** 2.67*** 6.77*** 3.77*** 5.88*** 3.25*** 6.65***

z-statistic

Electoral Autocracies – Electoral Democracies

0.01

0.04 0.01 0.05 0.08 0.04 0.05 0.17

0.07 0.08 0.02 0.02 0.05 0.02 0.02 0.06

0.03 0.03 0.01 0.01 0.04 0.02 0.03 0.02 0.07

Difference in proportions

8.50***

27.81*** 5.67*** 13.61*** 17.83*** 7.75*** 8.27*** 15.95***

14.67*** 12.97*** 4.70*** 8.22*** 10.23*** 4.70*** 4.13*** 9.23***

16.25*** 8.49*** 15.37*** 5.44*** 10.28*** 8.96*** 6.99*** 5.13*** 10.20***

z-statistic

Electoral Autocracies – Liberal Democracies

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

−0.02 0.03 0.01 0.02 0.03 0.09 −0.01 0.00 0.08 0.03 0.00 0.02 0.02 0.00 0.12

Shareholder China SOE Shareholder China SOE & Contractor China SOE Shareholder China SOE & Developer China SOE Shareholder China SOE & Consultant China SOE Shareholder China SOE & Financing Source China SOE

Contractor China SOE Contractor China SOE & Developer China SOE Contractor China SOE & Consultant China SOE Contractor China SOE & Financing Source China SOE

Developer China SOE Developer China SOE & Consultant China SOE Developer China SOE & Financing Source China SOE

Consultant China SOE Consultant China SOE & Financing Source China SOE

Financing Source China SOE

*, **, and *** indicate significance at the 0.10, 0.05, and 0.01 level.

−0.01 0.02 0.01 0.01 0.02

Difference in proportions

15.13***

6.82*** 4.00***

7.41*** 2.33** 6.19***

17.36*** 3.48*** 5.00*** 11.17***

11.23*** 9.04*** 4.75*** 5.04*** 7.83***

4.09*** 4.82*** 3.74*** 3.16*** 5.54***

z-statistic

Electoral Autocracies – Closed Autocracies

Owner China SOE & Shareholder China SOE Owner China SOE & Contractor China SOE Owner China SOE & Developer China SOE Owner China SOE & Consultant China SOE Owner China SOE & Financing Source China SOE

Table 7.3  (cont.)

0.01

0.02 0.01

0.06 0.00 0.04

0.05 0.02 0.01 0.04

0.05 0.01 0.03 0.01 0.04

0.00 −0.01 0.01 0.01 0.01

Difference in proportions

8.34***

4.92*** 3.17***

7.95*** 1.89* 6.87***

11.38*** 4.86*** 4.01*** 7.27***

9.00*** 5.45*** 5.49*** 3.73*** 6.27***

2.47** 1.55 3.00*** 2.31** 3.83***

z-statistic

Electoral Autocracies – Electoral Democracies

0.20

0.04 0.03

0.05 0.01 0.04

0.10 0.01 0.04 0.12

0.05 0.06 0.02 0.02 0.04

0.01 0.01 0.00 −0.01 0.00

Difference in proportions

18.43***

8.91*** 6.20***

9.12*** 2.83*** 7.42***

21.44*** 5.54*** 7.35*** 13.48***

15.76*** 11.39*** 6.30*** 5.89*** 9.60***

6.057*** 4.82*** 3.00*** 2.31** 5.54***

z-statistic

Electoral Autocracies – Liberal Democracies

Political Regimes and BRI Project Characteristics

193

scrutiny and to financially benefit without public scrutiny. One might expect China SOEs to make up for this difference, but the results for those characteristics do not compensate for the sizeable gap. Indeed, for the Owner China SOE characteristic, the difference between electoral autocracies and other regimes is small. The other China SOE characteristics are similar insofar as projects in electoral autocracies are more likely to involve China SOEs as a contractor, developer, consultant, or as a financing source than those from other regimes. The smallest difference is consistently with electoral democracies. Although these two political regimes differ from each other in terms of the autocratic – democratic dividing line, they are similar in terms of a heavy reliance on patronage and clientelist networks for political rulers to remain in power. With regard to financing specifically, the large difference between electoral autocracies versus closed autocracies (0.12) in relation to electoral democracies (0.01) may also be due to many closed autocracies possessing abundant hydrocarbon resources that reduce their need for financing from Chinese SOEs. An even larger difference with liberal democracies (0.2) is indicative of the much greater availability of private sources of financing in these economies, as discussed in Chapter 2. Overall, electoral autocracies consistently display the largest proportion of projects with each of the project characteristics measured and compared as reported in Table 7.3, which is consistent with the expectation that electoral autocrats will be the most welcoming of BRI projects. The consistently small differences between electoral autocracies and electoral democracies suggest their common concerns about electoral competition may be pushing leaders of these regimes to host BRI projects. 7.3.3

Findings for Electoral Autocrats with a Secure or Insecure Hold on Power

This section turns to an investigation of project characteristics depending on the durability of electoral autocrats hold on power as a proxy for whether their rule is secure or insecure. I also compare these results to those displayed by other political regimes and their durability. Table 7.4 displays the prevalence of BRI project characteristics in electoral autocracies for different time periods (up to and including 2019, before 2010, 2010–2014, and 2015–2019). It also reports the z-statistic for the difference in the average score of these characteristics for electoral autocracies that are above and below the median durability value for all electoral autocracies, for those in the top 25% and the bottom 25%, and for those

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

194

China’s Chance to Lead

in the top 10% and the bottom 10%. Results for closed autocracies, electoral democracies, and liberal democracies are reported in Table 7A.5, Panels A, B, and C, respectively. I posit that host country SOEs are more likely to be owners of BRI projects in electoral autocracies with low durability since leaders in these circumstances desire greater control over the distribution of patronage available from these projects. Results for the entire sample (all years up to and including 2019) reported in Table 7.4 are generally consistent with this prediction, with differences between the top and bottom 25% and 10% groups displaying the strongest effects. As host country SOE ownership rises, Chinese SOE ownership is expected to decline though it is likely to be a minority shareholder. The results for Chinese SOE ownership indicate this characteristic is more prevalent in electoral autocracies with durability above the median value. At the same time, Chinese SOEs are more likely to be shareholders (minority owners) in weakly durable electoral autocracies; the effect is most pronounced for regimes in the top versus bottom 10% durability scores. These results are consistent with the expectations. A similar pattern to Chinese SOE shareholding is displayed with regard to Chinese SOEs providing financing. Projects in the least durable electoral autocracies (bottom 25% and 10%) are significantly more likely to have Chinese SOE financing. This is consistent with the argument that these regimes will more aggressively pursue infrastructure projects, and the financing needed for them, in order to expand patronage resources. PPPs display an even stronger pattern with regard to regime durability. They display strongly significant results when an electoral autocracy is weakly durable (below the median, bottom 25%, and bottom 10%). Indeed, of all the BRI characteristics, PPPs display the strongest effects in relation to regime durability. These results are consistent with the expectation that weak rulers will seek to boost the delivery of infrastructure projects by leveraging the resources of external partners (private firms and/or Chinese SOEs) and the delivery of patronage these projects enable. Weak leaders in electoral autocracies can use BRI infrastructure projects not only to boost patronage resources, but also to lift economic growth. The successful completion of these projects can also blunt criticism by political opponents that money is being diverted toward unproductive uses. Completing BRI projects therefore offers an effective way to strengthen their appeal among their clientelist network and attract new support from those outside this network. The results are consistent with this prediction: project completion is more common in electoral autocracies with durability in the bottom 10% and the bottom 25%.

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

1.41 9.24 8.29

1.79 7.12 8.34 2.59 0.14 1.51 0.94 5.70 0.71 2.92 2.07 5.33 0.28 2.97 2.07

Owner China SOE, all years to 2019 (%) Owner China SOE, before 2010 (%) Owner China SOE, 2010–2014 (%) Owner China SOE, 2015–2019 (%)

Shareholder China SOE, all years to 2019 (%) Shareholder China SOE, before 2010 (%) Shareholder China SOE, 2010–2014 (%) Shareholder China SOE, 2015–2019 (%)

PPPs All years up to 2019 (%) PPPs before 2010 (%) PPPs, 2010–2014 (%) PPPs, 2015–2019 (%)

9.19 0.47 5.98 2.73

6.74 0.47 3.30 2.97

1.32 0.33 0.75 0.24

18.94

0.94 3.77 0.94

0.94 3.44 1.60 17.25

504 5.66

518 5.98

Owner Host Country SOE, all years to 2019 (%) Owner Host Country SOE, before 2010 (%) Owner Host Country SOE, 2010–2014 (%) Owner Host Country SOE, 2015–2019 (%)

Number of Projects Projects Completed, all years to 2019 (% of all BRI Projects) Projects Completed, before 2010 (%) Projects Completed, 2010–2014 (%) Projects Completed, 2015–2019 (%)

−4.85*** −1.00 −4.75*** −1.40

−1.40 1.00 −0.71 −1.86*

2.99*** −1.27 2.32** 3.00***

0.98 −2.52** 0.06

−1.44

0.00 −0.58 1.91*

0.46

2.36 0.05 1.08 1.23

2.73 0.09 1.37 1.27

1.08 0.00 0.47 0.61

0.33 2.73 5.80

8.86

0.14 1.56 1.27

271 2.97

5.84 0.33 4.10 1.41

4.15 0.33 1.98 1.84

0.66 0.19 0.42 0.05

0.85 5.84 5.47

12.16

0.61 2.40 0.52

306 3.53

113 0.94

−5.72*** −2.12** −6.18*** −0.54

−2.52** −1.66* −1.56 −1.49

1.49 −2.00** 0.23 3.21***

0.71 0.05 0.38 0.28

0.71 0.09 0.28 0.33

0.38 0.00 0.19 0.19

−2.20** 0.19 −5.00*** 1.04 0.47 2.54

−3.50*** 3.77

−2.50** 0.05 −1.98** 0.28 2.60*** 0.61

−1.04

3.58 0.28 2.87 0.42

2.54 0.24 1.13 1.18

0.47 0.19 0.24 0.05

0.42 3.86 2.36

6.64

0.28 1.89 0.28

167 2.45

−6.46*** −1.89* −6.43*** −0.78

−4.73*** −1.13 −3.29*** −3.19***

−0.47 −2.00** −0.33 1.34

−1.39 −5.95*** 0.40

−4.21***

−1.89* −5.04*** 1.61

−3.80***

Durable Durable z-statistic Durable z-statistic Durable z-statistic above below for Durable bottom for Durable bottom for median median difference top 25% 25% difference top 10% 10% difference

Table 7.4  Regime durability and differences in BRI project characteristics for electoral autocracies

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

0.24 4.10 3.58

0.71 3.86 2.83 10.65 0.80 5.04 4.81 2.17 0.05 1.41 0.71 1.56 0.05 0.80 0.71

Contractor China SOE, all years to 2019 (%) Contractor China SOE, before 2010 (%) Contractor China SOE, 2010–2014 (%) Contractor China SOE, 2015–2019 (%)

Developer China SOE, all years to 2019 (%) Developer China SOE, before 2010 (%) Developer China SOE, 2010–2014 (%) Developer China SOE, 2015–2019 (%)

Consultant China SOE, all years to 2019 (%) Consultant China SOE, before 2010 (%) Consultant China SOE, 2010–2014 (%) Consultant China SOE, 2015–2019 (%)

*, **, and *** indicate significance at the 0.10, 0.05, and 0.01 level.

2.21 0.09 1.32 0.80

1.70 0.19 0.52 0.99

10.60 0.61 4.81 5.18

7.92

7.40

−1.58 −0.58 −1.65 −0.35

1.12 −1.34 2.98*** −1.00

0.05 0.73 0.35 −0.56

−0.39 −1.39

2.24**

−0.64

0.71 0.05 0.19 0.47

1.04 0.00 0.42 0.61

6.03 0.33 2.26 3.44

1.51 1.56

0.24

3.30

1.46 0.05 0.71 0.71

0.38 0.05 0.28 0.05

5.42 0.38 2.12 2.92

1.98 2.36

0.14

4.48

−2.37** 0.00 −2.52** −1.00

2.56** −1.00 0.78 3.21***

0.86 −0.26 0.31 0.96

−1.17 −1.88*

0.71

−1.98**

0.24 0.05 0.05 0.14

0.24 0.00 0.14 0.09

2.73 0.14 0.66 1.93

0.42 0.47

0.05

0.94

0.71 0.05 0.24 0.42

0.19 0.05 0.09 0.05

2.73 0.33 0.61 1.79

0.85 0.71

0.14

1.70

−2.24** 0.00 −1.63 −1.73*

0.33 −1.00 0.45 0.58

0.00 −1.27 0.19 0.34

−1.73* −1.00

−1.00

−2.15**

Durable Durable z-statistic Durable z-statistic Durable z-statistic above below for Durable bottom for Durable bottom for median median difference top 25% 25% difference top 10% 10% difference

Financing Source China SOE, all years to 2019 (%) Financing Source China SOE, before 2010 (%) Financing Source China SOE, 2010–2014 (%) Financing Source China SOE, 2015–2019 (%)

Table 7.4  (cont.)

Political Regimes and BRI Project Characteristics

197

Because domestic firms frequently lack the capabilities and experience needed to successfully complete large infrastructure projects, it is expected that Chinese SOEs will act as consultants, developers, and contractors. This expectation is heightened for those leaders who are more concerned about their hold on power. The results for China SOEs as consultants are consistent with this expectation. Electoral autocracies in the bottom 25% and the bottom 10% in terms of regime durability are more likely to have Chinese SOEs involved as a consultant. However, results for China SOEs as developers or contractors do not yield significant results. Finally, we can compare the results for electoral autocracies to those in the other regime categories. I expect closed autocracies to be less sensitive to durability since rulers in these regimes do not hold elections; thus, their clientelist network is smaller and they do not need the large-scale patronage resources provided by BRI infrastructure projects. The results for several characteristics are opposite to those identified for electoral autocracies, notably project completion and PPPs. The strongest effect is for Chinese SOEs as contractors for low durability at the 10%, 25% and below the median thresholds. Chinese SOEs as developers are also used for low durability closed autocracies at the 25% and below the median thresholds. Overall, the characteristics of BRI projects in closed autocracies display some stark differences to those in electoral autocracies. Mirroring the results from the regression results, electoral democracies display far greater similarities to electoral autocracies with regard to BRI project characteristics. This is expected since leaders in electoral democracies also depend on relatively large patronage resources that can be delivered to their clientelist network; however, this will occur more through private business than via SOEs. Accordingly, host country SOE ownership is less prevalent for low durability electoral democracies than in low durability electoral autocracies. Likewise, Chinese SOEs as owners or shareholders are also less common. Chinese SOEs are most prevalent in low durability electoral democracies as a financing source, and as a contractor or developer. An opposite result to electoral autocracies is observed for PPPs; more durable electoral democracies are more likely to have PPPs. Finally, project completion also displays inconsistent results with regard to regime durability. Overall, electoral democracies are more similar to electoral autocracies than the other regime categories, but there are some clear differences likely attributable to the preference for private business over SOEs. Liberal democracies display the fewest significant results associated with regime durability. Only two categories are worth mentioning. The first is for China SOEs as owners, which are more likely in weakly durable liberal democracies (those in the bottom 10%). This is c­ onsistent with

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

198

China’s Chance to Lead

anecdotal evidence from Greece which was forced to sell off governmentowned assets in the wake of the 2008 financial crisis. The other category is China SOEs as contractors, which are more common in highly durable liberal democracies. This may indicate the willingness to call upon their expertise for large-scale infrastructure projects when political leaders are more secure in their elected position. Overall, the results for BRI project characteristics between weakly and strongly durable electoral autocracies corroborate the findings of the regression analyses. Specifically, the characteristics of BRI projects in weakly durable electoral autocracies are consistent with the expectations that these leaders seek to increase the delivery of infrastructure projects both to lift access to patronage resources and boost economic growth in order strengthen their hold on power. The characteristics display the greatest differences to those located in liberal democracies, followed by closed autocracies and then electoral democracies. The reliance on an institutionalized party apparatus for the distribution of patronage offers an explanation for the similarities observed between electoral autocracies and electoral democracies, and the differences observed in relation to closed autocracies and liberal democracies. 7.3.3.1 Pairwise Correlations for BRI Project Characteristics in Electoral Autocracies To gain insight into BRI project characteristics that are related to each other depending on the durability of the political regime, I examine pairwise correlations. Because of my specific interest in electoral autocracies for theoretical reasons, I restrict this analysis to electoral autocracies. For space considerations, the tables are reported in the chapter appendix (Tables 7A.6 and 7A.7 for above and below the median durability value; Tables 7A.8 and 7A.9 for the top and bottom 25% of durability values, respectively; Table 7A.10 and 7A.11 for the top and bottom 10% of durability values, respectively). For each pairwise correlation table, I report the total number of significant China SOE correlations in the bottom row (based on the bottom six characteristics listed in each table which refer to China SOEs). I discuss each pair of tables in turn, starting with above and below the median durability value. There are three points that merit attention. The first point is the positive significant correlation between PPPs and project completion in the below the median table but not in the above the median table, which is consistent with the emphasis leaders in these circumstances place on attracting foreign partners to increase the initiation of infrastructure projects (and the patronage and growth opportunities they represent), and the related need to successfully complete projects to woo voters not

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

Political Regimes and BRI Project Characteristics

199

already part of their clientelist network. We can observe that the proportion of PPPs in below the median regimes is 0.39 whereas it is only 0.22 in above the median. The second notable point is the correlation between China SOEs as shareholders with PPPs and with project completion. These correlations are positive and significant for both above and below the median samples indicating the important role China SOEs play in this role regardless of the host country’s regime durability. The third notable point is the relative similarity in the number of positive significant correlations for China SOE correlations. The main difference regards the greater prevalence of China SOE financing sources for above the median regimes, corresponding to significant positive correlations with China SOEs as owners or shareholders. This is consistent with leaders who are more secure in their rule allowing China SOEs to take a stronger role in the implementation of BRI projects. Starker differences emerge when comparing projects in the top 25% most durable electoral autocracies to those in the bottom 25%. There are four points to observe. First, as in the prior pair of tables, PPPs and projects completed are statistically significant for the bottom 25% but not the top 25%, but this time the size of the correlation coefficient is 0.288 in comparison to 0.149 for below the median. Second, PPPs and host country SOE as the owner displays a statistically significant positive correlation, indicating the increasing importance of host country SOEs in leading these projects presumably to ensure patronage resources are allocated in the most beneficial manner for political incumbents. Third, China SOEs as shareholders remain strongly related to PPPs for both the top and bottom 25% samples, but they display a stronger positive correlation with project completion for the bottom 25% sample, indicating an increasing importance attached to their involvement in more nondurable regimes. Fourth, there is a notably larger number of positive significant China SOE correlations (22 for the bottom 25% and 16 for the top 25%). The major difference is due to the consultant category which is consistent with the expectation that leaders of weakly durable regimes seek China SOE consultation to ensure the successful delivery of large infrastructure projects. The differences due to strongly versus weakly durable regimes are most clearly evident in a comparison of the top and bottom 10% of electoral autocracies. Four points deserve mention. First, project completion is correlated with PPPs again for the bottom 10%, with a coefficient of 0.476 (compared to 0.288 for the bottom 25%). Second, PPPs and host country SOE ownership are also correlated again for the bottom 10%, with a coefficient of 0.359 (compared to 0.319 for the bottom 25%). Third, host country SOE ownership and project completion are also

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

200

China’s Chance to Lead

correlated again for the bottom 10%, with a coefficient of 0.217 (compared to 0.127 for the bottom 25%). Finally, China SOE shareholding is significantly correlated with PPPs and project completion for the bottom 10% but not the top 10%. This trend was weakly observed in the top and bottom 25% samples, and not apparent in the above and below the median samples, indicating their particular importance to BRI projects in weakly durable electoral autocracies. It is also worth mentioning that the number of significant SOE correlations has declined relative to the 25% tables, indicating the importance of China SOE shareholding relative to the other China SOE roles. The key takeaway from these correlation tables is the increasingly strong push by leaders in weakly durable electoral autocracies to fund and complete projects, turning to whomever they can to do so. This is reflected in weakly durable electoral autocracy’s correlation between PPP projects with either home or Chinese SOE involvement, and that these projects are the most likely to be completed. 7.3.3.2 Differences in Proportions for BRI Project Characteristics in Electoral Autocracies This section examines differences in the proportions of BRI characteristics and their pairwise correlations for electoral autocracies at different levels of regime durability. This offers an additional indicator for the relative importance of BRI characteristics at different levels of regime durability. Rather than measuring the strength and direction of a linear relationship between two variables, difference in proportions indicates the frequency of BRI characteristics (or their simultaneous occurrence) being present between two different durability levels. It can help identify the importance of particular characteristics for weakly versus strongly durable regimes. To this end, it is helpful to identify where differences are large, and whether the differences increase in size for different levels of regime durability. Table 7.5 displays the results. First, PPPs display a large difference (0.17) for durability above and below the median which increases up to 0.32 between EAs with durability in the top versus bottom 10%. This indicates a very substantial difference in the reliance on PPPs. This difference is magnified for PPPs that also have a host country SOE owner (from 0.16 up to 0.35). The difference for PPPs and project completion also displays a large increase across levels of regime durability, starting at 0.07 for durability above and below the median and reaching 0.22 for the top 10% versus the bottom 10%. Indeed, of all the BRI characteristics, these three characteristics display the largest differences. These results highlight the particular importance of PPPs to

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

−0.01 0.03 −0.01 0.01 −0.02 −0.01 0.01 0.00 0.09 0.00 0.07 0.02 −0.01 0.03 0.07

Owner Host Country SOE Owner Host Country SOE & Owner China SOE Owner Host Country SOE & Shareholder China SOE Owner Host Country SOE & Contractor China SOE Owner Host Country SOE & Developer China SOE Owner Host Country SOE & Consultant China SOE Owner Host Country SOE & Financing Source China SOE

0.17 0.07 0.16 0.01 0.06 0.10 0.02 0.03 0.06

Projects Completed Projects Completed & Owner Host Country SOE Projects Completed & Owner China SOE Projects Completed & Shareholder China SOE Projects Completed & Contractor China SOE Projects Completed & Developer China SOE Projects Completed & Consultant China SOE Projects Completed & Financing Source China SOE

PPPs PPPs & Projects Completed PPPs & Owner Host Country SOE PPPs & Owner China SOE PPPs & Shareholder China SOE PPPs & Contractor China SOE PPPs & Developer China SOE PPPs & Consultant China SOE PPPs & Financing Source China SOE

Difference in proportions

1.60 0.00 2.50** 0.30 −0.92 1.83* 2.30**

−0.47 0.84 −0.86 0.44 −1.22 −0.86 1.70* 0.32

5.03*** 4.04*** 5.08*** 0.54 2.64*** 5.05*** 1.75* 3.29*** 3.15***

z-statistic

Electoral Autocracies: Durable Above and Below Median

Table 7.5  Differences in proportions for BRI project characteristics in electoral autocracies

0.15 0.03 0.10 −0.04 −0.03 0.05 0.14

0.01 0.07 −0.01 0.04 −0.05 −0.01 0.02 −0.01

0.22 0.12 0.24 0.03 0.07 0.08 0.00 0.04 0.08

Difference in proportions

3.70*** 2.19** 3.32*** 0.00 −1.53 2.61*** 4.41***

1.05 2.52** 0.83 2.08** −1.48 −1.00 1.90* −0.33

5.84*** 4.97*** 6.48*** 2.33** 2.85*** 3.84*** 0.33 3.21*** 3.68***

z-statistic

Electoral Autocracies: Durable top and bottom 25%

0.14 0.04 0.19 −0.19 −0.02 0.05 0.10

0.13 0.18 0.00 0.11 0.01 0.00 0.02 0.02

0.32 0.22 0.35 0.05 0.17 0.05 0.00 0.03 0.05

Difference in proportions

4.32*** 2.11** 4.87*** −0.87 −0.38 2.50** 3.34***

3.83*** 4.72*** 0.38 3.67*** 1.22 0.00 2.00** 1.51

6.52*** 5.87*** 7.03*** 2.53** 4.94*** 2.36** 0.58 2.23** 2.33**

z-statistic

Electoral Autocracies: Durable top and bottom 10%

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

0.01 0.02 0.02 0.04 −0.02 0.01 −0.03 0.03 0.02 0.03

Contractor China SOE Contractor China SOE & Developer China SOE Contractor China SOE & Consultant China SOE Contractor China SOE & Financing Source China SOE

Developer China SOE Developer China SOE & Consultant China SOE Developer China SOE & Financing Source China SOE

Consultant China SOE Consultant China SOE & Financing Source China SOE

Financing Source China SOE

*, **, and *** indicate significance at the 0.10, 0.05, and 0.01 level.

0.05 0.08 0.00 0.02 0.00

−0.05 −0.02 0.01 −0.01 0.01 −0.04

Difference in proportions

0.66

1.59 1.31

−1.13 1.14 −1.92*

0.05 1.47 1.68* 1.35

1.44 3.28*** 0.15 2.01** 0.31

−3.02*** −1.42 0.82 −1.39 1.00 −3.36***

z-statistic

Electoral Autocracies: Durable Above and Below Median

Shareholder China SOE Shareholder China SOE & Contractor China SOE Shareholder China SOE & Developer China SOE Shareholder China SOE & Consultant China SOE Shareholder China SOE & Financing Source China SOE

Owner China SOE Owner China SOE & Shareholder China SOE Owner China SOE & Contractor China SOE Owner China SOE & Developer China SOE Owner China SOE & Consultant China SOE Owner China SOE & Financing Source China SOE

Table 7.5  (cont.)

0.05

0.05 0.02

−0.06 0.01 −0.03

−0.10 −0.01 0.03 0.02

0.07 0.06 −0.02 0.04 −0.01

−0.04 0.00 0.00 −0.01 0.01 −0.03

Difference in proportions

2.02**

2.38** 1.35

−2.57** 1.14 −1.61

−0.88 −1.00 2.05** 1.31

2.56** 2.56*** −1.61 2.57** 0.00

−1.49 0.20 0.33 −0.33 1.34 −1.81*

z-statistic

Electoral Autocracies: Durable top and bottom 25%

0.04

0.05 0.01

−0.02 0.01 −0.02

−0.17 0.01 0.05 0.03

0.19 0.09 −0.01 0.02 0.00

−0.01 0.02 0.02 0.00 0.00 −0.04

Difference in proportions

2.16**

2.24** 1.00

−0.33 1.00 −0.38

0.00 1.41 2.72*** 1.83*

4.76*** 3.29*** −0.58 1.51 0.90

−0.47 1.61 1.64 0.00 0.00 −1.42

z-statistic

Electoral Autocracies: Durable top and bottom 10%

Political Regimes and BRI Project Characteristics

203

electoral autocracies with low levels of regime durability, reflecting the desire of political incumbents to work with foreign partners to enhance the delivery of infrastructure projects and related patronage. Three additional characteristics are noteworthy. The first is host country SOE ownership which displays a difference of 0.09 for above and below the median, and increases to 0.15 for the top and bottom 25% and declines slightly to 0.14 for the top and bottom 10%. The second is the difference for host country SOE owners and project completion which increases from 0.03 for above and below the median to 0.18 for the top and bottom 10%. These differences reinforce the importance of the host country regime controlling BRI projects and the accompanying patronage resources. The third noteworthy characteristic is China SOEs as shareholders. This characteristic displays an increasing and significant difference for durability between the top and bottom 25% (0.07) and between the top and bottom 10% (0.19). This is notable because it is the only China SOE characteristic to display such a large difference between levels of regime durability. It also displays a large difference when paired with host country SOE ownership (0.19 between the top and bottom 10%). All the aforementioned characteristics are noteworthy because they displayed significant pairwise correlations in the previous section. The differences in proportions offer a starker illustration of their importance to weakly durable electoral autocracies. Taken together, the pairwise correlations and the difference in proportions offer telling evidence for the importance of these particular BRI characteristics for electoral autocracies with declining levels of regime durability. 7.4 Conclusions This chapter has investigated the prevalence of numerous BRI project characteristics with respect to different political regimes as well as the durability of political regimes as a proxy for the security of leaders’ hold on power. The findings of this chapter, based on a new and different dataset from Chapter 6, corroborate and extend the findings of the previous chapter. I highlight six important takeaways. First and foremost, the findings confirm that electoral autocracies are the primary recipients of BRI spending, accounting for more than half of all BRI projects. Second, BRI projects initiated in electoral autocracies display a higher completion rate than other regimes, which is also consistent with the troubled transactions findings reported in Table 6A.1. Third, electoral autocracies more frequently rely on host country SOEs when engaging in BRI projects than other regimes. This feature is consistent with

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

204

China’s Chance to Lead

the expectation that election pressures incentivize leaders of electoral autocracies to call upon state resources to initiate these projects, and to use them for delivering patronage to a clientelist network. While leaders of electoral democracies also face election pressures, they do not control state resources to the same extent as autocrats. This heavy reliance on host country SOEs in electoral autocracies is also consistent with the high prevalence of state ownership in the sectors in which BRI projects most commonly occur, transportation and energy, as displayed in Table 5.6. Fifth, the relatively higher prevalence of PPPs for BRI projects in electoral autocracies is consistent with the election pressures explanation. Finally, this explanation may also account for the relatively higher prevalence of Chinese SOE financing and role as contractor in BRI projects located in electoral autocracies. The pairwise correlations between these BRI project characteristics display a stronger prevalence and statistical significance for electoral autocracies in comparison to other regimes, adding further evidence in support of the underlying election pressures argument. The consistency of the findings with respect to an election pressures argument raises the question of whether the security of a political incumbent’s hold on power will yield systematic differences in BRI project characteristics, particularly in the context of electoral autocracies. Indeed, because electoral autocracies comprise a large share of all emerging economies, it would be helpful to identify a specific subset that is the most avid recipient of Chinese BRI spending. This may yield important insights not only into dyadic-level relations between China and BRI recipients, but also the potential for China-led coordination globally. The second part of the chapter investigates evidence for this possibility. There are two important findings. The first regards evidence consistent with the theory that weakly durable political rulers of electoral autocracies more avidly seek Chinese foreign spending, especially in the context of BRI infrastructure projects. Specifically, weakly durable electoral autocracies display more host country SOE ownership, greater use of PPPs, and higher project completion rates. Further analysis with regard to pairwise correlations indicates strong relations between host country SOE ownership and PPPs, host country SOE ownership and project completion, as well as project completion and PPPs for weakly durable electoral autocracies. Together, these pairwise correlations point to the importance of the underlying electoral pressures driving the desire for Chinese BRI projects. The differences in proportions between strong and weakly durable electoral autocracies reinforce this point. The second important finding is that electoral autocracies and electoral democracies display greater similarities in attracting Chinese foreign

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

Political Regimes and BRI Project Characteristics

205

spending than the two autocratic regime types. The analysis of BRI project characteristics indicates electoral autocracies share the greatest similarities to electoral democracies (Table 7A.5, Panel B), with larger differences observed in the comparison with closed autocracies and then liberal democracies. These findings are what one would expect from the perspective of electoral autocrats’ and democrats’ shared electoral pressures and the pursuit of patronage to deliver to their clientelist networks.

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

Shareholder 0.10* China SOE N 28 Proportion 0.09

0.25*** 63 0.21

25 0.08

15 0.05

5 0.02

0.22***

0.17***

161 0.53

1

0.09

34 0.11

51 0.17

Owner China −0.01 SOE N 4 Proportion 0.01

0.15***

13 0.04

0.16***

48 0.16

0.02

Projects Completed N Proportion

Owner Host Country SOE N Proportion

1

1 76 0.25

PPPs N Proportion

PPPs

Owner Host Projects Country Completed SOE

15 0.05

0.32***

17 0.06

1

Owner China SOE

Table 7A.1  Closed autocracies (number of projects 305)

Appendix

87 0.29

1

5

2

7

3

4

1

1

Financing Contractor Developer Consultant Source Number Number Shareholder China China China China of pos of neg China SOE SOE SOE SOE SOE sig corrs sig corrs

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

0.34***

51 0.17

8 0.03

15 0.05

2 0.01

−0.03

0.14**

15 0.05

6 0.02

−0.04

0.22***

62 0.2

19 0.06

0.15***

0.04

0.03

4 0.01

0.03

0 0

−0.06

0 0

−0.05

1 0

−0.15***

Total Number of Significant China SOE Correlations

Financing 0.20*** Source China SOE N 25 Proportion 0.08

Consultant 0.01 China SOE N 5 Proportion 0.02

Developer 0.22*** China SOE N 10 Proportion 0.03

Contractor −0.12** China SOE N 20 Proportion 0.07

21 0.07

0.08

6 0.02

0.02

10 0.03

0.19***

31 0.1

−0.01

28 0.09

0.12**

16 0.05

0.25***

12 0.04

0.20***

112 0.37

1

9 0.03

0.24***

1 0

0.01

15 0.05

1

12 0.04

0.29***

19 0.06

1

58 0.19

1

24

5

2

7

3

3

2

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

Shareholder 0.18*** China SOE N 127 Proportion 0.11

0.01 221 0.2

0.08***

85 0.08

40 0.04

22 0.02

47 0.04

0.16***

92 0.08

−0.20*** 1

821 0.73

1

Owner China SOE

0.01

191 0.17

295 0.26

Owner China 0.06** SOE N 35 Proportion 0.03

0.03

87 0.08

0.26***

251 0.22

0.07**

Projects Completed N Proportion

Owner Host Country SOE N Proportion

1

1 322 0.29

PPPs N Proportion

PPPs

Owner Host Projects Country Completed SOE

Table 7A.2  Electoral autocracies (number of projects 1120)

297 0.27

1

5

2

7

3

4

1

1

Financing Contractor Developer Consultant Source Number Number Shareholder China China China China of pos of neg China SOE SOE SOE SOE SOE sig corrs sig corrs

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

0.01

259 0.23

92 0.08

73 0.07

17 0.02

0.06**

0.05*

65 0.06

22 0.02

−0.02

−0.01

41 0.04

0.08***

10 0.01

0.03

14 0.01

0.07***

26 0.02

367 0.33

0.01

−0.10***

−0.03

Total Number of Significant China SOE Correlations

Financing 0.04 Source China SOE N 110 Proportion 0.1

Consultant 0.02 China SOE N 29 Proportion 0.03

Developer 0.20*** China SOE N 54 Proportion 0.05

Contractor −0.18*** 0.02 China SOE N 100 121 Proportion 0.09 0.11

109 0.1

0.07**

40 0.04

0.12***

45 0.04

0.15***

150 0.13

0.05*

188 0.17

0.11***

59 0.05

0.12***

36 0.03

−0.03

513 0.46

1

60 0.05

0.22***

8 0.01

0.01

90 0.08

1

41 0.04

0.09***

89 0.08

1

348 0.31

1

24

5

2

7

3

3

2

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

Shareholder China SOE N Proportion

Owner China SOE N Proportion

0.01

21 0.04

41 0.07

7 0.01

16 0.03

0.07*

0.01

Owner China SOE

85 0.15

−0.02

22 0.04

26 0.05

122 0.22

5

2

1

2

4

2

1

Financing Contractor Developer Consultant Source Number Number Shareholder China China China China of pos of neg China SOE SOE SOE SOE SOE sig corrs sig corrs

0.28*** 1

41 0.07

−0.11*** 1

403 0.71

61 0.11

0.07*

1

24 0.04

−0.04

91 0.16

−0.01

Projects Completed N Proportion

Owner Host 0.24*** Country SOE N 137 Proportion 0.24

1

1 153 0.27

PPPs N Proportion

PPPs

Owner Host Projects Country Completed SOE

Table 7A.3  Electoral democracies (number of projects 564)

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

0.08**

10 0.02

9 0.02

2 0

8 0.01

−0.01

0.01

48 0.09

54 0.1

0.14***

0.10**

−0.07*

0.01

13 0.02

113 0.2

2 0

−0.01

2 0

0.06

16 0.03

−0.01

−0.07*

27 0.05

0.03

10 0.02

0.06

167 0.3

0.01

Total Number of Significant China SOE Correlations

Financing −0.12*** −0.01 Source China SOE N 32 27 Proportion 0.06 0.05

Consultant China SOE N Proportion

Developer China SOE N Proportion

Contractor China SOE N Proportion

35 0.06

−0.01

13 0.02

0.09**

6 0.01

0.11***

71 0.13

0.18***

74 0.13

0.03

23 0.04

0.13***

5 0.01

0.013

231 0.41

1

5 0.01

0.04

2 0

0.07*

11 0.02

1

17 0.03

0.10**

35 0.06

1

170 0.3

1

19

1

5

3

3

3

2

1

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

Shareholder China SOE N Proportion

Owner China SOE N Proportion

0.18*

7 0.06

9 0.08

0 0

2 0.02

0.12

−0.12

−0.01

17 0.15

−0.01

3 0.03

4 0.03

0.18**

8 0.07

−0.18** 1

80 0.69

29 0.25

11 0.09

1

6 0.05

0.35*** −0.07

18 0.16

0.07

Projects Completed N Proportion

Owner Host Country SOE N Proportion

1

1 30 0.26

PPPs N Proportion

PPPs

Owner Host Owner Projects Country China Completed SOE SOE

Table 7A.4  Liberal democracies (number of projects 116)

25 0.22

1

4

4

1

1

1

1

1

Financing Contractor Developer Consultant Source Number Number Shareholder China China China China of pos of neg China SOE SOE SOE SOE SOE sig corrs sig corrs

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

9 0.08

−0.01

2 0.02

2 0.02

4 0.03

*, **, and *** indicate significance at the 0.10, 0.05, and 0.01 level.

7 0.06

0.33*** 0.27***

2 0.02

0.32*** 0.13

2 0.02

0.38*** 0.17*

1 0.01

−0.13

Total Number of Significant China SOE Correlations

7 0.06

3 0.03

2 0.02 −0.11

0 0

1 0.01

−0.07

2 0.02

0.07

−0.08

−0.01

0 0

2 0.02

−0.01

29 0.25

7 0.06

−0.07

0.01

0.02

0.15

Financing Source −0.02 China SOE N 3 Proportion 0.03

Consultant China SOE N Proportion

Developer China SOE N Proportion

Contractor China −0.11 SOE N 8 Proportion 0.07

5 0.04

0.01

2 0.02

0.05

2 0.02

0.1

42 0.36

1

2 0.02

0.28***

0 0

−0.03

3 0.03

1

1 0.01

0.08

4 0.03

1

13 0.11

1

15

3

1

3

1

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

2.40 0.19 1.37 0.85

1.46 0.80 0.38 0.28

1.46

0.66

Shareholder China SOE, all years to 2019 (%) Shareholder China SOE, before 2010 (%) Shareholder China SOE, 2010–2014 (%) Shareholder China SOE, 2015–2019 (%)

1.89

1.89

0.42 0.05 0.24 0.14

0.24

0.99

0.38 0.33 0.00 0.05

3.58

0.24 0.94 0.19

0.57 0.05 0.28 3.53

139 1.37

145 0.90

Durable below median

Owner China SOE, all years to 2019 (%) Owner China SOE, before 2010 (%) Owner China SOE, 2010–2014 (%) Owner China SOE, 2015–2019 (%)

Owner Host Country SOE, all years to 2019 (%) Owner Host Country SOE, before 2010 (%) Owner Host Country SOE, 2010–2014 (%) Owner Host Country SOE, 2015–2019 (%)

Number of Projects Projects Completed, all years to 2019 (% of all BRI Projects) Projects Completed, before 2010 (%) Projects Completed, 2010–2014 (%) Projects Completed, 2015–2019 (%)

Panel A. Closed Autocracies

Durable above median 84 0.05

1.23

0.00

0.00 0.00 0.00 0.00

0.14

1.08

2.84*** 0.00 −3.46*** 0.00 −2.45** 0.00

−2.23**

−0.24 2.12** −2.23** 1.00

−2.54**

0.00

3.14*** 0.00

−0.08

1.70* 0.00 −4.16*** 0.05 0.63 0.00

−1.45

z-statistic for Durable difference top 25%

Table 7A.5  Regime durability and differences in BRI project characteristics

0.05 1.13 0.38

1.56

0.28 0.00 0.24 0.05

0.80

1.51

0.09

2.40

0.14 0.71 0.09

73 0.94

Durable bottom 25%

1.08

0.00

0.00 0.00 0.00 0.00

−1.00 0.00 −4.91*** 0.00 −2.83*** 0.00

−5.76*** 0.00

−2.23** −1.00

−2.45**

−3.13*** 0.14

−1.22

−1.41

−2.87*** 1.23

−1.73* 0.00 −3.51*** 0.05 −1.41 0.00

84 −4.15*** 0.05

z-statistic for Durable difference top 10%

0.05 0.28 0.24

0.57

0.19 0.00 0.19 0.00

0.24

0.38

0.09

0.71

0.14 0.09 0.00

29 0.24

Durable bottom 10%

−1.00 −2.45** −2.23**

−3.46***

−2.00**

−2.00**

−0.71

2.70***

−1.41

1.72*

−1.73* −0.58

−1.63

z-statistic for difference

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

Consultant China SOE, all years to 2019 (%) Consultant China SOE, before 2010 (%) Consultant China SOE, 2010–2014 (%) Consultant China SOE, 2015–2019 (%)

Developer China SOE, all years to 2019 (%) Developer China SOE, before 2010 (%) Developer China SOE, 2010–2014 (%) Developer China SOE, 2015–2019 (%) 0.52 0.00 0.19 0.33

0.14 0.05 0.14

0.05 0.42 0.19

0.00 0.00 0.05 0.33

0.66

0.05

0.09 1.56 1.56

0.19 0.19 1.18

0.42

0.24 3.20

0.80

0.57

1.56

0.14

0.24

Contractor China SOE, all years to 2019 (%) Contractor China SOE, before 2010 (%) Contractor China SOE, 2010–2014 (%) Contractor China SOE, 2015–2019 (%)

1.37

1.04

Financing Source China SOE, all years to 2019 (%) Financing Source China SOE, before 2010 (%) Financing Source China SOE, 2010–2014 (%) Financing Source China SOE, 2015–2019 (%)

1.32 0.33 0.61 0.38

2.21 0.57 1.37 0.28

PPPs All years up to 2019 (%) PPPs before 2010 (%) PPPs, 2010–2014 (%) PPPs, 2015–2019 (%)

0.05

0.00

0.00

0.05

0.94 0.00 0.85 0.09

1.73* −1.34 −1.26

−0.94

0.00 0.05 0.05

0.09

−1.00 0.00 −3.00*** 0.00 −1.34 0.00

−3.36*** 0.00

0.82 0.00 −4.78*** 0.14 −1.06 0.57

−3.52*** 0.71

−1.07

−0.93

0.71

−0.99

2.21** 1.15 2.48** −0.54

0.00 0.19 0.33

0.52

0.05 0.42 0.14

0.61

0.05 1.13 0.71

1.89

0.42

0.80

0.09

1.32

0.85 0.05 0.57 0.24

0.94 0.00 0.85 0.09

0.00

0.05

−1.34 −2.12**

−2.50**

0.00 0.05 0.05

0.09

−1.00 0.00 −3.00*** 0.00 −1.73* 0.00

−3.61*** 0.00

−1.00 0.00 −4.05*** 0.14 −0.58 0.57

−3.39*** 0.71

−2.53**

−4.13*** 0.00

−1.41

−5.03*** 0.05

0.33 −1.00 1.10 −1.13

0.00 0.00 0.09

0.09

0.05 0.00 0.00

0.05

0.05 0.09 0.47

0.61

0.19

0.05

0.09

0.33

0.28 0.05 0.19 0.05

1.00 −0.58

0.00

−1.00

−1.00

−1.00 0.45 0.43

0.38

−1.34

−1.00

−1.41

−2.12**

2.75*** −1.00 2.99*** 0.58

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

0.61 2.73 4.43

3.06 0.09 1.27 1.70

0.52 3.63 5.33 0.61 0.00 0.47 0.14 1.65 0.00 0.85 0.80 4.24 0.09

Owner China SOE, all years to 2019 (%) Owner China SOE, before 2010 (%) Owner China SOE, 2010–2014 (%) Owner China SOE, 2015–2019 (%)

Shareholder China SOE, all years to 2019 (%) Shareholder China SOE, before 2010 (%) Shareholder China SOE, 2010–2014 (%) Shareholder China SOE, 2015–2019 (%)

PPPs All years up to 2019 (%) PPPs before 2010 (%)

2.64 0.14

0.99 0.00 0.57 0.42

7.78

0.52 0.94 0.66

0.14 1.04 0.90 9.47

236 2.12

Durable below median

266 2.07

Durable above median

Owner Host Country SOE, all years to 2019 (%) Owner Host Country SOE, before 2010 (%) Owner Host Country SOE, 2010–2014 (%) Owner Host Country SOE, 2015–2019 (%)

Number of Projects Projects Completed, all years to 2019 (% of all BRI Projects) Projects Completed, before 2010 (%) Projects Completed, 2010–2014 (%) Projects Completed, 2015–2019 (%)

Panel B. Electoral Democracies

Table 7A.5  (cont.)

0.05 0.00 0.00 0.05

2.45 2.45

0.28

5.18

0.09 0.57 0.33

137 0.99

2.86*** 2.78 −0.45 0.09

−1.41 0.00 −1.35 0.38 −2.62*** 0.38

−3.03*** 0.75

−0.43 −1.73*

−1.38

1.66* 1.35

−0.41

1.96**

−2.14** 0.31 0.87

−0.11

z-statistic for Durable difference top 25%

1.13 0.14

0.09 0.42 0.99

1.51

0.33 0.00 0.09 0.24

1.08 3.25

0.42

4.76

0.28 0.47 0.19

132 0.94

Durable bottom 25%

0.00

2.17

0.00 0.00 0.00

55 0.00

0.00 0.05 0.09

0.14

0.00 0.00 0.00 0.00

3.87*** 1.04 −0.45 0.00

−1.41 −0.24 −2.42**

−2.32**

−1.41 −1.63

−2.12*

3.37*** 0.80 −1.57 1.37

−0.78

0.64

−1.42 0.43 0.91

0.16

z-statistic for Durable difference top 10%

0.57 0.00

0.00 0.09 0.75

0.85

0.09 0.00 0.05 0.05

0.33 1.93

0.19

2.45

0.00 0.09 0.14

63 0.24

1.72*

−0.58 −3.30***

−3.28***

−1.00 −1.00

−1.41

2.04** −1.45

−2.00**

−0.61

−1.41 −1.73*

−2.23**

Durable z-statistic bottom for 10% difference

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

Consultant China SOE, all years to 2019 (%) Consultant China SOE, before 2010 (%) Consultant China SOE, 2010–2014 (%) Consultant China SOE, 2015–2019 (%)

Developer China SOE, all years to 2019 (%) Developer China SOE, before 2010 (%) Developer China SOE, 2010–2014 (%) Developer China SOE, 2015–2019 (%) 0.80 0.00 0.33 0.47

0.00 0.05 0.57

0.05 0.24 0.00

0.00 0.05 0.14 0.61

0.28

0.19

0.38 2.03 2.92

0.19 1.23 2.40

2.12

1.60 5.33

1.89

1.46

3.82

0.19

0.05

Contractor China SOE, all years to 2019 (%) Contractor China SOE, before 2010 (%) Contractor China SOE, 2010–2014 (%) Contractor China SOE, 2015–2019 (%)

4.19

3.11

Financing Source China SOE, all years to 2019 (%) Financing Source China SOE, before 2010 (%) Financing Source China SOE, 2010–2014 (%) Financing Source China SOE, 2015–2019 (%)

1.32 1.18

2.54 1.60

PPPs, 2010–2014 (%) PPPs, 2015–2019 (%)

−2.12** 0.43

−0.73

−1.00 −1.63 1.73*

−0.63

−1.16 −2.06** −1.05

−2.35**

−1.25

−1.08

−1.34

−1.88*

0.00 0.00 0.38

0.38

0.00 0.05 0.09

0.14

0.09 0.47 0.94

1.51

0.75

0.28

0.05

1.08

2.89*** 1.65 1.18 1.04

0.00 0.05 0.24

0.28

0.00 0.09 0.00

0.09

0.24 1.04 1.65

2.92

1.23

0.94

0.14

2.31

0.33 0.66

0.00

0.24

−1.00 0.83

0.54

−0.58 1.41

0.45

−1.13 −2.12** −2.03**

0.00 0.00 0.00

0.00

0.00 0.00 0.00

0.00

0.00 0.00 0.05

−3.12*** 0.05

−1.55

−2.75*** 0.05

−1.00

−3.09*** 0.28

4.34*** 0.24 1.34 0.80

0.00 0.00 0.09

0.09

0.00 0.05 0.00

0.05

0.05 0.38 0.90

1.32

0.80

0.28

0.05

1.13

0.09 0.47

−1.41

−1.41

−1.00

−1.00

−1.00 −2.83*** −4.03***

−5.03***

−2.56**

−1.89*

−1.00

−3.29***

1.13 1.35

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

0.19 0.66 0.66

0.42 0.00 0.28 0.14

0.09 0.71 0.90 0.19 0.00 0.05 0.14 0.61 0.14 0.14 0.33 1.04 0.05

Owner China SOE, all years to 2019 (%) Owner China SOE, before 2010 (%) Owner China SOE, 2010–2014 (%) Owner China SOE, 2015–2019 (%)

Shareholder China SOE, all years to 2019 (%) Shareholder China SOE, before 2010 (%) Shareholder China SOE, 2010–2014 (%) Shareholder China SOE, 2015–2019 (%)

PPPs All years up to 2019 (%) PPPs before 2010 (%)

0.24 0.00

0.19 0.00 0.05 0.14

1.51

0.00 0.19 0.09

0.09 0.09 0.33 1.70

44 0.28

Durable below median

53 0.52

Durable above median

Owner Host Country SOE, all years to 2019 (%) Owner Host Country SOE, before 2010 (%) Owner Host Country SOE, 2010–2014 (%) Owner Host Country SOE, 2015–2019 (%)

Number of Projects Projects Completed, all years to 2019 (% of all BRI Projects) Projects Completed, before 2010 (%) Projects Completed, 2010–2014 (%) Projects Completed, 2015–2019 (%)

Panel C. Liberal Democracies

Table 7A.5  (cont.)

0.14 0.14 0.14

0.42

0.14 0.00 0.05 0.09

0.09 0.47

0.09

0.66

0.09 0.05 0.14

26 0.28

3.28*** 0.28 1.00 0.05

1.73* −1.00 1.27

0.86

0.00 0.00

0.00

0.19 0.87

−0.82

0.49

1.41 −0.82 1.66*

1.22

z-statistic for Durable difference top 25%

0.09 0.00

0.00 0.00 0.05

0.05

0.09 0.00 0.00 0.09

0.14 0.47

0.19

0.80

0.00 0.05 0.00

25 0.05

Durable bottom 25%

1.42 1.00

1.73* 1.73* 1.00

2.53**

1.00 0.00

0.45

−0.45 0.00

−0.82

−0.54

1.41 0.00 1.73*

1.89*

0.09 0.00

0.09 0.09 0.05

0.24

0.09 0.00 0.05 0.05

0.05 0.05

0.05

0.14

0.05 0.05 0.05

10 0.14

z-statistic for Durable difference top 10%

0.05 0.00

0.00 0.00 0.00

0.00

0.00 0.00 0.00 0.00

0.09 0.38

0.05

0.52

0.00 0.05 0.00

11 0.05

0.58

1.41 1.41 1.00

2.23**

1.00 1.00

1.41

−0.58 −2.33***

0.00

−2.14**

1.00 0.00 1.00

1.00

Durable z-statistic bottom for 10% difference

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

0.14 0.00 0.05 0.05

0.09 0.00 0.05 0.05 0.05 0.00 0.00 0.05

Developer China SOE, all years to 2019(%) Developer China SOE, before 2010 (%) Developer China SOE, 2010–2014 (%) Developer China SOE, 2015–2019 (%)

Consultant China SOE, all years to 2019 (%) Consultant China SOE, before 2010 (%) Consultant China SOE, 2010–2014 (%) Consultant China SOE, 2015–2019 (%)

*, **, and *** indicate significance at the 0.10, 0.05, and 0.01 level.

0.05 0.00 0.00 0.05

0.00 0.57 0.33

0.05 0.19 0.42

0.09

0.14 0.90

0.14

0.05

0.66

0.00

0.09

Contractor China SOE, all years to 2019 (%) Contractor China SOE, before 2010 (%) Contractor China SOE, 2010–2014 (%) Contractor China SOE, 2015–2019 (%)

0.24

0.28

Financing Source China SOE, all years to 2019 (%) Financing Source China SOE, before 2010 (%) Financing Source China SOE, 2010–2014 (%) Financing Source China SOE, 2015–2019 (%)

0.14 0.09

0.75 0.24

PPPs, 2010–2014 (%) PPPs, 2015–2019 (%)

−1.00 0.00

−1.00

1.00 0.00

0.58

1.00 −2.00** 0.50

−0.87

0.45

−1.00

1.41

0.30

0.00 0.00 0.05

0.05

0.09 0.00 0.05 0.05

0.05 0.14 0.28

0.47

0.14

0.05

0.09

0.28

2.98*** 0.14 1.13 0.09

0.00 0.00 0.05

0.05

0.05 0.00 0.00 0.05

0.00 0.14 0.19

0.33

0.09

0.09

0.00

0.19

0.05 0.05

0.00

0.00

1.00 0.00

0.58

1.00 0.00 0.63

0.73

0.45

−0.58

1.41

0.63

1.00 0.58

0.00 0.00 0.00

0.00

0.05 0.00 0.05 0.00

0.05 0.09 0.05

0.19

0.05

0.05

0.05

0.14

0.09 0.00

0.00 0.00 0.00

0.00

0.00 0.00 0.00 0.00

0.00 0.09 0.00

0.09

0.00

0.05

0.00

0.05

0.05 0.00

1.00

1.00

1.00

1.00 0.00 1.00

0.82

1.00

0.00

1.00

1.00

0.58

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

Shareholder China SOE N Proportion

Owner China SOE N Proportion

0.110**

40 0.08

45 0.09

13 0.03

14 0.03

0.206***

−0.00700

89 0.17

102 0.2

0.0300

−0.00700

25 0.05

0.227***

127 0.25

−0.0290

Projects Completed N Proportion

Owner Host Country SOE N Proportion

1

1 113 0.22

PPPs N Proportion

PPPs

Owner China SOE

84 0.16

−0.0150

18 0.03

25 0.05

0.180***

55 0.11

−0.287*** 1

366 0.71

1

Owner Host Projects Country Completed SOE

121 0.23

1

6

3

1

2

3

2

1

1

Financing Contractor Developer Consultant Source Number Number Shareholder China China China China of pos of neg China SOE SOE SOE SOE SOE sig corrs sig corrs

Table 7A.6  Pairwise correlations for electoral autocracies with durable above the median (number of projects 518)

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

0.083*

47 0.09

35 0.07

5 0.01

5 0.01

0.00800

−0.0570

13 0.03

18 0.03

−0.0420

0.0270

0.131***

105 0.2

−0.0550

27 0.05

0.0640

33 0.06

0.00700

165 0.32

23 0.04

66 0.13

0.0450

−0.248*** 0.096**

Total Number of Significant China SOE Correlations

Financing Source China SOE N Proportion

Consultant China SOE N Proportion

Developer China SOE N Proportion

Contractor China SOE N Proportion

28 0.05

0.154***

3 0.01

−0.0130

9 0.02

0.091**

10 0.02

51 0.1

0.142***

12 0.02

0.080*

21 0.04

0.164***

49 0.09

−0.177*** −0.0350

78 0.15

0.080*

20 0.04

0.089**

11 0.02

34 0.07

0.296***

2 0

−0.0260

46 0.09

−0.124*** 1

226 0.44

1

15 0.03

0.086*

33 0.06

1

157 0.3

1

25

6

3

4

3

6

1

3

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

Shareholder China SOE N Proportion

Owner China SOE N Proportion

0.103**

44 0.09

73 0.14

9 0.02

17 0.03

0.160***

0.0470

100 0.2

182 0.36

0.110**

0.0500

62 0.12

0.268***

120 0.24

0.149***

Projects Completed N Proportion

Owner Host Country SOE N Proportion

1

1 195 0.39

PPPs N Proportion

PPPs

Owner China SOE

118 0.23

0.0430

18 0.04

16 0.03

143 0.28

6

3

3

2

5

1

1

1

Financing Contractor Developer Consultant Source Number Number Shareholder China China China China of pos of neg China SOE SOE SOE SOE SOE sig corrs sig corrs

0.155*** 1

28 0.06

−0.093** 1

402 0.8

1

Owner Host Projects Country Completed SOE

Table 7A.7  Pairwise correlations for electoral autocracies with durable below the median (number of projects 504)

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

0.0400

44 0.09

66 0.13

12 0.02

22 0.04

0.00900

0.0130

9 0.02

30 0.06

0.0530

0.00800

0.254***

139 0.28

0.0520

42 0.08

0.077*

26 0.05

−0.0520

8 0.02

−0.0240

6 0.01

0.101**

4 0.01

0.0670

14 0.03

170 0.34

71 0.14

53 0.11

−0.094** 0.0260

−0.132*** −0.00500

Total Number of Significant China SOE Correlations

Financing Source China SOE N Proportion

Consultant China SOE N Proportion

Developer China SOE N Proportion

Contractor China SOE N Proportion

48 0.1

0.00300

24 0.05

0.161***

22 0.04

0.201***

86 0.17

0.196***

95 0.19

0.169***

32 0.06

0.151***

19 0.04

0.0450

225 0.45

1

20 0.04

0.131***

5 0.01

0.0440

36 0.07

1

23 0.05

0.106**

47 0.09

1

168 0.33

1

23

3

5

3

3

3

2

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

Shareholder China SOE N Proportion

Owner China SOE N Proportion

0.117*

19 0.07

22 0.08

8 0.03

3 0.01

0.262***

0.0830

Owner China SOE

41 0.15

0.0150

4 0.01

12 0.04

0.228***

23 0.08

−0.343*** 1

188 0.69

43 0.16

−0.0420

1

11 0.04

−0.0130

63 0.23

−0.0140

Projects Completed N Proportion

Owner Host 0.151** Country SOE N 42 Proportion 0.15

1

1 50 0.18

PPPs N Proportion

PPPs

Owner Host Projects Country Completed SOE

58 0.21

1

5

3

1

1

2

2

2

1

Financing Contractor Developer Consultant Source Number Number Shareholder China China China China of pos of neg China SOE SOE SOE SOE SOE sig corrs sig corrs

Table 7A.8  Pairwise correlations for electoral autocracies with durable in the top 25% (number of projects 271)

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

Total Number of Significant China SOE Correlations

41 0.15

20 0.07

11 0.04 −0.138**

2 0.01

1 0

0.0210

14 0.05

0.0740

−0.0570

6 0.02

4 0.01

−0.0740

0.0280

−0.00200

−0.0370

92 0.34

9 0.03

34 0.13

0.0510

−0.279*** 0.0740

Financing 0.0240 Source China SOE N 14 Proportion 0.05

Consultant China SOE N Proportion

Developer China SOE N Proportion

Contractor China SOE N Proportion

11 0.04

0.153**

1 0

−0.0160

5 0.02

0.152**

4 0.01

24 0.09

0.185***

5 0.02

0.0700

10 0.04

0.174***

24 0.09

−0.182*** −0.0610

38 0.14

0.0830

9 0.03

0.0620

6 0.02

−0.119*

128 0.47

1

13 0.05

0.226***

2 0.01

0.0460

22 0.08

1

7 0.03

0.115*

15 0.06

1

70 0.26

1

16

4

1

3

7

1

1

3

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

Shareholder China SOE N Proportion

Owner China SOE N Proportion

0.209***

34 0.11

45 0.15

5 0.02

12 0.04

0.137**

0.0570

76 0.25

0.0360

13 0.04

0.0510

258 0.84

69 0.23

0.202***

1

49 0.16

0.120**

75 0.25

0.288***

Projects Completed N Proportion

Owner Host 0.319*** Country SOE N 122 Proportion 0.4

1

1 124 0.41

PPPs N Proportion

PPPs

Owner Host Projects Country Completed SOE

13 0.04

0.310***

14 0.05

1

Owner China SOE

88 0.29

1

5

4

3

3

4

1

Financing Contractor Developer Consultant Source Number Number Shareholder China China China China of pos of neg China SOE SOE SOE SOE SOE sig corrs sig corrs

Table 7A.9  Pairwise correlations for electoral autocracies with durable in the bottom 25% (number of projects 306)

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

Total Number of Significant China SOE Correlations

4 0.01

90 0.29

4 0.01

18 0.06

27 0.09

0.134**

4 0.01

0.192*** −0.0120

8 0.03

13 0.04

0.0260

7 0.02

0.356***

5 0.02

24 0.08

−0.0520

17 0.06

0.193***

4 0.01

0.0770

45 0.15

−0.00800 0.178***

−0.0870

0.0100

3 0.01

5 0.02

0.0100

0.0490

0.0730

0.0140

92 0.3

34 0.11

23 0.08

−0.0920

−0.173*** −0.0810

Financing 0.0360 Source China SOE N 41 Proportion 0.13

Consultant China SOE N Proportion

Developer China SOE N Proportion

Contractor China SOE N Proportion

50 0.16

0.209***

20 0.07

0.187***

3 0.01

0

115 0.38

1

6 0.02

0.156***

5 0.02

0.284***

8 0.03

1

13 0.04

0.0790

31 0.1

1

95 0.31

1

22

3

4

3

3

1

1

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

Shareholder China SOE N Proportion

Owner China SOE N Proportion

0.160*

5 0.04

0.0780

3 0.03

10 0.09

−0.0360 4 0.04

0.299***

8 0.07

2 0.02

3 0.03

1 0.01

80 0.71 −0.278*** 1

13 0.12

11 0.1

1

Owner China SOE

−0.00600 0.143

−0.0590

3 0.03

0.0220

20 0.18

0.0240

Projects Completed N Proportion

Owner Host Country SOE N Proportion

1

1 15 0.13

PPPs N Proportion

PPPs

Owner Host Projects Country Completed SOE

15 0.13

1

5

2

1

1

2

2

1

1

Financing Contractor Developer Consultant Source Number Number Shareholder China China China China of pos of neg China SOE SOE SOE SOE SOE sig corrs sig corrs

Table 7A.10  Pairwise correlations for electoral autocracies with durable in the top 10% (number of projects 113)

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

−0.0330

3 0.03

3 0.03

0 0

0 0

0.0240

−0.100

1 0.01

1 0.01

−0.0840

0.0130

0.0430

11 0.1

−0.161*

3 0.03

−0.0510

4 0.04

0.0440

48 0.42

4 0.04

6 0.05

0.270***

−0.193** −0.198**

Total Number of Significant China SOE Correlations

Financing Source China SOE N Proportion

Consultant China SOE N Proportion

Developer China SOE N Proportion

Contractor China SOE N Proportion

6 0.05

0.414***

1 0.01

0.108

1 0.01

0.108

1 0.01

8 0.07

0.365***

3 0.03

0.296***

2 0.02

0.169*

5 0.04

−0.214** −0.141

10 0.09

−0.0120

1 0.01

−0.135

0 0

−0.221**

58 0.51

1

4 0.04

0.351***

1 0.01

0.163*

5 0.04

1

3 0.03

0.238**

5 0.04

1

20 0.18

1

18

4

3

3

1

7

1

1

3

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

Owner China SOE N Proportion

0.0480

4 0.02

9 0.05

50 0.3

75 0.45

0.225***

0.217***

42 0.25

0.359***

52 0.31

0.476***

Projects Completed N Proportion

Owner Host Country SOE N Proportion

1

1 76 0.46

PPPs N Proportion

PPPs

9 0.05

0.0390

141 0.84

1

Owner Host Projects Country Completed SOE

10 0.06

1

Owner China SOE

2

2

3

4

1

1

2

Financing Contractor Developer Consultant Source Number Number Shareholder China China China China of pos of neg China SOE SOE SOE SOE SOE sig corrs sig corrs

Table 7A.11  Pairwise correlations for electoral autocracies with durable in the bottom 10% (number of projects 167)

https://doi.org/10.1017/9781009385879.007 Published online by Cambridge University Press

−0.101

8 0.05

12 0.07

4 0.02

5 0.03

−0.128*

−0.0300

1 0.01

2 0.01

−0.0770

−0.0210

0.0140

11 0.07

33 0.2

0.105

13 0.08

0.0190

3 0.02

−0.0410

2 0.01

−0.0100

1 0.01

0.00900

1 0.01

0.126

5 0.03

40 0.24

10 0.06

14 0.08

46 0.28

0.365***

−0.311*** 0.0810

25 0.15

32 0.19

0.0140

−0.313*** −0.192**

0.226***

0.191**

Total Number of Significant China SOE Correlations

Financing Source China SOE N Proportion

Consultant China SOE N Proportion

Developer China SOE N Proportion

Contractor China SOE N Proportion

Shareholder China SOE N Proportion

12 0.07

0.0110

8 0.05

0.141*

1 0.01

−0.0250

22 0.13

0.0870

54 0.32

1

20 0.12

0.229***

10 0.06

0.211***

2 0.01

0.0500

58 0.35

1

3 0.02

0.204***

3 0.02

0.362***

4 0.02

1

6 0.04

0.141*

15 0.09

1

36 0.22

1

16

3

3

2

2

4

4

1

3

8

Case Studies of Political Regimes and the BRI

Chapters 6 and 7 have presented quantitative evidence linking political regimes to Chinese foreign construction and investment spending. The findings offer tantalizing evidence for the explanatory power of political regimes, and the specific importance of electoral autocracies. However, the persuasiveness of those findings remains limited without qualitative analysis that identifies how Chinese foreign spending is impacted by specific political regime characteristics. The aim of this chapter is to trace the link between political regimes and the initiation and implementation of BRI projects. The chapter includes five country cases that map evolving policies toward Chinese foreign investment and construction projects from before 2013 to after. Each case represents a strategically important partner to China’s Belt and Road Initiative. The political regimes illustrate how China’s foreign spending varies according to the context of each country’s political arrangements. The cases focus on Chinese SOE involvement in projects in each of the four regime types and include two electoral autocracy cases corresponding to leaders with a secure and an insecure hold on power. All cases include an overview of Chinese foreign investment and construction spending during the past two decades, and a focused examination of a construction project led by a Chinese SOE. An important mechanism by which China is rolling out the BRI is through the construction of port-led development models. To compare how these are implemented differently across political regimes, I examine such port projects in each of the case studies. Four of five cases focus on SOE-led port-related projects, the fifth case presents a private-led port-park project in addition to a SOE-led project. Although each country case examines an SOE-led project, how they are initiated and implemented is expected to vary by political regime, as theorized in Chapter 4. The cases in this chapter provide context and detail for the patterns observed in Chapters 5 through 7. They illustrate that the closed autocracy of the UAE possesses very different characteristics and incentives for its political leaders in attracting Chinese foreign spending in comparison 232

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

Case Studies of Political Regimes and the BRI

233

to the political incumbents of Malaysia, an electoral autocracy with strong political opposition threatening to end the ruling party coalition’s over five-decade hold on power. In the latter, the political incumbents use the power of the state to aggressively court Chinese foreign spending on numerous projects. And although political incumbents in electoral democracies, such as Indonesia, also face strong political opposition, Chinese foreign spending is more likely to occur in collaboration with local private firms than host country SOEs. In the Malaysia case, stateled patronage and clientelist networks are critical. In Indonesia, these patronage opportunities and clientelist networks are more private-sector oriented. Electoral autocracies in which the leader’s hold on power is more secure, such as Djibouti, resemble closed autocracies insofar as the state centralizes deal-making with China but does not engage in excessive risk-taking in an effort to boost and direct patronage to key constituencies in an effort to bolster support ahead of the next election. But in liberal democracies, such as Greece, the political power of patronage and clientelist networks is reduced. This yields far more attention to local stakeholders’ concerns, including workers, businesses, environmental groups, and the local community. In essence, local stakeholders tend to have far greater power in influencing the nature of Chinese foreign spending in liberal democracies. This typically leads to the privileging of private Chinese firms over SOEs, but the Greece case illuminates the unusual circumstances of a large Chinese SOE-led project, due to Greece’s debt crisis, in order to provide an equivalent comparison to the SOE-led projects in each of the other political regime cases. To preface the cases, I provide background discussions regarding BRI projects and political opposition as well as China’s international port investments via the BRI. As mentioned in earlier chapters, a core argument of the book regards the role of organized political opposition and concomitant reliance on clientelism in distinguishing electoral from closed autocracies. I therefore provide further discussion regarding the incentives of political incumbents to court BRI projects depending on the political regime in which they are located, the likely problems to emerge with BRI projects as a result of these incentives, and the capacity for political opponents to exploit these problems. In the second background section I discuss China’s port-related investments, which offer an important development model to implement the objectives of the BRI as articulated in the Vision and Actions statement with regard to promoting trade, investment, and connectivity both with China and with other members of the BRI. The port-park-city model is the most ambitious expression of the BRI’s aspirations, involving numerous projects and collaborative arrangements among public and private

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

234

China’s Chance to Lead

actors. Because these individual projects are often structured as PPPs, and involve multiactor and multiproject coordination, port-park-city development models (and related port-led development models) will last decades into the future, granting China the potential to cultivate longlasting ties with recipient countries and their political establishment. 8.1

Background on BRI Projects, Political Opposition, and Clientelism

I argue the key mechanism to yield differences between closed and electoral autocracies is the strength of political challengers. Strong political opponents force incumbents to institutionalize the ruling party and its patronage machinery. Weak electoral autocrats face strong pressures to expand and speed up the implementation of BRI projects that can deliver patronage to the ruling party’s clientelist network and promote economic development to attract new supporters. The speedy delivery of patronage is possible due to the autocratic structure of the political regime which enables the ruler to circumvent or ignore weak institutional checks on his power and skirt due diligence procedures. But this, in turn, raises the risks of longer-term problems such as inadequate assessment of a project’s social and environmental impact. At the same time, political regimes with weak institutions and strong political opposition are more likely to have such problems uncovered and exploited to attack political incumbents, as in a weak electoral autocracy (Malaysia) or an electoral democracy (Indonesia). For political incumbents, their strategy is to either forgo such risky behavior to begin with, try to prevent problems from being uncovered (typically before the next election), or gamble that the benefits generated by the project outweigh its drawbacks. In a closed autocracy (the UAE) or a strong electoral autocracy (Djibouti), by contrast, political opposition is nonexistent or weak, reducing the pressures on incumbent rulers to quickly initiate projects and lowering the likelihood of problems being exposed. In liberal democracies, institutions are typically strong enough to reduce, if not prevent, such problems from occurring due to adherence to due diligence processes but will also require more time before the project is authorized and completed. Common problems relating to BRI projects that often result from rushed implementation and the skirting of due diligence processes include an excessive reliance on imported Chinese labor and businesses, ignoring or failing to consult the local community about their interests and concerns, environmental damage, corruption, as well as financing problems and unsustainable debt. With regard to imported Chinese

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

Case Studies of Political Regimes and the BRI

235

labor and businesses, frequent complaints about BRI projects occur relating to inadequate local employment opportunities, a lack of training programs, ignoring opportunities to integrate local firms into new development projects and for knowledge transfer.1 While Chinese developers often argue that local workers and businesses lack the adequate skills, experience, or resources, inadequate host government attention to addressing these issues typically results in dissatisfaction and resentment which can be tapped by political opponents. Likewise, BRI project assessments may fail to adequately engage local communities and stakeholders to inform, consult, solicit input, or address their concerns. The top-down, government-to-government approach of Chinese SOEs is especially prone to disregard local stakeholders and affected communities. To speed projects toward implementation, environmental and social impact assessments may be completed in a haphazard manner or ignored altogether which can produce expensive environmental damage that only emerges later in the project’s life cycle. Additionally, BRI project contracts are often awarded without proceeding through an open, transparent competitive bidding process, resulting in corruption and waste through inflated prices and bribes. Although Xi Jinping introduced strict anticorruption rules in China, there do not appear to be anticorruption mechanisms in place for BRI projects. Finally, inadequate due diligence procedures potentially resulting in a series of problems as enumerated above will deter outside investors from participating in such projects. The additional problems of moral hazard on the part of Chinese lenders who can count on an implicit or explicit government guarantee leads Chinese developers to take excessive risks or pursue unprofitable projects. Consequently, host governments unable to share financing costs with outside investors may confront unsustainable debt problems. A major BRI project that illustrates these dynamics in the context of a weak electoral autocracy is Malaysia’s East Coast Rail Line (ECRL) which is planned to connect the Kuantan Port and the Malaysia-China Kuantan Industrial Park (MCKIP) to Malaysia’s larger national economy, following the prescribed Port-Park-City development model. Around the time the 1MDB corruption scandal was being exposed in 2015 and 2016 (Brown 2018), the ECRL was rushed to approval. Less than two weeks after Malaysia’s cabinet approved the project on October 21, 2016, Najib traveled to Beijing to sign the financing and construction agreement. China Communications Construction Company was awarded the contract on November 1, 2016 without an open bidding 1

In the context of the Asia Pacific, see Russel and Berger (2019) for examples of how these problems are manifested.

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

236

China’s Chance to Lead

process and without the involvement or input of local Malaysian stakeholders or even the state governments where the railway would be built.2 With a looming election to be held in May 2018, the groundbreaking ceremony was moved up to August 2017. Due to the rushed process, problems with the project’s feasibility and design soon emerged. For example, the pre-project assessment used five-year-old exchange rates to calculate the total costs of the project, which climbed from USD $7 billion to roughly $13 billion as work began.3 The initial project design also resulted in a 65-kilometer discrepancy in the length of the track that was ultimately implemented by the contractors. Following the ruling party’s surprise loss in the 2018 election, Mahathir’s administration also uncovered massive corruption and the need to pay roughly USD $5.4 billion in compensation and penalty charges if the project were to be canceled.4 China and Malaysia resumed construction on the ECRL after a yearlong suspension and following an agreement to cut its cost by nearly a third to about $11 billion. Malaysia Rail Link, the project’s local partner, said in a statement that domestic contractors would get 40 percent of the civil works (up from 30 percent in the initial contract) and that 70 percent of the workers would be local, with new programs focused on training and developing Malaysian skilled workers. Additionally, environment and social impact assessments would be more thoroughly completed prior to construction.5 As this brief snapshot reveals, political rulers can exploit BRI projects for short-term political gain, but such shortcuts often result in longerterm problems. The primary cause of such problems is the fear on the part of political incumbents of losing power which incentivizes them to engage in more risky projects, speedy approvals, and rushed implementation. The political regime also grants incumbent rulers varying resources and tools to do this. The combination of strong political opposition and centralized political control of state resources is theorized to produce the most extensive problems, as in weak electoral autocracies. Electoral democracies will also manifest problems, but to a lesser extent due to less centralized political control and stronger checks on executive power. Closed autocracies will have fewer problems exposed, but also 2

“You Cannot Rush a Slow Train, S’gor Exco Tells Liow,” Malaysiakini, March 20, 2018, www.malaysiakini.com/news/416408 (accessed August 4, 2021). 3 See www.brimonitor.org/wp-content/uploads/2021/07/CS_ECRL.pdf (accessed August 4 2021). 4 “Malaysia Must Pay S$7.4 Billion If It Terminates East Coast Rail Link,” Today (Singapore), June 5, 2018, www.todayonline.com/world/malaysia-must-pay-s74-billionif-it-terminates-east-coast-rail-link (accessed August 4, 2021). 5 See www.brimonitor.org/wp-content/uploads/2021/07/CS_ECRL.pdf (accessed August 4 2021).

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

Case Studies of Political Regimes and the BRI

237

have reduced pressures to push projects toward implementation due to the proscribing of political parties and elections. Liberal democracies are more likely to prevent problems from occurring due to more inclusive institutions and strong adherence to due diligence processes, but they will take longer to initiate and complete projects. 8.2

Background on China’s International Port Investments via the BRI

Starting with the launch of the BRI in the autumn of 2013, Chinese port investments increased at a substantially higher rate compared to those made by other global port companies. Between 2003 and 2017, Huo et al. (2019) identify thirty-nine foreign port projects by Chinese firms, including twenty-nine by China Merchants Group and COSCO (both SOEs), and ten by local port groups6 (predominantly SOEs).7 They find that twenty-eight out of thirty-nine projects were initiated from 2013 onward. Following the global financial crisis, the international shipping and maritime industry experienced declining growth. Port privatization initiatives provided investment opportunities for port-related companies, which Chinese firms more aggressively capitalized on than other global port companies. For example, investments initiated by the four major global terminal operators during 2013–2017 (PSA International, Hutchison Ports, APM Terminals, and DP World) ranged from 1.9 percent (HPH) to 20 percent (DP World) (Huo et al. 2019).8 Cosco Shipping Ports, by contrast, increased its investments by 100 percent. Ports and port-related development projects effectively match the aims of the BRI as articulated in the Vision and Actions statement by promoting regional economic development and thereby generating opportunities to expand and deepen cooperation with China and other members of the BRI. China Merchants Group points to its success in Guangdong as a template for other countries via the “transformative power of ports.”9 The most comprehensive version of port-related development initiatives is known as a Port-Park-City (PPC) model. This model entails building 6 These include the Dalian Port Group, Guangxi Beibu Gulf Port Group, Shenzhen Yantian Port Group, Rizhao Port Group, Hebei Port Group, Qingdao Port Group, and the Shanghai International Port Group. 7 Chinese private companies have also been involved in international port investment projects, such as Shandong Landbridge Group which is involved in the port logistics business through its subsidiary Landbridge Port. In 2015, the company won the bid for the 99-year lease of Port of Darwin, Australia. Subsequently, in 2016 it acquired the Margarita Island Port in the Colón Free Trade Zone in Panama. 8 Investments are measured according to the number of terminals. 9 See www.chinadailyasia.com/asia-weekly/article-11619.html (accessed August 21 2021).

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

238

China’s Chance to Lead

port infrastructure, industrial parks, logistics and export processing zones, developing commercial services and residential areas adjacent to the port and linking the port to the wider region with transportation infrastructure. China Merchants dubs this the “Shekou Model,” which is named after a once tiny fishing village in Guangdong Province (Dutton et al. 2020). In the early 1980s, China Merchants developed a seaport and export processing zone there, propelling neighboring Shenzhen’s growth into a world-class commercial, financial, and industrial hub. Drawing on this experience, the PPC model is designed to promote not just the transportation of goods, but also to foster the development of a wider business ecosystem. Building such infrastructure entails a long-term commitment on the part of both China and the host country government that will last decades into the future. By providing access to a large pool of loan capital from central policy banks (e.g., the Export-Import Bank and the China Development Bank) and BRI-specific lending facilities such as the “Silk Road Fund,” Chinese government policy encourages such overseas investment activities. For port-related projects requiring long-term coordination among multiple actors, projects, and a large financial and operational commitment, diplomatic and political support often accompany these projects, such as the Forum on China-Africa Cooperation or even direct government-to-government discussions (e.g., on the sidelines of China’s annual BRI Forum). Another port-based model is for a major port to serve as a regional hub at which the biggest container ships can dock, which then unload and dispatch their cargo onto smaller vessels serving other regional ports or onto trains traveling to land-based destinations. This model is being successfully deployed by COSCO in the Greek port of Piraeus. The port serves as a trans-shipment hub with goods arriving in Piraeus via the Suez Canal that are subsequently shipped to other parts of the Mediterranean. This saves time and money compared with unloading in the giant ports of northern Europe. COSCO also has plans to invest in a rail route to enable freight shipments between Piraeus and Central and Eastern European destinations. Chinese international port investments often occur via public-private partnerships (PPPs) that entail collaboration between Chinese companies as private entities and the public sector in a host country (Huo et al. 2019). The host country typically decides the structure of the PPP mode depending on their specific needs and objectives. The PPP modes most commonly used in Chinese international port investments, and in the port sector generally (Farrel 2010), include acquisition (typically of partial port assets owned by an SOE in the host country via partial

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

Case Studies of Political Regimes and the BRI

239

divestiture),10 followed by concession, public-private joint venture when the “private entity” is a Chinese SOE, and finally Build-Own-Transfer (BOT).11 Concession or acquisition is typically used when a firm’s main objective is to quickly increase market share. Concessions are limited to a predefined duration of time (though it often lasts several decades) whereas acquisitions provide an indefinite ownership claim to an asset. Joint ventures (and partnerships) enable the project to leverage the strengths of each member and to allocate risks to those most capable of handling them. Often, joint ventures involve partnerships with multiple foreign firms with each participant contributing to the project with their particular technical expertise and experience in port construction, operations and management, as well as their own network of shipping lines and alliances. BOT, as a form of greenfield investment, entails higher risks but is beneficial if port infrastructure is otherwise unavailable. This also lasts for a finite time period, typically lasting two or more decades. With regard to the country cases and port-related projects examined in this chapter, a concession was used when COSCO obtained a 35-year lease for two of three container terminals of Greece’s Piraeus Port in 2008 which was followed by the acquisition of a 51 percent stake from the state-owned Piraeus Port Authority (PPA) in 2016. China Merchants also acquired a 23.3 percent stake from the Djibouti government for the Port of Djibouti in 2013. A concession was also used when COSCO signed a 35-year agreement with Abu Dhabi Ports in the UAE for operating Khalifa Port Container Terminal 2 in 2016. Guangxi Beibu Gulf Port Group used a 30-year concession for Kuantan Port in Malaysia starting in 2015 which could be extended for another 30 years subject to the fulfilment of certain obligations. A joint venture was used between China’s Guangxi Beibu Gulf Port Group and Malaysia’s IJM Corporation in a 60–40 split for Kuantan Port in 2013 and by China Merchants Port Holdings (23.5 percent equity) and the public port authority of the Djibouti government, Port de Djibouti, S.A. (PDSA), finalized in 2012.12 Finally, none of the cases examined in this chapter involves the use of the BOT mode.13 10

See the Public-Private Infrastructure Advisory Facility (2007, 2009a, and 2009b) for details about these PPP modes. 11 BOT refers to a government or public sector actor granting a concession to a private sector actor (often a consortium) to finance, build, operate, maintain, and manage the facility, then transfers the facility to the public sector at the end of the concession period. 12 “Acquisition of 23.5% interests in Joint Venture in Djibouti,” The Stock Exchange of Hong Kong, December 30, 2012, www1.hkexnews.hk/listedco/listconews/ sehk/2012/1230/ltn20121230025.pdf. (accessed August 4, 2021). 13 The BOT mode is used by China Merchants Group for the Colombo International Container Terminal Sri Lanka. CMGroup is responsible for the construction and

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

240

China’s Chance to Lead

8.3

Closed Autocracy: The United Arab Emirates

8.3.1

Political Context

Since the discovery of oil and the formation of the Emirates in 1971, business and political elites have maintained a tightly knit community which has sought to preserve the stability of the UAE and their hold on power and wealth. Wealthy merchant families sought the help of ruling families to stop foreign merchants from competing against them. In return, members of the ruling families would participate in the running and managing of the merchant families’ businesses. Following the decline in oil revenue during the 1980s, the UAE implemented liberalizing reforms that attracted foreign direct investment. The share of the private sector in the UAE’s GDP increased from around 15 percent during the 1970s, in nominal terms, to over 50 percent in 2012, led by Abu Dhabi and Dubai. Despite this economic modernization, the UAE has remained a closed autocracy. Traditionally, the ruling families granted patronage in the form of direct income, land, and privileges to loyal supporters. Although merchant families dominate a large segment of the private sector, they also occupy high government positions, and these connections have further cemented their position in the private sector by allowing them access to various sources of revenue such as land ownership as well as limiting competition in the private sector. The strong interlinkages between the merchant and ruling families, and among the merchant families themselves, are evident from their presence on corporate boards and in government positions (Almezaini 2013). While there is a substantial share of SOEs, there are also a very large number of private firms linked to the ruling elite and wealthy merchant families. With a large fraction of Emirati citizens employed in the public sector, and political parties forbidden, incumbent rulers secure political quiescence from the people via rising incomes, social welfare, and minimal taxes. During the Arab Spring in 2011, for example, the UAE increased public salaries anywhere from 50 percent up to 120 percent and provided further material benefits to citizens (Assad and Barsoum 2019). But this overreliance on public sector employment has led to persistent recommendations to create more private sector opportunities for citizens that are attractive and help to diversify the economy (World Bank 2018). 35-year operation of the terminal which will be returned to the government of Sri Lanka after the expiration of its operation. A BOT model is also used for the deep sea port and industrial park of Kyaukpyu special economic zone in Myanmar with a fifty-year operation period.

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

Case Studies of Political Regimes and the BRI

241

For the political and business elite of the UAE to preserve stability and the status quo, the UAE’s small size requires maintaining strong relations with external powers to deter larger neighbors. Traditionally, the United States has played this role with 5,000 troops stationed in the UAE and a significant air force presence at the Al Dhafra Air Base. More recently, the UAE has pursued the development of bilateral ties with China in order to reduce a perceived overreliance on the United States, as well as a recognition that China would become an important trade and investment partner as its energy requirements steadily increase. Indeed, China’s growing economic ties with the Gulf region elevate the opportunities for the UAE as a regional base of operations for Chinese businesses, especially in the context of the BRI. 8.3.2

Chinese Foreign Spending and the Belt and Road Initiative

China’s interests in the Gulf region are anchored in access to energy, expanding trade and investment ties, and cooperation in the implementation of BRI projects. A stable regional environment is necessary to each of these, and the UAE is well-positioned to contribute to them. While both Saudi Arabia and Iran are important to China, neither meets the same range of Chinese interests with the depth that the UAE can provide, making it Beijing’s most mature Middle East relationship. With regard to energy, oil has played a historically important role in the UAE-China relationship. Although the UAE accounts for only 2.5 percent of China’s total crude oil imports (Sim and Fulton 2021), the UAE’s value stems from its being the only Gulf country open to foreign investment in its hydrocarbon sector. Chinese energy companies hold a 12 percent stake in Abu Dhabi’s onshore oil concession, the largest share granted to a foreign partner. The value of trade between China and the UAE has grown from USD2.5 billion in 2000 to nearly USD51 billion in 2019 (IMF Direction of Trade Statistics). China consistently ranks as the UAE’s top source of imports and among its top five export destinations with a balance of trade that favors China. The UAE also acts as a re-export hub for Chinese products to the rest of the Middle East, Africa, and Europe, with 60 percent of the UAE’s imports from China subsequently re-exported. Indicative of the deepening ties, China’s expatriate community in the UAE is estimated at over 200,000, the largest in the Middle East, with over 4,200 Chinese companies operating in the Emirates.14 14

Abdul Kader, “Chinese community in UAE Grows Fourfold in 10 Years”, Gulf News, 10 August 2016.

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

China’s Chance to Lead

9,000 8,000 7,000 6,000 5,000 4,000 3,000 2,000 1,000 0

12 10 8 6 4 2

Construction

Combined

Projects Initiated

19

18

Investment

20

17

20

16

20

15

20

14

20

13

20

12

20

11

20

10

20

09

20

08

20

07

20

06

20

20

20

05

0

Number of projects initiated

Millions of USD

242

Figure 8.1  Chinese foreign spending into the UAE, 2005–201915

Despite the increase in economic relations between the two countries, China never entered the top ten countries as a source of foreign investment into the UAE prior to 2016, when government data availability stops.16 However, Chinese investment increased substantially in 2017, as seen in Figure 8.1, due to investments in Khalifa Port and KIZAD. Chinese construction projects in the UAE took off in 2008 with the Etihad Rail Project involving the China Railway Construction Corporation. Since the launch of the BRI in 2013, nearly three dozen construction projects have been initiated (Table 6A.1). Notably, the host country owner for many of these projects is a private firm from the UAE, but with close links to the ruling elite. A common occurrence in closed autocracies is for members of the ruling families and close friends to own private companies that are effectively controlled by the state. Because they are not government-owned entities that may be required to disclose financial information, they can retain a greater level of opacity regarding their business and financial activities. This is observed in other closed autocracies such as the Suharto regime in Indonesia, Brunei, and other Middle East monarchies. Also notable among BRI projects in the UAE is the large number of real estate development projects. These tend to be privately owned, which is indicative of the role the wealthy elite play in striking deals with Chinese firms to drive up the value of their real estate holdings, 15

See Table 8A.1 for the data used to construct this figure. 16 Data about the UAE’s foreign investment inflows is available from the UAE’s Federal Competitiveness and Statistics Centre: https://fcsa.gov.ae/en-us/Pages/Statistics/ Statistics-by-Subject.aspx#/%3Ffolder=Economy/National%20Account/Foreign%20 Investment (accessed August 5 2021).

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

Case Studies of Political Regimes and the BRI

243

a key repository of wealth among local elites. This can partly account for the prevalence of real estate development appearing among the top two industries for investment and construction among closed autocracies in Table 6A.3. Due to the substantial state revenues from oil and the reduced pressure on ruling elites due to the absence of political parties, the UAE does not need to rely on PPPs to increase infrastructure projects on a limited budget like most emerging economies. Instead, PPPs are used to include partners for their infrastructure development capabilities with large, complex projects and their technical expertise. This accounts for the relatively lower level of PPPs observed among closed autocracies as displayed in Table 7.3. At the same time, the UAE displays a stronger commitment to the preparation of PPP projects according to the World Bank 2020 Benchmarking Infrastructure scores. In relation to the average score for closed autocracies in parentheses, the UAE scores are: Preparation 46 (35), Procurement 54 (55), Contract Management 58 (55), and Unsolicited Projects 42 (56). The stronger commitment to PPP preparation reflects the UAE’s efforts to attract foreign investment since it launched liberalizing reforms in the 1980s. To illuminate the effects of the UAE’s closed autocracy on BRI projects, I now turn to a port-park-city project considered a critical link on the maritime route connecting China to Europe, the Khalifa Port and the Khalifa Industrial Zone. 8.3.3

Khalifa Port and the Khalifa Industrial Zone (KIZAD)

Khalifa Port and the Khalifa Industrial Zone (KIZAD) were created as part of the Abu Dhabi government’s plan to achieve economic diversification by 2030. It is intended to attract foreign investments that will generate job opportunities for skilled Emiratis and expatriates alike. By 2030, KIZAD is expected to contribute up to 15 percent of Abu Dhabi’s non-oil GDP. They are about one-hour driving time to the cities of Abu Dhabi and Dubai. Khalifa Port is also planned to be connected to the new Etihad Rail network,17 being completed via a joint venture with the China Railway Construction Corporation.18 Thus, the project is typical of port-park-city initiatives with links to the wider region.

17

“BRI: Khalifa Port’s CSP Abu Dhabi Terminal crosses 540,000 TEU milestone,” Hellenic Shipping News Worldwide, April 21 2020. www.hellenicshippingnews.com/ bri-khalifa-ports-csp-abu-dhabi-terminal-crosses-540000-teu-milestone/(accessed August 4, 2021). 18 www.thenationalnews.com/business/economy/etihad-rail-awards-dh4-6bn-contract-tochinese-joint-venture-1.953254 (accessed August 4, 2021).

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

244

China’s Chance to Lead

An upgrade to Khalifa Port was completed in 2012 prior to Chinese involvement, and has since handled all of Abu Dhabi’s container traffic.19 KIZAD is the onshore portion of the port and was also opened in 2012. The park is necessary to boost trade through the port since Dubai’s Port of Jebel Ali, the world’s ninth busiest port and the biggest and by far the busiest port in the Middle East, is just 70 kilometers away. In 2017, COSCO signed a 35-year concession agreement to operate Khalifa Port Container Terminal 2 (Huo et al. 2019), which included a USD738 million investment to double the container-handling capacity of the port by 2020. Soon after, a consortium of five Chinese companies known as the Jiangsu Provincial Overseas Cooperation and Investment Company Limited (JOCIC) signed a fifty-year lease in KIZAD with an investment of $300 million (Fulton 2020). This was followed by another agreement announced by JOCIC in early 2018 that over fifteen Chinese companies had signed agreements in KIZAD worth $1 billion in sectors including construction, manufacturing, and trade and logistics. Khalifa Port and KIZAD illuminate the likely path other Port-Park-City projects will follow once completed with regard to a large influx of investment by private Chinese firms. A series of visits by government officials illustrate the importance attached to this project by both Abu Dhabi and China. For example, during President Xi’s state visit to the UAE in July 2018, China’s stateowned Industrial Capacity Co-Operation Financial Group Limited became the first Chinese financial services company granted access to Abu Dhabi Global Market. Its principal mission is to provide financial services for Chinese firms in KIZAD, support BRI projects in Abu Dhabi, and support the internationalization of the RMB for these projects.20 Shortly thereafter, a delegation from Abu Dhabi flew to Nanjing where JOCIC held an investment promotion conference attended by Chinese government officials and representatives from ninety Chinese companies looking to invest in KIZAD.21 In 2019, the deputy director 19

www.nytimes.com/2012/11/29/world/middleeast/a-port-to-keep-up-with-growth-inabu-dhabi.html (accessed August 4, 2021). 20 “Jiangsu Government Establishes State-Owned Financial Services Firm in Abu Dhabi Global Market to Support Belt-and-Road Initiative”, Abu Dhabi Global Market, July 20, 2018, www.adgm.com/media/announcements/jiangsu-government-establishesstate-owned-financial-services-firm#:~:text=share-,Jiangsu%20Government%20 establishes%20state%2Downed%20financial%20services%20firm%20in%20 Abu,support%20Belt%20and%20Road%20Initiative&text=The%20ICCFG%20 will%20support%20the,route%20including%20the%20MENA%20region (accessed August 4, 2021). 21 “Jiangsu Provincial Government and Potential Investors Explore Lucrative Investment Opportunities at China-UAE Industrial Capacity Cooperation Demonstration Zone Located in Khalifa Industrial Zone Abu Dhabi (KIZAD)”, Abu Dhabi Ports, September

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

Case Studies of Political Regimes and the BRI

245

of the NDRC, a member of the CCP’s Central Committee, attended the groundbreaking ceremony for COSCO’s KIZAD terminal in 2019, signaling the importance of the project to Beijing (Fulton 2019). Another notable feature of BRI projects in the UAE is the absence of public criticism. In other countries with effective political opposition, issues relating to opportunities for local workers, local business, environmental effects, the impact on local communities, as well as potential corruption and financing and debt issues would receive substantial media attention. But because over 70 percent of UAE workers are noncitizens, this virtually eliminates concerns about stealing jobs from Emiratis. This, in turn, reduces the need to ensure local firms participate in BRI projects as contractors or subcontractors, and evades the need for potential delays or obstacles to initiating them, such as environmental impact assessments. State control of the media prevents reporting on corruption, while the substantial oil revenue reduces concerns about unsustainable debt. 8.3.4 Summary China and the UAE’s commercial and governing elite share a common interest in the preservation of the UAE’s status quo political structure. For the UAE elite, shifting from an overdependence on the United States to cultivating stronger ties with China is necessary as China’s regional influence rises and as US commitments wane. For China, the UAE offers a business-friendly base of operations for its commercial interests in the region that is suitable for both its private firms and SOEs. The UAE’s liberalizing reforms have attracted private firms from around the world, contributing to the development of a thriving business and financial hub connected to the wider region. At the same time, the political structure of the UAE ensures the state retains residual control rights with regard to public-private engagements and is therefore a suitable location for hosting Chinese SOEs. The UAE’s closed autocracy political regime that focuses on the sharing of patronage among a small clientelist network coupled with a lack of urgency for the governing elite to foster close ties with China relative to political regimes that hold elections contributes to particular characteristics for BRI projects indicative of other closed autocracies. Specifically, there are lots of real estate projects connected to 11, 2018, https://adports.ucglab.com/kizad-an-abu-dhabi-ports-company-boosts-tradeties-with-china-during-jiangsu-provinces-investment-promotion-conference/ (accessed August 14, 2021).

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

246

China’s Chance to Lead

commercial families and a lack of criticism about BRI projects. Likewise, PPPs are not relied on as heavily as in other regimes with tight budgets and looming elections. Finally, there is a lack of public discussion about the socio-environmental impact of infrastructure projects in comparison to more democratic regimes, as will be seen later. Overall, BRI projects initiated in the UAE offer insight into the political mechanisms driving relations between closed autocracies and China more generally. 8.4

Strongly Durable Electoral Autocracy: Djibouti

8.4.1

Political Context

Djibouti is a small low-income country in a strategically important geopolitical location in the Horn of Africa. Historically, Djibouti has been a gateway to East Africa, a maritime cross-road for trade, and a refueling and transshipment center. Blessed with the region’s only deep-water port, Djibouti has served as a critical point of access to the sea for neighboring landlocked Ethiopia and as a transshipment hub for other East African countries. More recently, Djibouti’s strategic location close to major maritime transportation routes, conflict areas in the Middle East, and piracy activity zones, coupled with its relative political stability, has made it an important location for foreign military bases. These features have generated an economy heavily reliant on port services, logistics, and rents from military bases. Political parties are divided primarily along ethnic lines, the Afar and Issa Somali, making up about 35 percent and 60 percent of the less than 1 million population, respectively (Leta et al. 2015). The first president after independence from France in 1977, Gouled, was a member of the Issa. Although the ruling coalition included members from all groups, the Issa ethnic group retained the controlling power. Ismail Omar Guelleh succeeded his uncle as president in 1999, enabling the ruling party, the Union for the Presential Majority (UMP), to retain political control for over four decades. Power is highly personalized and increasingly centralized in the hands of President Guelleh, with control exercised via a patronage-based system. Funds for this patronage network primarily come from fees paid by port fees and foreign military bases (France, Germany, Spain, NATO, AfriCom, Netherlands, Operation Atalanta-EU, Japan, China, US).22 22

The United States pays USD63 million annually as of 2015, for example. See www.bbc​ .com/news/world-africa-33115502 (accessed August 4, 2021).

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

Case Studies of Political Regimes and the BRI

247

The economy is characterized by a lack of public services, with electricity that is unaffordable to the poor and blackouts occurring every day. More than 95 percent of agricultural products are imported and nearly 30 percent of the population is in a food vulnerable position.23 These conditions contributed to protests following the 2013 election. Six opposition parties formed a coalition to compete in the 2013 election, the Union for National Salvation (USN). The party won ten of sixty-five seats in the election, the first time since independence in 1977 that opposition parties gained representation in the legislature, although the USN cited what it referred to as cases of mass fraud and ballot tampering, and called for demonstrations. Around 100 protesters and opposition leaders were arrested and at least six protesters were killed during protests following the election. The main opposition parties boycotted the 2018 election, yielding fifty-seven of sixty-five seats in Parliament to the UMP. These features have contributed to a private sector that has remained underdeveloped since independence (Styan 2020). Typically, a private sector capable of funding political opposition is necessary for challengers to compete effectively in national elections (Carney 2018). This has contributed, in part, to the ruling party’s unwavering hold on power for over four decades. 8.4.2

Chinese Foreign Spending and the BRI

China’s interest in Djibouti is due to three main factors. First, Djibouti serves as a critical point of access to the sea for neighboring Ethiopia where Chinese firms have made substantial investments and constructed numerous industrial parks. Second, Djibouti’s strategic location along a heavily trafficked maritime route serves as a critical location for combating piracy. Third, Djibouti has made concrete steps with the 2008 opening of the Djibouti Free Trade Zone and of the Doraleh Container Terminal Port to revive its status as a maritime logistical hub for container transshipments, elevating its regional commercial importance. The contemporary period of bilateral ties between China and Djibouti started in 2012 with President Guelleh’s second state visit to the Forum on China African Cooperation conference. This led to Ethiopia, China, and Djibouti agreeing to fund the renovation of the Ethio-Djibouti railway and construct associated port facilities in Djibouti. This relationship was soon cemented with the acceleration of Chinese infrastructure investments into the Horn of Africa. 23

See www.acaps.org/es/country/djibouti/crisis/drought (accessed August 4, 2021).

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

248

China’s Chance to Lead

The proximate cause for Djibouti’s ruling party to engage with China in the development of projects was the need to address political unrest in the wake of the 2013 election and its underlying causes, including chronic and widespread poverty, lack of business and employment opportunities, and extreme income inequality. In response, Djiboutian authorities developed a strategy called “Vision Djibouti 2035,” which aims to transform the country into a middle-income economy and a logistics and commercial hub for East Africa. The strategy includes large-scale public sector investments focused on infrastructure, such as new ports, free-trade zones, and upgrading the railroad to Addis Ababa, among other projects. It was formally launched in June 2014 following a high-level meeting between the World Bank and Djibouti’s Ministry of Economy and Finance.24 Around this time, Chinese construction projects and the financing of other infrastructure facilities increased as displayed in Figure 8.2. Chinese infrastructure projects in Djibouti have been concentrated primarily in ports, industrial parks, railways, and energy-related investments. The first major port investment was the Port of Djibouti. The original concession was won by DP World in 2006 for a thirty-year concession but was taken over by the Djibouti government and given to China Merchants in 2012.25 In 2013, construction of the USD590 million Doraleh Multipurpose Port began, which is an extension of the Port of Djibouti, with terminals for handling oil, bulk cargo, and containers (to be discussed further below). Construction on the USD70 million Damerjog Port for livestock also started in 2013 with funding from China Merchants Group.26 There is also the USD64 million Port of Ghoubet which is for salt exports and was launched in 2017, also with funding from China Merchants Group.27 24

See www.worldbank.org/en/news/feature/2014/11/05/djibouti-turns-to-its-peers-foradvice-on-reaching-its-goals (accessed August 4, 2021). 25 For details about the dispute, see “A legal tussle over a strategic African port sets up a challenge for China’s Belt and Road plan,” Feb. 8 2019, https://qz.com/africa/1560998/ djibouti-dp-world-port-case-challenges-chinas-belt-and-road/ (accessed August 4, 2021). China Merchant acquired 23.5 percent of the Port de Djibouti for a total of $185 million. The acquisition also includes two-thirds of the Doraleh Container Terminal; source: “China Merchants Buys into Africa’s Port of Djibouti,” Jan. 2 2013, www.joc.com/port-news/international-ports/china-merchants-buys-africas-port-dji​ bouti_20130102.html (accessed August 4, 2021). 26 “Djibouti Starts Construction of Two Major Ports,” Sept 13, 2013, www.offshoreenergy.biz/djibouti-starts-construction-of-two-major-ports/ (accessed August 4, 2021). 27 “Port of Goubet opens up Djibouti to salt exploration and export,” June 27, 2017, https://africabusinesscommunities.com/news/port-of-goubet-opens-up-djibouti-tosalt-exploration-and-export/; accessed August 4, 2021. “Djibouti: Small country, big stakes,” August 21, 2018, www.theafricareport.com/541/djibouti-small-country-bigstakes/ (accessed August 4, 2021).

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

Case Studies of Political Regimes and the BRI

249

1,200

4

Millions of USD

3 800 2

600 400

1 200

Investment Combined

18 20 19

17

20

16

20

15

20

14

20

13

20

12

20

11

20

10

20

09

20

08

20

07

20

20

20

20

06

0

05

0

Number of projects initiated

1,000

Construction Projects Initiated

Figure 8.2  Chinese foreign spending in Djibouti, 2005–201928

Industrial parks are being constructed with funding and project implementation by China, including the Djibouti International Free Trade Zone which will be the largest industrial park in Africa once completed. This is discussed in the next section. There is also the Djibouti Damerjog Industrial Development Free Trade Zone (DDID FTZ)29 project, which focuses on heavy industry including oil, gas, ship repair, and livestock to be completed by 2032.30 To promote the use of the new port facilities, new railways, airports, and highways are also being built. These will be critical to moving goods to and from northern Ethiopia, in addition to avoiding congestion and delays as shipment volumes rise. The main Chinese-funded transportation project is the USD3.4 bn Addis Ababa–Djibouti Railway, the first all-electric inter-country railway system in Africa. It began operations in 2018,31 cutting travel time from about two days on the road to around twelve and a half hours. It is owned by the governments of Ethiopia and Djibouti, largely financed by China (the Export-Import Bank), and constructed by China Railway Group (CRG) and China Civil Engineering Construction Corporation (CCECC). 28

See Table 8A.2 for the data used to construct this figure. 29 For more details about the Djibouti Damerjog Industrial Park, see https://ddip.dj/ (accessed August 4, 2021). 30 “Djibouti pushes ahead with Damerjob project”, September 7, 2020, http://country​ .eiu.com/article.aspx?articleid=1520113535&Country=Djibouti&topic=Economy&su btopic=_2 (accessed August 4, 2021). 31 “Djibouti’s massive boom in transport infrastructure”, April 6, 2020, https://newafri​ canmagazine.com/22974/ (accessed August 4, 2021).

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

250

China’s Chance to Lead

With regard to energy-related investments, there is the Damerjog liquefied natural gas terminal at a cost of USD4 billion with construction that began in 2016. The financier, developer, and operator of the project is China’s POLY-GCL Petroleum Group Holdings Limited, a joint venture of China POLY Group and GCL Group.32 A crude oil terminal is also being developed nearby at a cost of USD200 million and oil will be shipped to the terminal from Ethiopia via a newly constructed pipeline.33 Djibouti exhibits patterns consistent with those identified among electoral autocracies more generally in Chapters 6 and 7 that indicate a large influx of Chinese foreign spending as part of the BRI and the use of PPPs to initiate more projects given finite resources. The political incentives are twofold, distribute patronage and pursue growth. The dramatic increase in PPPs for infrastructure projects is indicative of the need to respond to domestic challenges to political incumbents (as seen in Table 7.3 with PPPs distinguishing electoral autocracies from closed autocracies). Its World Bank PPP scores are typical of electoral autocracies, indicating strong political oversight but with greater attention to the needs of “private” partners. The World Bank Benchmarking study assigns Djibouti the following scores for its PPP framework in relation to electoral autocracies in parentheses: Preparation 40 (38), Procurement 51 (56), Contract Management 79 (63), and Unsolicited Proposals 50 (51). Due to Djibouti’s underdeveloped private sector, Chinese firms have been primarily responsible for the construction of BRI projects. Once approved, Djibouti’s contract management oversight is relatively more effective than other electoral autocracies. Not only does Djibouti seek to benefit from initiating new projects, which offers a way to deliver patronage to regime supporters, but also to grow the economy and create more jobs and thereby raise the popular standing of political rulers among all citizens via the completion of those projects (Table 7.3). Djibouti’s strategic importance to international trade is also evident by the number of overseas military bases that it hosts. China opened its first permanent overseas naval facility in Djibouti in August 2017. According to Beijing, it is a “logistics center” that allows China’s navy to conduct multilateral anti-piracy operations in the region.34 But by 32

“Construction Starts at Damerjog LNG Terminal in Djibouti, March 3, 2016, www​.off​ shore-energy.biz/construction-starts-at-damerjog-lng-terminal-in-djibouti/ (accessed August 4, 2021). 33 “Investment in Djibouti’s new port infrastructure continues”, https://oxfordbusiness​ group.com/analysis/docking-space-investment-new-port-infrastructure-capitalisecountry%E2%80%99s-location (accessed August 4, 2021). 34 “China hints more bases on way after Djibouti,” March 8, 2016, www.reuters.com/ article/us-china-djibouti/china-hints-more-bases-on-way-after-djibouti-idUSKC​ N0WA0IT (accessed August 4, 2021).

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

Case Studies of Political Regimes and the BRI

251

locating China’s naval facility adjacent to western powers demonstrates the autonomy and agency that Djibouti’s President wields in relation to China. For China, Djibouti is its first and only overseas military base; for Djibouti, China is one of numerous foreign militaries with a fee-paying presence in the country. Djibouti’s importance to China was reflected in China’s high-profile reception of the Djiboutian President during a state visit to Beijing in November 2017, the first foreign head of state to be received by Xi Jinping following the 19th National Congress of the Chinese Communist Party.35 Djibouti also became a formal participant in the MSRI on September 2, 2018 following a meeting between Guelleh and Xi Jinping in Beijing. However, the substantial number of Chinafunded projects and the high levels of debt owed to Chinese lenders have contributed to a rebalancing of the relationship. 8.4.3

The Doraleh Multipurpose Port (MPP) and the Djibouti International Free Trade Zone (DIFTZ)

The Doraleh MPP is located next to the original Port of Djibouti and was designed to be a major expansion that vastly improves the efficiency and ease of doing business in the Horn of Africa. It has helped alleviate congestion at the Port of Djibouti and greatly expanded cargo handling volumes in preparation for the DIFTZ and the rising volume of freight from Ethiopia due to Chinese firms’ growing presence in the country’s industrial parks. As of 2019, 90 percent of landlocked Ethiopia’s imports transited through Djibouti, giving the country with a population of less than a million leverage over its gigantic, 100 million strong neighbor.36 Due to the lack of domestic construction capabilities, the Doraleh MPP contractor was constructed by the China State Construction Engineering Corporation with the heavy equipment provided by the Zhenhua Port Machinery Co (SOE).37 Construction of the Doraleh MPP was completed in 2017. It is a public-private partnership with the Chinese state-owned firms China Merchants Group and Dalian Port Group as minority shareholders. The port can accommodate up to six of the world’s largest cargo ships simultaneously. It cost USD590 million, 35 “China, Djibouti agree to establish strategic partnership,” November 23, 2017, www​ .xinhuanet.com/english/2017-11/23/c_136774764.htm (accessed August 4, 2021). 36 www.washingtonpost.com/world/africa/in-strategic-djibouti-a-microcosm-of-chinasgrowing-foothold-in-africa/2019/12/29/a6e664ea-beab-11e9-a8b0-7ed8a0d5dc5d_ story.html (accessed August 4, 2021). 37 www.pairault.fr/sinaf/index.php/les-afriques-et-la-chine/pays-et-regions/chine-djibouti (accessed August 5, 2021).

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

252

China’s Chance to Lead

jointly financed by the Djibouti Free Zone Authority, China Merchants Group, and the China Export-Import Bank. The Djibouti International Free Trade Zone (DIFTZ) is owned by Djibouti Ports and Free Zones Authority (DPFZA) (60 percent); the China side retains a minority ownership stake (40 percent) and is composed of a consortium of investors including China Merchants Group (SOE), Dalian Port Corporation (SOE), and IZP Technologies (private).38 It is a ten-year, USD3.5 billion project and will be the largest free trade zone on the African continent when completed in 2028. It focuses on logistics, maritime, construction, automotive, and home appliance and electrical industries. Djiboutian authorities estimate the DIFTZ will create 350,000 new jobs over the next 10 years, with the government introducing a 70 percent cap on foreign workers during the first five years of the project, to be reduced to 30 percent thereafter. To attract international businesses to the DIFTZ, the site is offering a range of incentives, including zero percent rates on property tax, dividend tax, value-added tax, corporate income tax and personal income tax for foreign employees. As of 2021, over 100 companies have agreed to establish facilities in the zone.39 As in other Port-Park-City projects, the Doraleh MPP and the DIFTZ are connected to the wider region via a railway. In this case, an upgrade to the Ethio-Djibouti railway, renamed the Addis Ababa-Djibouti Railway which terminates at the Doraleh MPP.40 Both Chinese state-owned and private firms were involved in its construction. The political dominance of the ruling party coupled with an underdeveloped private sector and weak political opposition have contributed to 38 China Merchants Port Holdings Company Limited 2016 Annual Report provides additional details, www.cmport.com.hk/UpFiles/bpic/2017-04/20170427050724234.pdf (accessed August 4, 2021). 39 Additional details are available at the Djibouti International Free Trade Zone website: https://dpfza.gov.dj/facilities/Free-trade-area/djibouti-international-free-trade-zone (accessed August 4, 2021). 40 Companies involved in the construction of the Addis Ababa-Djibouti Standard Gauge Railway include both state-owned and private enterprises, such as Bank of China, China Merchants Group, China Resources Property, Nanjing China Railway, CLP Holdings, K Wah Construction Materials, Kerry Logistics Network, Lan Kwai Fong Group, Swire Group and the Hong Kong and China Gas Company. The Ethiopia-Djibouti railway, contracted by two Chinese companies China Civil Engineering Construction Corporation (CCECC) and China Railway Group Limited (CREC), is currently managed by a consortium of Chinese companies for a period of six years. See “China and the East Africa railways: Beyond full industry chain export,” July 6, 2017, www.brook​ ings.edu/blog/africa-in-focus/2017/07/06/china-and-the-east-africa-railways-beyondfull-industry-chain-export/ (accessed August 4, 2021). Also see “Chinese engagement in Djibouti’s economic transformation in a nutshell”, January 10 2020, www.xinhuanet​ .com/english/2020-01/10/c_138694414.htm; (accessed August 4, 2021).

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

Case Studies of Political Regimes and the BRI

253

a lack of criticism about BRI projects not offering adequate opportunities for domestic firms as subcontractors, for example. Likewise, there is little public criticism about Chinese workers taking job opportunities from local Djiboutians or the substantial increase in debt owed to China. It is therefore unsurprising that environmental effects of BRI projects would also go unmentioned, despite obvious risks concerning the construction of fuel and oil-related facilities. Indeed, the overwhelming dominance of Djibouti’s ruling party, despite holding elections (which are often uncontested), contributes to a political system that resembles a closed autocracy. 8.4.4 Summary China possesses strong interests in Djibouti, including the need to protect its ships from piracy and securing access to industrial parks in Ethiopia which are landlocked and therefore rely on Djibouti’s ports for access to the sea. Djibouti’s leaders are interested in patronage opportunities via rents from China’s military base, ports fees, and BRI projects, as well as pursuing its Vision 2035 initiative that seeks to promote the development of the private sector, create jobs and reduce poverty. Although Djibouti holds semi-competitive elections, opposition parties are weak. Often, national elections are uncontested. Consequently, Djibouti’s political regime bears close resemblance to a closed autocracy. The strong showing of political opposition in the 2013 national elections warned Djibouti’s leaders about the threat of political challengers and the need to pay greater attention to citizens’ and other stakeholders’ sources of discontent. Nevertheless, public criticism of BRI projects is uncommon. These dynamics have contributed to a political regime that is stable and relatively confident about retaining power. At the same time, the centralization of political control and absence of a well-developed commercial sector has contributed to strong government-to-government negotiations regarding China’s involvement in the country, both with its naval base and its substantial investments in BRI projects. Djibouti’s strategic location and Vision 2035 plan match the economic development model China Merchants Group has proposed with its “Port-Park-City” approach to infrastructure development. This involves the development of port facilities, coupled with industrial parks and transportation infrastructure to link the parks and surrounding region to the port. The major infrastructure projects at the heart of the ChinaDjibouti relationship are the Multi-Purpose Port (MPP) at Doraleh, the Djibouti International Free Trade Zone (DIFTZ), and the Addis

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

254

China’s Chance to Lead

Ababa-Djibouti Railway. All have relied heavily on Chinese SOEs for their construction, maintenance, and financing in strong coordination with the Djibouti government, and President Guelleh specifically. In summary, Djibouti’s dealmaking with China reflects the characteristics theorized to permeate Chinese infrastructure spending in an electoral autocracy with a secure hold on power. Specifically, the relationship involves a large financial commitment to the building of new infrastructure supported by government-to-government negotiations, a relatively high level of opacity in the structure and implementation of these agreements, greater weight assigned to political priorities relative to profits, and a relative inattention to these projects’ socio-environmental impact based on a lack of public discussion about these issues. Because Djibouti’s leader is secure in his hold on power, pressure to quickly secure financing and construct new infrastructure is reduced in relation to an electoral autocracy in which the leader has a more insecure hold on power as will be seen in the case of Malaysia. 8.5

Weakly Durable Electoral Autocracy: Malaysia

8.5.1

Political Context

Najib, Malaysia’s prime minister from 2009 to 2018, assumed his leadership position during a period of serious turmoil for the ruling party, Barisan Nasional (BN), which had held uninterrupted power for more than four decades. The 2008 election resulted in the BN’s parliamentary presence falling nearly 30 percentage points, to 63 percent, and the opposition securing a majority in five states for the first time in what was popularly termed a “political tsunami.”41 Najib’s predecessor, Abdullah, was blamed for the party’s disastrous showing, paving the way for Najib’s ascent. However, BN did not recover in the 2013 election, winning only 133 parliamentary seats compared to the 140 in 2008. Corporate scandals associated with the selective awarding of lucrative state contracts to political allies combined with selective patronage to ethnic Malay small and medium-sized enterprises contributed to the alienation of Chinese business owners and a growing segment of the middle class. Over the course of 2014–2016, news broke about 1MDB (a sovereign wealth fund established by Najib soon after he became prime minister and controlled directly by him) borrowing vast sums of money through questionable 41

“Malaysia’s Political Tsunami”, March 9 2008, Newsweek, www.newsweek.com/malay​ sias-political-tsunami-84183 (accessed August 4, 2021).

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

Case Studies of Political Regimes and the BRI

255

deals, funneling much of it to political operations, with around USD7 billion missing or misappropriated (Brown 2018).42 This news threatened the survival of the BN government, intensifying pressure on Najib in the lead up to the 2018 elections. While UMNO’s decades-long rule was sustained partly through a reliance on gerrymandering, media controls, and other authoritarian measures, a central part of this electoral dominance was due to targeted patronage primarily to ethnic Malays. Historically, this occurred through a Malay business class that received preferential access to government resources (tendering of contracts, privatization of state assets, lending). After the Asian financial crisis of 1997–1998, many formerly UMNO-linked businesses were nationalized by the government and patronage became centralized through government-linked corporations (GLCs) and government-linked investment companies (GLICs), enabling development spending to serve the political goals of the prime minister (Gomez et al. 2017). Consequently, the leading construction companies and private banks that lent to them became GLCs (Lim 2015; Gomez et al. 2017; Carney 2018). In this way, the construction industry continued its historical role of serving UMNO’s political interests, though with a sharper focus on the interests of the prime minister. Inflated government contracts accompanied by heavily discounted land and permits directed to GLCs enabled UMNO to fill its coffers in order to fight increasingly competitive elections. Often, these lucrative contracts were awarded to GLCs that had formed joint ventures with foreign companies (Gomez et al. 2017). Indeed, UMNO’s legitimacy rested both on its ability to deliver patronage to ethnic Malays and on rapid economic growth which in turn depended on high levels of foreign investment (Liu and Lim 2019). This institutionalized system of patronage in the form of contracts to the construction industry was sustained and amplified through the building of megaprojects. As a result of the government’s reliance on partnering with private firms in the construction of these large projects, public-private partnerships became an increasingly common form of project administration. Between 1990 and 2013, Malaysia had the second highest number of PPP projects in the Asia Pacific region (17 percent of the region’s total) after China (33 percent) (World Bank 2013; Osei-Kyei and Chan 2017). 42

On July 2, 2015, British journalist Clare Rewcastle Brown published an investigative article in her Sarawak Report blog that said Najib had diverted into his own bank account over US$700 million from 1MDB, a figure later confirmed by the US Federal Bureau of Investigation.

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

256

China’s Chance to Lead

8.5.2

Chinese Foreign Spending and the Belt and Road Initiative

China’s interest in Malaysia is due to its strategic location next to the Melaka Strait and South China Sea, and its demand for infrastructure projects and related commercial opportunities. The Melaka Strait is the world’s busiest maritime area, with nearly a third of world seaborne trade passing through it each year, including 80 percent of China’s energy imports.43 Because the American navy dominates the passage, China worries America and its allies could deny China access to it in a time of crisis or war. Malaysia was not traditionally a significant recipient of Chinese outward foreign direct investment. However, shortly after Najib won the 2013 election held on May 5, and the day after Xi Jinping formally announced the launch of the Maritime Silk Road Initiative in neighboring Indonesia on October 3, the Najib administration announced the First Five-Year Program for Economic and Trade Cooperation between Malaysia and China (2013–2017). Bilateral visits between the top leaders, ministers, and provincial/state chiefs of both countries became very extensive and frequent, with Najib himself visiting China eight times throughout his tenure, six times between 2016 and 2017 (Ngeow 2019). Investment from China to Malaysia increased from among the lowest levels in the region to quickly become the third-largest recipient in Southeast Asia. After the 2008 election, Chinese spending was initially directed toward transportation and power generation, followed by a large petrochemical project at Pengerang in 2011. A surge in Chinese investment into Malaysia started in 2013, as displayed in Figure 8.3, when the construction of Chinese-backed industrial parks was initiated, coinciding with a substantial increase in the percentage of government contracts awarded to foreign contractors (Todd and Slattery 2018).44 Investments were spread across a diverse range of sectors (manufacturing, infrastructure, real estate, telecommunications and other services) and included varying types of collaborations between private and public firms from both the China and Malaysia sides, as well as standalone investments by Chinese public and private firms into Malaysia (Gomez et al. 2020). But the largest projects were typically overseen directly by the federal 43

“The Malacca Dilemma: A hindrance to Chinese Ambitions in the 21st Century,” August 26, Berkeley Political Review, https://bpr.berkeley.edu/2019/08/26/the-malaccadilemma-a-hindrance-to-chinese-ambitions-in-the-21st-century/ (accessed August 4, 2021). 44 Notably, this increase did not occur at the expense of local contractors based on historical trends.

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

Case Studies of Political Regimes and the BRI

257

Millions of USD

20 15 10 5

Investment

Construction

Combined

Projects Initiated

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

0

Number of projects initiated

25

10,000 9,000 8,000 7,000 6,000 5,000 4,000 3,000 2,000 1,000 0

Figure 8.3  Chinese foreign spending into Malaysia, 2005–201945

government, continuing the established pattern of Malaysia’s previous megaprojects. Chinese megaprojects initiated in Malaysia include the Malaysia-China Kuantan Industrial Park (MCKIP), Kuantan Port’s expansion, the East Coast Rail Link (ECRL), Melaka Gateway, GemasJohor Bahru Rail, KL-Singapore High-speed Rail, Xiamen University Malaysia, Kuala Linggi International Port, Trans-Sabah Gas Pipeline, and the Multi-Product Pipeline. All were initiated between 2013 and 2017, ranging in cost from USD300 million (for Xiamen University) to USD20 billion (ECRL). The major Chinese investor in all these projects was a Chinese SOE. Figure 8.3 also displays a dramatic decline in 2018 and 2019, coinciding with the first loss of the federal election by the ruling party coalition in over six decades. This led to cancellation of the two oil pipelines and the renegotiation of the ECRL by the new Mahathir administration due to corruption. However, Mahathir expressed strong support for attracting new Chinese investments into Malaysia and supported the continuation of other Chinese projects in the country, which is reflected by the upward spike in the projects initiated line in 2019.46 Chinese investment into megaprojects that were part of a Port-ParkCity model nicely matched Malaysia’s existing project management and patronage system. Indeed, UMNO welcomed Chinese investment as a means to feed the politically important construction sector and enhance the delivery of patronage in advance of the 2018 election which was 45

See Table 8A.3 for the data used to construct this figure. 46 “Mahathir seeks to attract Chinese investment,” August 19, 2018, Global Times, www​ .globaltimes.cn/content/1116133.shtml (accessed August 4, 2021).

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

258

China’s Chance to Lead

certain to be a highly competitive battle based on UMNO’s declining support in the 2008 and 2013 elections. Indicative of their desire for Chinese spending, all of the major Chinese-backed projects in Malaysia were initiated by Malaysian government-linked actors (Jones and Hameiri 2020). Although the Malaysian government established a PPP unit in 2009, the Privatization and Private Finance Initiative Unit, to bolster its support for private partners, PPPs have remained heavily under government control. As a reflection of this, the World Bank Benchmarking Infrastructure (2020) study reports Malaysia achieved the following scores in relation to electoral autocracies, in parentheses: Preparation 20 (38), Procurement 23 (56), Contract Management 31 (63), Unsolicited Proposals 0 (51). These sizeable deviations from the average scores for electoral autocracies (or even from closed autocracies) demonstrate the substantial importance of political control to PPPs in Malaysia, especially since the volume of PPPs has increased alongside substantial increases in Chinese foreign spending. As seen in Figure 8.3, Malaysia also exhibits a dramatic increase in Chinese foreign spending following the introduction of the BRI (Chapter 6). The avid welcoming of BRI spending is indicative of general patterns observed concerning differences between electoral autocracies and other regimes, as seen with PPPs and a range of related correlates in Table 7.3. Malaysia also illuminates the dominant role of host country SOEs in electoral autocracies as compared with other regimes, as seen in Table 7.3. But Malaysia is especially useful in illustrating differences between weak versus strong electoral autocracies (Djibouti), and the relatively more aggressive tactics rulers of weak electoral autocracies will use to attract Chinese spending, as discussed and examined in Chapter 6 (Table 6.2, for example). This includes the initiation of more projects in the context of PPPs, the use of SOEs in this effort and a higher completion rate than strong electoral autocracies. To illuminate these patterns in the context of a specific project, I turn to an examination of a PortPark-City initiative, the Kuantan Port and the Malaysia-China Kuantan Industrial Park. 8.5.3

Kuantan Port and the Malaysia-China Kuantan Industrial Park (MCKIP)

Chinese commitments to build the MCKIP and upgrade the neighboring port were launched in 2013. The location of the port and park serves both political and economic development objectives. Politically, it is located in Najib’s home state of Pahang (Yean and Negara 2020). The

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

Case Studies of Political Regimes and the BRI

259

project was launched in February 2013, just prior to the national elections in May. The park is Malaysia’s first to be granted national status, granting it access to federal funding as part of the government’s East Coast Economic Region initiative.47 Economically, it has the potential to substantially benefit Malaysia’s less developed East Coast Peninsula, and is located in Malaysia’s ninth largest city. Kuantan presented an ideal location for achieving both Chinese and Malaysian objectives. It takes three days shipping time to reach China Beibu Gulf Port, the fastest shipping time between Malaysia and China. The project consisted of constructing a New Deep Water Terminal to accommodate ships with five times greater capacity than originally available, and included the construction of the MCKIP in order to provide for the needed industrial activities to stimulate demand for cargo through the port. Kuantan is also planned to be a stop on the East Coast Rail Line once constructed. The ECRL connects Port Klang on the west coast of Peninsular Malaysia to Kuantan Port on the east coast, and will continue up to Northeastern Malaysia.48 The projects reward both elites, via financial rewards with government contracts and new business opportunities via China’s projects, and the people via jobs and subcontracting work. In addition to matching UMNO’s preexisting patronage delivery model via large construction megaprojects, they also matched China’s aspirations for “Port-Park-City” development plans. This is reflected in the ownership structure of the projects and the provision of jobs and subcontracting opportunities. Kuantan Port was originally built and operated by the federal government in 1984, later privatized to Kuantan Port Consortium in 1998, and subsequently acquired by IJM Corporation in 2008. The Kuantan Port Consortium (KPC) is owned by the IJM Corporation and Hong Kong’s Beibu Gulf Holding (BGH) on a 60:40 equity split. It is a public-private partnership with a concession of the Kuantan Port for thirty years plus another thirty years subject to fulfillment of certain obligations by the KPC since 2015. BGH belongs to China’s Guangxi Beibu Gulf International Port Group, an SOE involved in the construction and operation of ports, railways, and roads. The Malaysian government retains a golden share in the KPC allowing it a veto on decisions that impact the national interest. IJM is one of Malaysia’s top ten construction companies, and 47

This contributed to the MCKIP receiving RM10.5 billion in new investments and 8,500 job opportunities (Gomez et al. 2020). 48 For more details about the ECRL, see the discussion in the Background on BRI Projects, Political Opposition, and Clientelism section.

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

260

China’s Chance to Lead

although it is a private company, its list of substantial shareholders includes several GLICs that collectively own more than 48.8 percent of the equity as of March 2021.49 With regard to the Kuantan Port expansion, the construction costs were paid for by the Malaysian government (RM1 billion) (Hutchinson and Yean 2021). Because IJM is a construction company, KPC was responsible for the dredging, reclamation, and construction of the berth and terminal facilities under the supervision of IJM, with technical consultation provided by the Chinese partner. Work on the port, including the breakwater, was completed in 2016. With regard to the MCKIP, a joint venture was formed between Malaysia (51 percent) and China (49 percent) to develop it. On the Malaysia side, Kuantan Pahang Holding Sdn. Bhd. (KPH), which is a public-private partnership from Malaysia, holds 51 percent equity share in the joint venture. KPH, in turn, is owned by IJM Land Bhd., a subsidiary of IJM Corporation (40 percent), Sime Darby Property (30 percent) and the Pahang State Government (30 percent) (Hutchinson and Yean 2021). Sime Darby is a Malaysian GLC. This group of Malaysian owners represents a mix of private, federal government-owned and state of Pahang-owned firms. The engagement of Pahang state is important as land acquisition is under the jurisdiction of state governments in Malaysia. The China consortium consists of two provincial state-owned enterprises (SOEs), namely China Guangxi Beibu Gulf International Port Group (CGBGIP) (95 percent) and Qinzhou Jinqu Investment Company Ltd (QJIC) (5 percent). QJIC, was established in 2012, wholly owned by the Qinzhou City Development and Investment Group Co Ltd, and it is owned by the Qinzhou City Government (Khor 2013). It is involved in the operation and management of state-owned construction assets. Qinzhou is the location of MCKIP’s sister industrial park in China, launched in collaboration with Malaysian firms.50 In addition to financing from the Malaysian federal government, financing for the MCKIP was also provided by China Development Bank Corporation. IJM also borrowed from China Construction Bank (Malaysia) Bhd., which offered a lower interest rate and was granted a 49

The shareholders include: the Employees Provident Fund Board (EPF, 16.7 percent); Kumpulan Wang Persaraan (KWP, 9.15 percent); Permodalan Nasional Bhd. (PNB, 6.9 percent); Urusharta Jamaah Sdn. Bhd (UJ, 6.15 percent), Public Mutual Bhd (PMB; 4.07 percent); AIA Bhd (4.03 percent); and Bumiputera Investment (1.81 percent). 50 The sister park is the China-Malaysia Qinzhou industrial Park (CMQIP) in Guanxi. Source: https://chinareportasean.com/2019/07/16/sustaining-sibling-partnership-chinamalaysia/ (accessed August 4, 2021).

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

Case Studies of Political Regimes and the BRI

261

banking license by Malaysia in 2017 to facilitate financial cooperation between the two countries. The financial package included the development of the primary infrastructure in MCKIP, such as the construction and upgrading of a direct road linking MCKIP to the Kuantan Port, as well as the necessary water, electrical and telecommunications facilities to support the development of the park (Gomez et al. 2020). MCKIP provides financial incentives for Chinese businesses to locate their facilities there (Gomez et al. 2020), and as of 2021 MCKIP has more than 13 committed companies with a total investment exceeding RM18 billion (RMB 30 billion).51 Alliance Steel was the first company to initiate manufacturing operations in the park, and is expected to export 80 percent of its output contributing to an increase in cargo throughput and profitability for the port (Yean and Negara 2020). It is expected to create 20,000 jobs for the locals in the area. Due to strong political opposition to Malaysia’s incumbent rulers, there have been numerous vocal attacks directed at Chinese projects, with specific criticisms varying by project. Major critiques have focused on the lack of opportunities for Malaysian workers and businesses, a lack of local stakeholder consultation, excessive debt, and corruption. A primary concern regards the use of foreign workers. Media reports indicated that 1,000 Chinese workers were being used compared to 2,600 local workers at Alliance Steel. This yields a ratio of foreign to local of 0.38 which is two times higher than the national ratio for the manufacture of basic iron and steel in 2016 (Gomez et al. 2020). However, Alliance Steel has stated that the acquisition of skills by local workers through training and technology transfer programs will contribute to an increase in the number of local workers over time. Additionally, investors in the MCKIP are required to provide training for Malaysian workers employed to operate the equipment and machinery used at the new plants, as in the case of Alliance Steel. Through these foreign 51 Alliance Steel was the first company to locate in the park. Additional companies include the Zhongli Group www.360powder.com/info_details/index/1967.html; Hunan Zhongke Hengyuan Technology Co., Ltd. www.chinanews.com/gj/2016/0327/7813281.shtml; Hangzhou Hangyang Group www.chenhr.com/company/NdfUM/; Guangxi Investment Group www.chinanews.com/gj/2016/03-27/7813281.shtml; Sundi Tire Manufacturing Malaysia Co., Ltd. (Sindi Tire) www.mckip.com.my/215/; NewOcean Energy www.mckip.com.my/28/; ND Paper (M) Co., Ltd www.mckip​ .com.my/270/; TNP Power Sdn Bhd(Cogeneration, combined heat and power, CHP) www.mckip.com.my/270/; Meixian China Travel Service Co., Ltd. www.mckip.com​ .my/270/; CITIC Resources Recycling Technology Pte Ltd www.mckip.com.my/270/; Wuxi Jinxie Industrial Co., Ltd. www.mckip.com.my/270/; China Power Construction Group www.mckip.com.my/270/. All links accessed August 4, 2021.

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

262

China’s Chance to Lead

investments, Malaysia aims to gain from technology transfer, as well as through domestic sourcing of component products from Malaysian companies.52 With regard to opportunities for small or local firms, various factors can lead Chinese firms to favor Chinese contractors, such as the need for specific skills, language, capacity (such as possessing the necessary equipment) to complete the project, political direction (implied or explicit) from the Chinese government to maximize economic opportunity for Chinese nationals, and financing availability for larger (typically Chinese) contractors. To ensure Malaysian firms benefit from these projects, in most cases the Malaysian government has reportedly negotiated specific clauses that reserves work for local subcontractors (typically around 30 percent) (Todd and Slattery 2018). This is a critical role by which host governments deliver patronage to their constituencies via foreign investment and construction projects. In Algeria, another electoral autocracy, the government provides a list of contractors that Chinese firms should work with and rewards the company and its project for doing so.53 Nevertheless, local entrepreneurs complained about the lack of opportunities, leading many to switch allegiance to the opposition party.54 Criticisms about a lack of consultation with local stakeholders gained traction with regard to the ECRL’s route which was initially planned without consulting local jurisdictions and communities through which it would be built. A lack of consultation led to a backlash with regard to the Forest City project, with Chinese nationals accounting for about 70 percent of apartment buyers, producing a property market glut, and land reclamation activities that negatively impacted the local fishing industry. And local residents expressed grave concerns about environmental damage and negative health consequences for local residents due to the numerous manufacturing facilities being constructed in the MCKIP.55 Dozens of unpermitted bauxite mines were recently involved in transporting red clay-like ore to the Kuantan port for export to China, producing streams that ran red and red dust that covered homes, cars, and 52

This technology transfer remains to be seen as of the end of 2019 (Gomez et al. 2020). 53 Interview with Chinese SOE employee on March 19, 2021. 54 “Selling the country to China? Debate spills into Malaysia’s election,” April 27, 2018, Reuters, www.reuters.com/article/us-malaysia-election-china/selling-the-country-tochina-debate-spills-into-malaysias-election-idUSKBN1HY076 (accessed August 4, 2021). 55 “Modernizing Malaysia: Port City See Pitfalls of Chinese Investment,” September 13 2018, The News Lens, https://international.thenewslens.com/article/103999 (accessed August 4, 2021).

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

Case Studies of Political Regimes and the BRI

263

businesses with potentially severe long-term health and environmental consequences.56 But the most damaging criticisms for the ruling party, and Najib in particular, focused on excessive debt and corruption. The many costly mega-projects largely financed by China created risks for Malaysians in the long term not only due to the cost of repaying the loans with interest but also due to the potential vulnerability of being so indebted to a single country, especially if the projects were not completed (Ngeow 2019).57 Even more damaging were the revelations regarding the misuse of funds intended for BRI projects. To help pay off 1MDB’s debts, the price tags of several Chinese projects (the East Coast Rail Link and two oil pipelines) were deliberately inflated in order for the Chinese SOE contracted to build them (CCCC) to funnel money through a shell company to help repay 1MDB’s debts (Brown 2018).58 As evidence piled up verifying these claims, Najib’s re-election prospects dwindled. These events clearly illustrate the desire on the part of a weak electoral autocrat to tap Chinese BRI funds to defend his hold on power. 8.5.4 Summary Malaysia is an important partner to China and the Belt and Road Initiative for at least three reasons. First, it neighbors the Melaka Strait which is a critical choke point for the passage of 80 percent of Chinese energy resources. Second, it borders a large portion of the South China Sea. Third, it offers an attractive location for Chinese business, both with regard to traditional BRI projects and increasingly as a partner in the Digital Silk Road which includes 5G technology, artificial intelligence, and their integration into smart cities projects. Malaysia stands out as one of the most enthusiastic participants in the BRI and the adoption of Chinese digital technologies which will be examined in Chapter 9. Malaysia’s interest in China has been primarily driven by its political leaders’ desire to access funds for the building of infrastructure, industrial parks, and local business opportunities. The aggressive courting of Chinese investment since 2013 was motivated by the ruling party’s, and 56

“Kuantan facing severe danger,” January 3, 2016, New Straits Times, www.nst.com​ .my/news/2016/01/120276/kuantan-facing-severe-danger (accessed August 4, 2021). 57 “China investments put Malaysia at risk, warn MP, economist,” August 25, 2017, The Malaysian Insight, www.themalaysianinsight.com/s/12213 (accessed August 4, 2021). 58 It is estimated the price was inflated from USD7 billion to USD14 billion, making it 120–140 percent costlier than comparable projects. “PM must explain if rail project inflated to pay 1MDB, MP says,” July 28, 2016, Malaysiakini, www.malaysiakini.com/ news/350223 (accessed August 4, 2021).

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

264

China’s Chance to Lead

especially the prime minister’s, weak hold on power. BRI projects offer a tremendous source of patronage that can be directed to the ruling party’s clientelist network and used as inducements to attract new supporters. The history of large construction projects in Malaysia as sources of patronage create an ideal platform to exploit China’s BRI opportunities for political gain. As a result, the interests of China and Malaysia are aligned, though Malaysia’s leaders are more aggressive in their demand for Chinese investment and construction projects due to their short-term electoral pressures. In contrast to Djibouti, the political opposition is much stronger in Malaysia. This has contributed to the ruling party’s aggressive pursuit of Chinese funding and projects. This can be seen by the central role of the federal government in rapidly initiating numerous BRI projects. The ruling party’s electoral pressures have contributed to extensive use of PPPs to increase the number of projects within the constraints of the government’s finite resources. At the same time, the strength of the political opposition has contributed to far louder criticisms of these projects than in Djibouti, including a lack of opportunities for local Malaysian businesses, workers, environmental effects, and corruption. The loss of the 2018 national election, the first in over five decades for the ruling party, demonstrates the legitimacy of their concerns and explains the intensity of the government’s efforts in attracting Chinese spending. 8.6

Electoral Democracy: Indonesia

8.6.1

Political Context

Following the end of Suharto’s rule in 1998, Indonesia implemented democratic reforms. The election of Joko Widodo (Jokowi) to president in 2014 seemed to affirm Indonesia’s credentials as a democracy in three ways. First, his political rise occurred by winning competitive elections following the introduction of democracy; second, he defeated a Suhartoera general who would have shifted the regime in an authoritarian direction; third, the post-election transfer of power occurred peacefully. Yet, Indonesia’s democratic institutions remain fragile and problematic due primarily to its clientelist system of politics (Aspinall and Hicken 2020; Berenschot and Aspinall 2020). Whether through their funding of expensive political campaigns or through influence on courts, regulators, or politicians, oligarchs wield tremendous influence. Added to this is the sharpest polarization that Indonesia has seen since the 1950s between religious pluralists and proponents of a stronger role for Islam.

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

Case Studies of Political Regimes and the BRI

265

The 2019 election revealed “a hardening of a religious cleavage across the country: Prabowo’s campaign appealed to Muslims, and Jokowi’s to nonMuslims” (Pepinsky 2019). These two characteristics of Indonesian politics have contributed to its placement in the electoral democracy category rather than liberal democracy for three reasons. First is corruption of the judiciary. While judges are relatively well protected from political interference, corruption has been a persistent problem with judges from even the highest levels being convicted of taking bribes to influence court rulings. In 2017 alone, the country’s Judicial Commission recommended fiftyone judges undergo further investigation.59 Additionally, there is growing concern about politics influencing judicial decisions as religious polarization dominates elections and with judicial decisions being influenced by religious considerations. For example, in March 2019 the Supreme Court rejected the appeal of a Buddhist woman who received an eighteen-month prison sentence for blasphemy after she privately complained about the volume of a nearby mosque’s call to prayer.60 The second problem is corruption of elected officials and those working in public administration positions. In the Parliament’s 2014–2019 term of office, for example, around two dozen legislators were charged with corruption, including the former Speaker (Russell 2020). In elections, vote buying is commonplace (Aspinall and Mietzner 2019). Although there is little firm evidence of vote rigging, voters often remain suspicious of the results, as witnessed by violent protests by supporters of the defeated candidate (Prabowo) after Jokowi’s 2019 election victory. According to a 2018 survey by Charta Politika, fewer than half of Indonesians trust the Supreme Court, the Parliament, and political parties.61 Since 2003, the Corruption Eradication Commission (KPK) has led the fight against corruption, but the Parliament voted for a law to water down its powers which Jokowi passed into law in 2019. The third problem is the Jokowi government’s crackdown on opposition figures that is unprecedented in Indonesia’s history as a democracy since 1998. The election marked a decoupling of the politics of religious pluralism from those of democracy. Jokowi criminalized the most extreme figures in the Islamist mobilizations of 2016 and targeted 59

www.state.gov/wp-content/uploads/2019/03/INDONESIA-2018.pdf (accessed August 4, 2022). 60 “Indonesia Supreme Court Rejects Appeal from Woman Convicted of Blasphemy,” April 10, 2019, VOA News, www.voanews.com/east-asia-pacific/indonesia-supremecourt-rejects-appeal-woman-convicted-blasphemy (accessed August 4, 2021). 61 “Pollster: TNI Gains Indonesian Citizens’ Highest Trust,” Tempo.co, August 28, 2018, https://en.tempo.co/read/921167/pollster-tni-gains-indonesian-citizens-highesttrust (accessed August 4, 2021).

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

266

China’s Chance to Lead

opposition figures involved in the Change the President protests leading up to the 2019 election. Such figures were charged with treason, corruption, and the spreading of pornographic images. Jokowi then circumvented proper legal processes to ban Hizbut Tahrir Indonesia (HTI), whose radical Islamists were involved in protests against a non-Muslim political opponent. In doing so, Jokowi introduced a presidential regulation, in lieu of a law, that gives immense authority to the executive to ban groups it deems to have contravened the nation’s Pancasila (religious tolerance). After winning re-election in 2019, the Jokowi government then encouraged a purge of Islamist elements from state agencies and several prominent opposition figures were threatened with criminal charges for criticizing the government’s pandemic response.62 A final challenge for Indonesian democracy stems from the growing presence of the military in politics. To break free of the military’s political grip, the post-1998 reforms took away the parliamentary seats reserved for army representatives and barred active security forces personnel from voting or standing in elections. However, Jokowi’s 2019 government prominently features several former generals, and since April 2019 active military officers can be promoted to civil service positions. Following his presidential election victory, Jokowi named Prabowo as defense minister. Prabowo, a one-time son-in-law to Suharto, held high military rank during the final years of Suharto’s authoritarian regime (1966–1998), is known as a heavy-handed enforcer with a record of violating human rights.63 8.6.2

Chinese Foreign Spending and the Belt and Road Initiative

China’s interest in Indonesia centers on the country’s substantial infrastructure opportunities, natural resources, and strategically important location next to the Melaka Strait and South China Sea. Following the introduction of democracy in 1998, Indonesia became a major natural resources exporter to China. Meanwhile, Chinese development finance and foreign investment grew slowly during the first years of Indonesia’s democratic rule. The country faced an enormous gap in the amount of investment needed to sustain its economic development, with the government budget deficit restricted to no more than three percent of GDP due to new laws implemented following the Asian Financial Crisis (Damuri et al. 2019). Susilo Bambang Yudhoyono’s (SBY) regime (2004–2014) 62

See Aspinall and Mietzner (2019) for discussion of historical reasons for this division. 63 www.lowyinstitute.org/the-interpreter/indonesia-s-new-cabinet-and-its-human-rightsimplications (accessed August 4, 2021).

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

Case Studies of Political Regimes and the BRI

267

worked primarily with Indonesian oligarchs to promote infrastructure development via Public Private Partnerships (Camba 2020). Often however, local businesses with the requisite expertise and experience to complete large, complex, and long duration infrastructure projects were unavailable. SOEs can address some of these difficulties, but they may also affect the government’s fiscal position. Consequently, foreign investment offers an attractive solution. Partly for this reason, investment from China increased during SBY’s presidency, especially in utilities (electricity, gas, and water supply), and mining projects. Although many of these projects involve Chinese companies bringing their own workers (e.g., Surabayan Madura Bridge, power plants in Kalimantan, and nickel smelters in Southeast Sulawesi),64 political opposition groups did not voice objections, in contrast to Jokowi’s presidency (2014-present). This muted reaction was due, in part, to a broad support coalition including Islamic parties in addition to continuing tolerance from radical Islam at the time (Suryadinata 2017). Although Chinese investment was welcomed, there has remained a distinct preference for investments by private business where possible (Damuri et al. 2019). For example, Indonesia’s Finance Minister stressed the government’s desire for more private sector involvement in infrastructure development, with the expectation that the private sector would fund nearly half of the estimated total USD430 billion over the 2019–2024 time period.65 The finance minister also announced that state enterprises had to divest infrastructure assets built in the past five years or make them tradable in order to open them to private sector participation. This policy statement is reflected in Indonesia’s substantially stronger PPP framework relative to Malaysia’s, which offers confidence to private investors. According to the World Bank Benchmarking Infrastructure (2020) report, Indonesia’s PPP scores in relation to electoral democracies (in parentheses) are: Preparation 51 (51), Procurement 65 (67), Contract Management 63 (69), Unsolicited Proposals 58 (62). A notable reflection of this preference for private sector involvement is the Indonesia Morowali Industrial Park (IMIP) which is one of the largest Chinese investments in Indonesia to date and was initiated by private firms from China and Indonesia. In 2009, the Shanghai Decent Group of China together with Sulawesi Mining Investment (SMI) of Indonesia invested in nickel ore mining and export. Due to government restrictions 64

See Suryadinata (2017) for details about these projects. 65 “Indonesia government wants more private sector investment in infrastructure: minister”, The Business Times, Nov 26 2019. www.businesstimes.com.sg/government-econ​ omy/indonesia-government-wants-more-private-sector-investment-in-infrastructure (accessed August 4, 2021).

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

268

China’s Chance to Lead

9,000

Millions of USD

7,000 6,000 5,000 4,000 3,000 2,000 1,000

09 20 10 20 11 20 12 20 13 20 14 20 15 20 16 20 17 20 18 20 19

08

20

07

20

06

20

20

20

05

0

Number of projects initiated

20 18 16 14 12 10 8 6 4 2 0

8,000

Investment Combined

Construction Projects Initiated

Figure 8.4  Chinese foreign spending into Indonesia, 2005–201966

on the export of mineral ore, their activities later switched to ferronickel smelting industries. In 2014 the Shanghai Decent Group and the owner of SMI, the Bintang Delapan Group, established the IMIP as a hub for nickel-based industries (Indonesia has around 24 percent of the world’s nickel reserves). Ownership is split 66.25 percent for Shanghai Decent and 33.75 percent for Bintang Delapan. By its expected completion date in 2024, total investments in the park are projected to reach USD29 billion. The park has access to land, sea and air transportation and provides tenants with supporting infrastructure facilities such as electricity and telecommunications.67 As of 2020, the park employed 35,000 Indonesian workers plus tens of thousands more indirectly through local companies that supply and serve companies in the park (Yean and Negara 2020). By 2013, when Xi made his public announcement about launching the MSRI while visiting Indonesia, China had already become Indonesia’s top trading partner. As reflected in Figure 8.4, China’s foreign direct investment (FDI) into Indonesia jumped from no. 9 in 2010 (US$0.17 billion) to no. 2 in 2020 (US$4.8 billion), though it still lags far behind Singapore’s (US$9.8 billion). This dramatic increase reflects 66

See Table 8A.4 for the data used to construct this figure. 67 Shareholders of IMIP include: Tsingshan Group, GEM Co Ltd, Contemporary Amperex Technology Ltd (CATL), Shareholders: PT Bintang Delapan Investama; Shanghai Decent Investment (Group) Co Ltd. For more information, see “Indonesia Morowali Industrial Park: how industrial policy reshapes Chinese investment and corporate alliances,” January 17, 2021, https://pandapawdragonclaw.blog/2021/01/17/ indonesia-morowali-industrial-park-how-industrial-policy-reshapes-chinese-invest​ ment-and-corporate-alliances/ (accessed August 4, 2021).

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

Case Studies of Political Regimes and the BRI

269

Jokowi’s efforts to boost infrastructure development by turning to foreign assistance since becoming president in 2014 (Suryadinata 2017). China, Japan, and Australia have played the leading roles, with China the most active among them. Consistent with the preference for investment by private firms, the Indonesian partners in Chinese projects are predominantly private firms that are affiliated with one of Indonesia’s family-owned conglomerates (Negara and Suryadinata 2018).68 The top industries attracting Chinese foreign investment include mining, energy and utilities (electricity, gas, and water supply), and transportation (Damuri et al. 2019; Yean and Negara 2020). The dramatic rise in Chinese spending reflects the electoral pressures of leaders in electoral democracies, and the substantial political benefits associated with attracting Chinese foreign spending, including patronage to the ruling party’s clientelist network and promoting growth. By privileging private business over SOEs, as in electoral autocracies, Indonesia displays characteristics consistent with those of electoral democracies more generally, as seen in Tables 5.6 and 6.1. Indicative of the more hands-off approach by the government, President Jokowi only visited China four times between 2014 and 2020, in comparison to Najib’s eight times as prime minister of Malaysia (six times between 2016 and 2017). To provide a more focused analysis of how Indonesia’s political regime impacts BRI projects, I turn to the examination of a Chinese SOE-led project, the Jakarta-Bandung HSR. 8.6.3

Jakarta-Bandung High-Speed Rail Project

The Jakarta-Bandung HSR project illuminates that leaders of electoral democracies face similar electoral pressures to weak electoral autocrats, but produce different outcomes. Shortly after winning the 2014 presidential election, Jokowi proposed a massive infrastructure plan known as the “Global Maritime Fulcrum” (GMF) that consisted of hundreds of projects with the aim of boosting growth.69 Jokowi immediately faced a series of challenges to implementing his plan, including rising discontent as he slashed fuel subsidies to pay for new infrastructure projects, coupled with the depreciation of the Rupiah. Jokowi became anxious to secure a quick deliverable to counter his plummeting approval ratings. 68

Negara and Suryadinata (2018) provide a list of 50 of the largest projects with the identities of the Chinese and Indonesian companies involved in each project. 69 “Jokowi’s Global Maritime Fulcrum: 5 More Years?” June 11, 2019, The Diplomat, https://thediplomat.com/2019/06/jokowis-global-maritime-fulcrum-5-more-years/ (accessed August 4, 2021).

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

270

China’s Chance to Lead

Soon after entering office, Jokowi visited China to attract support for the GMF plan. Beijing pledged support and expressed strong interest in the Jakarta-Bandung HSR project as an opportunity to promote their newly developed HSR capabilities and agreed to a currency swap to prop up the Rupiah. Although the HSR project was already awarded to Japan, the Jokowi government invited China to conduct a feasibility study. In just three months, China responded with a proposal that was faster and at lower cost than what Japan proposed (Liao and Katada 2021). After further negotiation with Japan and China, Jokowi awarded the contract to China due, in large part, to their new suggestion of a contract between SOEs of the two countries to help circumvent Indonesia’s debt limit (Salim and Negara 2016). The HSR thus became the first GMF project to begin construction, starting in 2016. Numerous Chinese firms were involved in the HSR’s construction, with Indonesian firms also participating. On both sides, they were predominantly SOEs. Notably, and in contrast to the other Chinese SOEled projects examined in other cases, firms from other countries were also involved such as HSBC and the Japan International Cooperation Agency (JICA) as consultants.70 From an electoral strategy perspective, the HSR project was very popular in Bandung, the capital of West Java and the country’s most populous province, which Jokowi had lost to his opponent in the presidential election. The HSR project was soon followed by more than 30 additional infrastructure projects directed to West Java (Liao and Katada 2021). Both Chinese and Japanese funding was secured, contributing to a major boost in Jokowi’s approval ratings. However, the Jakarta-Bandung HSR encountered two major obstacles to its speedy implementation. The first major problem related to the appropriation of land. The government’s land payments were capped, producing substantial delays as property owners contested the government in courts (Camba 2020). President Jokowi implemented an executive order in 2015 that would speed up land appropriation with better financial terms, expedite court hearings, and allow the government to 70

PT Kereta Cepat Indonesia-China was the EPC contractor and the developer of the Jakarta-Bandung High Speed Rail project. PT Kereta Cepat Indonesia-China consists of a consortium of SOEs named PT Pilar Sinergi BUMN Indonesia, led by a construction company, PT Wijaya Karya (Persero) Tbk. and a consortium of Chinese companies, named Beijing Yawan HSR Co Ltd. Gemac Engineering Machinery Co Ltd (Chinese firm) supplied JW-4G catenary operation cars. HSBC Bank plc and Japan International Cooperation Agency were the consultants of the project. China Development Bank was the financial investor of the project. Pangang Group Co Ltd (Chinese firm) was the steel supply Contractor. China Railway Materials Co was an equipment supplier.

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

Case Studies of Political Regimes and the BRI

271

call upon the military to enforce the policy. The executive order led to 95 percentof Java landowners agreeing to leave; the remaining 5 percent were forcibly evicted, but the projected completion date was pushed back by two years, to 2021. The second major challenge related to political opposition is the increasingly polarized electorate. Prabowo Subianto, Jokowi’s rival in the recent presidential election and leader of the main opposition party, publicly criticized Jokowi for awarding the project to the Chinese. He attacked Jokowi for kowtowing to the Chinese and criticized China’s use of “illegal Chinese workers” and “debt trap” diplomacy. Long-simmering anti-Chinese sentiment rooted in the Suharto era, the rise of identitybased politics in Indonesia (Aspinall and Mietzner 2019) coupled with the substantial increase in Chinese foreign investment and reminders of excessive foreign debt prompting the financial crisis of 1997 converged to generate a popular backlash to Chinese investment and Chinese workers in particular. Additionally, Indonesia is struggling to create more than two million jobs annually to accommodate new entrants into the labor market and, hence, wants to limit the entry of foreign workers (Damuri et al. 2019). Prabowo also encouraged civil society organizations (CSOs) to oppose the HSR. This tapped into discontent both from Japanese firms and Indonesian SMEs. In Java, there are dozens of industrial parks with Japanese firms making up 80–85 percent of the companies that would be affected by the HSR project (Dwiatmoko et al. 2018). These firms paid for CSOs to mobilize and “delay Jakarta-Bandung until the next election, which might create an opportunity for the project to get canceled if the opposition wins.” And although the Indonesian government has been actively promoting SME involvement in all economic activities, problems often emerge with finding companies capable of completing tasks the main contractors demand (Damuri et al. 2019). This has contributed to further accusations that the government favors SOEs to speed up the completion of large projects, denying opportunities to Indonesian private firms.71 Although some conglomerates initially considered the project via a PPP during SBY’s government, the oligarchs opted against pursuing it themselves, instead focusing on the downstream industries that accompanied the HSR. Concerns about the HSR also emerged with regard to its environmental impact. The environmental impact assessment was completed in just seven days, fueling further stakeholder and CSO complaints. The 71

www.salaki-salaki.com/update-1-indonesian-parliamentary-commission-approves-taxamnesty-bill/ (accessed August 4, 2021).

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

272

China’s Chance to Lead

project developer was forced to make a major rerouting of the rail terminus, raised the project budget from $5.5 billion to $6.1 billion, similar to Japan’s original offer, and more than doubled the proposed ticket fares to absorb cost overruns. Due to numerous delays, Indonesian president Joko Widodo ordered a reevaluation of the Jakarta–Bandung Railway’s design and feasibility in January 2018. The review revealed a worrisome lack of due diligence and described the project as deeply troubled.72 Political attacks, cost overruns, and admission of the project as seriously flawed, led protestors to block the construction of the HSR (Camba 2020), prompting the Jokowi government to mobilize the police to arrest them and jail some of the instigators. In March 2019, the construction of the line temporarily stopped due to elections. While project construction resumed following Jokowi’s 2019 reelection, the Jokowi government asked Japan to join the project and integrate it into the JIBA-led Jakarta-Surabaya rail project as the only economically viable option to address the mounting costs.73 8.6.4 Summary China’s interest in Indonesia with regard to the BRI centers on two main areas, natural resources and infrastructure development. Its relative proximity to China and large, but underdeveloped, economy offer enormous opportunities for Chinese businesses. Indonesia’s interest in Chinese foreign investment is primarily to help it build much-needed infrastructure, and in turn create business opportunities and jobs. But aside from the Jakarta-Bandung HSR project, Indonesia has displayed far less direct government involvement in the implementation of BRI projects compared to its autocratic counterparts. Indonesia’s electoral democracy political system yields stronger political challengers than typically found in electoral autocracies, forcing political incumbents to respond to sharper and more sustained attacks and to be more responsive to a wider range of stakeholders. This contributes to the privileging of private business and a reduction in the

72

“Jokowi Demands Evaluation of Jakarta-Bandung Fast Train Project after Setbacks,” Jakarta Globe, 10 January 2018, https://jakartaglobe.id/context/jokowi-demands-evalua​ tion-jakarta-bandung-fast-train-project-setbacks (accessed August 4, 2021); and Karlis Salna and Arys Aditya, “Indonesia May Be Next Asian Country to Spurn China in Election,” Bloomberg, 31 March 2019, www.bloomberg.com/news/articles/2019-03-31/ indonesia-may-be-next-asian-country-to-spurn-china-in-election (accessed August 4, 2021). 73 www.scmp.com/week-asia/opinion/article/3092487/indonesia-moving-towards-japanand-away-china-just-follow-railway (accessed August 4, 2021).

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

Case Studies of Political Regimes and the BRI

273

reliance on SOEs wherever private business is capable of and willing to complete the project. Overall, Indonesian BRI projects rely on a relatively more broad-based, private sector-oriented support network (typically via reliance on oligarchs). This contributes to BRI projects that are more tailored to decentralized, private sector participation rather than centralized state-led implementation, as in Malaysia. Nevertheless, the short-term electoral pressures coupled with electoral democracies’ weak institutional checks on executive power can lead to the government promoting infrastructure opportunities while potentially generating longerterm problems by rushing due diligence processes. But the heightened political competition can help to address these problems through stronger scrutiny of how government resources are used, with calls to ensure the private sector benefits, local workers participate, and environmental effects and local communities’ concerns are adequately addressed. In summary, Indonesia displays the relatively greater privileging of the private sector and attention to civil society expected of a more democratic polity. While the government strongly encourages and enables the initiation of BRI projects, its direct involvement is reduced in comparison to what typically occurs in electoral autocracies and its potential excesses more effectively counterbalanced by political challengers and civil society. 8.7

Liberal Democracy: Greece

8.7.1

Political Context

Despite enduring a crisis starting in 2008 that lasted over five years and resulted in an economy 25 percent smaller than when it began, Greece’s political institutions have retained the core institutional features of a liberal democracy. Nevertheless, the magnitude of the crisis tested the country’s institutions, revealing two weaknesses linked to its authoritarian past, including clientelism and a rise in populist nationalism. The first weakness in Greece’s democratic institutions is clientelism. Prior to democratization in 1974, patrons provided clients with a range of preferential services in exchange for their vote, such as assistance with public hospitals and welfare services, convenient transfers for male conscripts during their military service, and access to bank loans (Mouzelis 1986). After the transition to democracy, clients turned to party bureaucracies. Political party offices at the local, regional and national levels acted as intermediaries between party voters and state authorities, providing party supporters with public sector jobs as well as state subsidies and welfare benefits (Featherstone 1990, Sotiropoulos 1996).

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

274

China’s Chance to Lead

Despite specific requirements by Greece’s lenders during the crisis to de-politicize public administration, political clientelism increased. The number of political appointees and the recruitment of temporary personnel to Greece’s public sector continued unabated even at a time of severe economic crisis (Sotiropoulos 2018). Similar problems occurred in relation to business, such as the offering of state subsidies, new business opportunities, or tailor-made regulations to specific business owners in exchange for their long-term political support. To combat longstanding networks of corruption among political and administrative officials, new anti-corruption agencies were established in 2011. This led to former ministers being prosecuted and tried in connection with the onset of the crisis. However, some mechanisms, such as the courts, have remained inefficient and susceptible to political influence (Papaioannou and Karatza 2018). Other anticorruption mechanisms have been too weak with regard to the resources they can mobilize to effectively combat corruption (Bratu, Sotiropoulos and Stoyanova 2017). The second problem concerns political and historical legacies of clientelism and corruption fueling a rise in populist nationalism during the Greek debt crisis. Populist nationalism weakened Greek democracy in two ways. First, it contributed to a reduction in judicial independence. For example, the Greek ruling party would publicly discredit opposition MPs and former ministers, and also attack judges if their decisions were contrary to government policy, thus putting pressure on judges to converge with government policies. Judicial appointments also became tainted by political influence under the Syriza-ANEL government (2015–19) (Sotiropoulos 2019). The second deterioration to Greek democracy due to populist nationalism concerns an intolerance of views critical of the government. This occurred with the ruling party coalition marginalizing critical voices through the control of state-owned and private media, thereby relegating anti-government views to social media (Sotiropoulos 2018). For example, in 2013, the ND-Pasok coalition government shut down the public broadcaster, ERT, due to its journalists criticizing government policies and replaced it with a new broadcaster. This was in turn dissolved by the Syriza-Anel coalition government and replaced by the previous broadcaster (ERT), which then acted as a government mouthpiece. Together, clientelism and populism have contributed to the backsliding of Greek democracy. Yet, Greece has preserved the core features of a liberal democracy that distinguish it from electoral democracies, such as Indonesia. Both Greece and Indonesia suffer from clientelism and populism, but the depth of these problems is greater in the Indonesian context, and the threats to its democratic institutions magnified by the rising influence of the military.

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

275

4,000

8

3,000

6

2,000

4

1,000

2

0

0 20

20

20

20

20

09 20 10 20 11 20 12 20 13 20 14 20 15 20 16 20 17 20 18 20 19

10

08

12

5,000

07

6,000

06

14

05

7,000

Number of projects initiated

Millions of USD

Case Studies of Political Regimes and the BRI

Investment Combined

Construction Projects Initiated

Figure 8.5  Chinese foreign spending in Greece, 2005–201974

8.7.2

Chinese Foreign Spending and the Belt and Road Initiative

When China joined the WTO and initiated its “Going Out” policy, Chinese president Jiang Zemin made the first official visit of a Chinese head of state to Greece. From that first visit in April 2000 until 2019, there have been ten high-level visits between leaders of the two countries.75 These visits have facilitated bilateral cooperation and contributed to a major inflow of Chinese foreign investment. While the first instance of Chinese FDI occurred in 2003,76 the more meaningful starting point of Chinese foreign investment is the year 2008 when COSCO Shipping (an SOE) signed a concession agreement with the Greek government for a sizeable part of the Piraeus Port, as displayed in Figure 8.5.77 To recover from the crisis, Greece actively sought foreign investment. However, not everyone favored investment from China. For example, 74

See Table 8A.5 for the data used to construct this figure. 75 George Papandreou was the only recent Greek prime minister, in office from October 2009 to November 2011, not to visit China. It is likely that the severe economic crisis that broke out in Greece during his term of office prevented him from traveling to China. 76 This is according to the Bank of Greece: www.bankofgreece.gr/en/statistics/externalsector/direct-investment (accessed August 4, 2021). 77 Bank of Greece data that includes data for both mainland China and Hong Kong indicate higher levels of Chinese foreign spending between 2016 and 2019, presumably because COSCO is headquartered in Hong Kong (CGIT data measures spending originating in mainland China and rerouted through Hong Kong, but does not capture spending originating in Hong Kong).

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

276

China’s Chance to Lead

once Syriza became the main opposition party in 2012, it voiced strong opposition to Chinese investment until prime minister Alexis Tsipras (Syriza party leader) signed the third bailout agreement with the country’s international creditors in July 2015 (following the party’s victory in the January 2015 election). Tsipras visited China in July 2016 and since then the Greek government has expressed a welcoming attitude to Chinese investment despite unwavering opposition to foreign investment from members of the Communist Party. The center-left Social-Democratic parties of PASOK (Pan-Hellenic Socialist Movement) and Potami (River) have been receptive to foreign investment, though not unconditionally. The center-right party New Democracy, the main opposition party to Syriza since 2012, has remained supportive of Chinese investment since the 2000s, reflecting the favorable views of Greek business associations (Hellenic Federation of Enterprises and the Athens Chamber of Commerce and Industry). The ultra-right xenophobic party Golden Dawn has remained silent on Chinese investment in Greece, restricting its agenda to immigration and attacking the political establishment. While China was one of the few countries to increase investments into Greece during the 2010s, the three bail-out agreements signed by Greek governments with the European Commission, the European Central Bank and the IMF between 2010 and 2015, set clear limits on the terms of investment deals. For example, the European Commission forced the Greek government to strip COSCO of some undue privileges in 2015 (Tonchev and Davarinou 2017). It is therefore necessary to view Chinese investments in Greece from the context of Greece’s EU membership. Due to stringent external oversight coupled with the need to assure external investors about the safety of investing in the country, Greece developed a strong PPP framework, particularly with regard to Procurement and Contract Management.78 The World Bank Benchmarking Infrastructure (2020) report scores Greece, in relation to liberal democracies in parentheses, as follows: Preparation 51 (50), Procurement 91 (71), Contract Management 74 (63), and Unsolicited proposals: NA (66). China’s flagship project in Greece is the investment made by COSCO in Piraeus Port exceeding 1 billion euros for a 35-year management lease starting in 2008. This “anchor investment” has helped attract additional Chinese investments in sectors beyond maritime ports and container shipping such as energy, logistics, real estate, and tourism (Tonchev and Davarinou 2017). In energy, the most notable investment is by China State 78

Procurement focuses on the strength of legal and regulatory measures to identify a private partner. Contract Management concerns monitoring and evaluation of PPP projects.

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

Case Studies of Political Regimes and the BRI

277

Grid International Development Ltd (SOE), the world’s largest electricity transmission and distribution operator, for a 24 percent stake in Greece’s Independent Power Transmission Operator in 2017. In addition, Shenhua (SOE), China’s largest coal producer, signed an agreement with the Copelouzos Group, one of the largest investment groups in Greece, for the acquisition of 75 percent of the shares of four wind parks.79 In logistics, Huawei (private) and ZTE (SOE) developed logistics hubs in Piraeus in concert with COSCO’s Consolidation & Distribution Center (PCDC); a number of other foreign corporations also started using PPA services. In real estate, Fosun International (private) was involved in a consortium led by the Greek group Lambda Development which won a public tender in 2014 for a USD8 billion project to develop Athens’ former airport at Hellenikon. However, Fosun sold their stake to Lambda in 2019 due to mounting frustrations about delays with the project coming from elements within the Syriza party. Finally, tourism has also become an increasingly attractive sector for Chinese companies. For example, Air China (SOE) started direct flights between Beijing and Athens in 2017, with onward embarkation for Chinese passengers at Piraeus Port onto cruise ships. 8.7.3

Piraeus Port

China’s strong interest in Greece and the Port of Piraeus is due to its unique position in the region, linking Europe to Asia via the MSRI. China’s strategic aim is to turn Greece into “China’s gateway to Europe.”80 The port serves as an entry point for Chinese products heading to Southeastern, Eastern, and Central Europe and as a transshipment hub for the Eastern Mediterranean. It could also serve as a leading example in promoting Chinese efforts at cooperation with other European countries in the context of the BRI. As China’s president, Xi Jinping, told the Greek Prime Minister Alexis Tsipras during the first Belt and Road Forum on International Cooperation in 2017, the Chinese government was willing to work together with Greece to build the Port of Piraeus as an important focal point for the BRI.81 For Greece, interest in Chinese investment intensified during the 2008 crisis, following initial discussions during the center-right Prime Minister Karamanlis’ visit to China in 2006. 79

“Shenhua buys 75 pct of Copelouzos Group wind farms” November 2, www.amna​.gr/ en/business/article/201236/Shenhua-buys-75-pct-of-Copelouzos-Group-wind-farms (accessed August 4, 2021). 80 “China’s Mediterranean Odyssey”, The Diplomat, 19 April 2016; https://thediplomat​ .com/2016/04/chinas-mediterranean-odyssey/ (accessed August 4, 2021). 81 “Xi calls for expanded cooperation between China, Greece”, Xinhuanet, 13 May 2017; www.xinhuanet.com/english/2017-05/13/c_136280026.htm (accessed August 4, 2021).

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

278

China’s Chance to Lead

During President Hu Jintao’s official visit to Greece in November 2008, COSCO Shipping Ports Ltd, a subsidiary of the Chinese stateowned shipping company COSCO, signed a 35-year lease agreement worth USD1 billion to operate piers II and III at Piraeus’ container terminal; Pier I remained under the management of the state-owned Piraeus Port Authority (PPA). With funding from the China Development Bank, COSCO invested in new deeper water docks and larger cranes, enabling Piraeus to accommodate the latest generation of large containerships and increase capacity. COSCO also improved the port’s transit facilities and built a new railway link between the port and the Greek national rail system. Several additional commitments soon followed, including a USD5 billion shipping fund in 2011 to enable Greek shipowners to place new building orders at Chinese shipyards. Following the launch of the BRI in 2013, COSCO invested another USD270 million to further upgrade the two leased piers and to construct a new adjacent oil product terminal. During 2013–2014, COSCO signed agreements with a number of foreign companies to channel a substantial portion of their European distribution through Piraeus. During Prime Minister Li Keqiang’s visit to Greece in June 2014, the two countries signed trade agreements worth USD4 billion and 19 bilateral agreements promoting political, economic, and cultural cooperation. Needing funds demanded by the EU in return for a third bailout, the government (a New Democracy-led coalition with center-left parties PASOK and DYMAR) initiated the privatization of a 67 percent stake in Piraeus Port Authority in the spring of 2014. The PPA controlled the repair docks, the ferry and cruise terminal as well as Pier I of the container terminal. Of the five bidders, COSCO was expected to win. However, the January 2015 elections produced a new coalition government made up of the radical left Syriza and the right-of-center Independent Greeks, which stopped the privatization process. Strong resistance came from the dockworkers’ union and the Communist Party. However, the Greek deputy prime minister announced the government was ready to restart the privatization process during a visit to China in the spring of 2015. COSCO won the tender, enabling it to acquire full management control of Piraeus through a 51 percent stake in the Port Authority with the promise of another 16 percent after five years, subject to certain investments. Since then, COSCO has invested USD5 billion with more promised, following a master plan that was proposed in early 2018 that would turn Piraeus into one of the five largest ports in Europe. The plan includes a fourth pier at the container terminal, and a new shipyard facility that specializes in repair and servicing of cruise ships and mega yachts. The plan also

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

Case Studies of Political Regimes and the BRI

279

proposed a large, new cruise terminal that would turn Piraeus into the homeport in the Mediterranean for Chinese cruise ships and passengers, plus four new hotels and a shopping mall. COSCO also indicated interest in acquiring smaller ports on Greek islands in the Aegean and Adriatic to establish a cruise network. According to the plan, the “ultimate goal is for Piraeus port to be established as a worldwide transportation hub and in this way enhancing the development of the Greek economy in general.”82 Despite approving COSCO’s acquisition of the PPA, the government rejected parts of the plan at the end of 2018 with reference to environmental impact studies, as well as archaeological and cultural interests. Objections also stemmed from domestic interests fearing competition and negative effects on local economies, such as construction companies, local businesses and government ministries.83 In November 2019, following the July 7, 2019 election that brought the center-right New Democracy Party to power, President Xi visited Greece and signed a MOU supporting COSCO’s expansion plan for Piraeus84 including USD660 million to develop the port and the construction of a fourth container terminal, worth USD224 million.85 Although segments of the Greek electorate have expressed concerns about COSCO’s investments in Greece, the overall impact on the Greek economy has been hugely positive. By introducing the most advanced equipment and upgrading the infrastructure of the port, container volumes increased by more than 700 percent, making Piraeus the fastest-growing container port in the world, the second largest port in the Mediterranean and the seventh largest in Europe.86 COSCO’s investments have 82 “Cosco reveals $620 million Piraeus development plan,” Seatrade Maritime News, January 29, 2018, www.seatrade-maritime.com/asia/cosco-reveals-620m-piraeus-devel​ opment-plan; (accessed August 4, 2021). 83 “Cosco Shipping’s Piraeus plans dealt a blow”, Seatrade Maritime News, December 3, 2018, www.seatrade-maritime.com/asia/cosco-shippings-piraeus-plans-dealt-blow (accessed August 4, 2021). 84 “Greek Shipping Ministry Welcomes Updated Euro800m Piraeus Port Plan,” September 2, 2019, https://news.gtp.gr/2019/09/02/greek-shipping-ministry-welcomesupdated-e800m-piraeus-port-plan/ (accessed August 4, 2021). 85 Officials announced during Mr. Xi’s visit that the European Investment Bank would provide a $155 million loan to back Cosco’s expansion plan for Piraeus, with the loan guaranteed by the Export-Import Bank of China; “China’s Cosco Pours More Money into Greek Port”, www.wsj.com/articles/chinas-cosco-pours-more-money-into-greekport-11573581625; (accessed August 4, 2021). 86 “China is making substantial investment in ports and pipelines worldwide,” February 6, 2020, The Economist, www.economist.com/special-report/2020/02/06/china-ismaking-substantial-investment-in-ports-and-pipelines-worldwide (accessed August 4, 2021). “Piraeus becomes the second largest port in the Med,” January 7, 2019, Seatrade Maritime News, www.seatrade-maritime.com/asia/piraeus-becomes-secondlargest-port-med (accessed August 4, 2021).

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

280

China’s Chance to Lead

generated 3,000 direct local jobs, more than 10,000 indirect jobs, and greatly contributed to Greece’s economic development.87 8.7.4

COSCO and Piraeus Port: Addressing Stakeholders’ Concerns

Although COSCO’s investments have proven to be very beneficial to the Greek economy, vocal criticisms erupted from a wide range of stakeholders, which were amplified by Greece’s severe and prolonged crisis. Despite intense political pressure, COSCO and the Greek government avoided the problems that plagued Indonesia’s HSR and Malaysia’s ECRL projects largely due to the strength of the Greek government’s institutions that prevented the ruling party from bypassing or rushing due diligence processes. The strength of Greece’s domestic institutions and adherence to due diligence procedures were bolstered by strict EU oversight. COSCO’s initial proposal to operate Piers II and III in 2008 elicited angry protests and strikes. With the unemployment rate in the Athens suburb at around 70 percent,88 dockworkers were very concerned about job losses and a worsening of working conditions. Even after the ratification of the concession contract by the Greek Parliament in 2009, rallies continued, with criticisms continuing over the use of subcontractors that employed unskilled, nonunion workers.89 However, protests did not occur when COSCO acquired Piraeus Port in 2016. COSCO executives held dialogues with employees, and carried out personnel restructuring that resulted in a collective agreement, and avoided wage cuts or redundancies of local personnel. Out of a total workforce of more than 3,000 people, just ten were Chinese (Qianqian and Davarinou 2019). COSCO has also assisted with the revitalization of Greece’s ship repair industry, an important pillar of the Greek maritime sector, by investing more than USD65 million to renovate the local ship repair zone and upgrade the infrastructure.90 COSCO also addressed local social responsibility concerns through the construction of schools and roads, aiding orphanages of the Piraeus 87

“Xi’s visit can propel Greece’s role in BRI,” November 11, 2019, Global Times, www​ .globaltimes.cn/content/1169675.shtml (accessed August 4, 2021). “Chinese company revitalizes Greek port,” November 11, 2019, People’s Daily, http://en.people.cn/ n3/2019/1111/c90000-9631168.html (accessed August 4, 2021). 88 www.wsj.com/articles/chinese-transform-greek-port-winning-over-critics-1416516560 (accessed August 4, 2021). 89 www.nytimes.com/2012/10/11/business/global/chinese-company-sets-new-rhythm-inport-of-piraeus.html (accessed August 4, 2021). 90 “Cosco determined to revitalize ship repair in Piraeus port,” October 1, 2020, Seatrade Maritime News. www.seatrade-maritime.com/shipbuilding/cosco-determined-revitaliseship-repair-piraeus-port (accessed August 4, 2021).

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

Case Studies of Political Regimes and the BRI

281

Region, and by working with the Holy Metropolis of Piraeus to provide food for the poor (Piraeus Port Authority 2018).91 With regard to environmental impact, COSCO adheres to the environmental management system PERS (Port Environmental Review System) of the European Sea Ports Organization. It also meets minimum European environmental standards and Greek regulations regarding waste management and relies on photovoltaic power generation to reduce carbon emissions. Although COSCO has displayed considerable attentiveness to local stakeholders’ concerns, there are two major challenges for future Chinese investment in Greece. The first regards domestic political opposition from local businesses and communities negatively affected by Chinese investments, as arose in the initial reaction to COSCO’s master investment plan in 2018. For example, the initial proposal to construct a mall inside the cruise terminal was feared to be a final blow to the shopping stores of Piraeus forcing the mall to restrict its access exclusively to tourists (Qianqian and Davarinou 2019). And although the current ruling party may be receptive to Chinese investment, this could change following the next election. Swings from one ruling party coalition to another create uncertainty about Greece’s future receptivity to Chinese investment in the long run. The second challenge is at the regional level. The EU plays an important role in creating and enforcing rules governing foreign investment. EU member states, particularly France and Germany, have expressed growing concerns about Chinese investments in Greece. A key issue for countries with advanced technology and know-how is the long-term implications of Chinese investment for their economic competitiveness (Meunier 2019). For example, Sigmar Gabriel, German Minister of the Economy at the time, proposed an expansion of the 2009 Foreign Trade Law to give the German government more flexibility in deciding to review and block certain foreign investments following the takeover of German robot-maker Kuka by Chinese home appliance maker Midea in 2016. This was followed by France, Germany, and Italy writing to the EU Trade Commissioner in February 2017 to request a European debate regarding the conditions that would permit countries to investigate or block foreign investments, especially in relation to sensitive high-tech products or when the investor is state-owned. Concerns have also arisen with regard to China’s influence beyond foreign investment. For example, Sigmar Gabriel, as German foreign minister, blamed Greece for not supporting the EU’s proposal to bring 91

www.worldreginfo.com/wdoc.aspx?file=Piraeus_Port/2/13F11B98-07E2-4580871C-F03720179186/394262_rfa_2017_en_grs470003013.pdf (accessed August 4, 2021).

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

282

China’s Chance to Lead

the freedom of navigation in the South China Sea to the International Tribunal for the Law of the Sea out of fear that Chinese investment might be affected.92 Rising concerns about China’s growing economic and political influence in Europe contributed to a review of EU-China relations by the European Commission and the High Representative leading to the publication of the EU–China Strategic Outlook that specifies ten concrete actions for EU member states to follow when dealing with challenges with China (European Commission 2019). The EU also approved new rules for the screening of foreign investment that raises the complexity and difficulty of doing business, especially for high-tech and strategic sectors (Chan and Meunier 2021). Coinciding with this initiative, investment by China’s state-owned companies into Europe remained at a ten-year low in 2020, while private sector investment plummeted 49 percent from the year before (Kratz et al. 2021). 8.7.5 Summary China’s primary interest in Greece is due to its geographical location as a point of entry into Central, Eastern, and Southeastern Europe, as well as its potential to serve as a regional transshipment hub for the Eastern Mediterranean. Greece’s interest in Chinese foreign investment has been driven by short- and long-term factors. The short-term factor was the desire to attract Chinese spending during the Greek debt crisis. The long-term factor is to promote growth, and in turn create jobs and business opportunities. As shown in Chapters 6 and 7, liberal democracies are the least likely recipients of Chinese SOE-led infrastructure projects and investments. The huge investment into Piraeus Port by COSCO (an SOE) was due primarily to the debt crisis. This granted an opportunity for China to gain a foothold and to then overcome resistance to Chinese SOEs. Indeed, COSCO’s and China’s long-term success in Greece has crucially depended upon generating positive outcomes for Piraeus Port stakeholders, including workers, local businesses, the local community, environmental groups, and the Greek federal government. The strength of Greece’s democratic institutions helped to ensure that shortterm political pressures did not compromise adherence to regulations designed to safeguard the project’s long-term success. 92

“German Foreign Minister Wants China to Abide by One-Europe Principle: Chinese Foreign Ministry Is Shocked by the Remarks,” August 31, 2017, Global Times, https:// world.huanqiu.com/article/9CaKrnK505G (accessed August 4, 2021).

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

Case Studies of Political Regimes and the BRI

283

Due to the exceptional circumstances of the Greek debt crisis, government-to-government negotiations occurred in relation to Piraeus Port. But the ongoing objections by various groups within the ruling party coalition suggest the potential for future problems. The main attraction to continued Chinese investment following the end of the debt crisis, and the election of the center-right party, has been the generation of business opportunities and economic growth. COSCO successfully demonstrated that it could inject growth into the local Greek economy to the benefit of Greek private businesses and workers. But return to normalcy has also led to stronger oversight and influence on the approval of new investments by the EU, particularly by Germany and France. They are liberal democracies that favor private firms over SOEs. Consequently, more stringent regulatory approvals coupled with stronger anti-China political pressures suggest further challenges to China’s investments in Greece. 8.8 Conclusions A core finding illuminated by the case studies is the agency of recipient countries with regard to BRI projects. The nature of this agency is due to the country’s political regime and is manifested in two ways. First, recipient countries’ political regimes influence the level of enthusiasm for BRI projects. Second, recipient countries’ political regimes influence how BRI projects are implemented. Political regimes’ level of enthusiasm for BRI projects is primarily due to clientelist demands that vary according to the strength of political opposition combined with the public-private control of the corporate sector. In countries with a high demand for clientelist resources and highly centralized state control, political incumbents will aggressively pursue BRI projects with the government and host country SOEs as the main partners, as in weak electoral autocracies. Malaysia offers a clear illustration of this phenomenon. Political incumbents of strong electoral autocracies do not confront the same worries about losing power and are therefore less inclined to engage in risky projects to preserve their hold on power, as in the case of Djibouti. Although political opposition displayed a surprisingly strong result in the 2013 election, Guelleh has since consolidated his hold on power and did not face political challengers from the main opposition parties in the 2018 election. Leaders of closed autocracies confront even weaker electoral incentives to attract BRI projects since multiparty elections are forbidden, as in the UAE. The political regime most similar to weak electoral autocracies and political incumbents’ electoral pressures is electoral democracy. As was

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

284

China’s Chance to Lead

seen in the case of Indonesia, Chinese foreign spending accelerated quite dramatically during President Joko Widodo’s tenure. Despite the case about the Jakarta-Bandung HSR, Indonesian SOEs are less frequently involved as the key partner with their Chinese counterpart in comparison to Malaysia. Both countries engage in a range of public and private collaborations with Chinese projects, but there is a clear state-led orientation in Malaysia whereas Indonesian projects are more often private-firm led, as in the case of the Indonesia-Morowali Industrial Park. Finally, electoral pressures in liberal democracies are also strong, but the strength of their political institutions reduces the skirting of due diligence procedures for short-term political gain. The Greece case is unusual among liberal democracies due to the country enduring a deep and prolonged debt crisis, leading it to welcome Chinese SOE spending in the absence of foreign investment from other sources. But this experience offers helpful insights into how the institutions of liberal democracies influence the implementation of BRI SOE-led projects differently from other political regimes. The case studies in this chapter offer qualitative evidence consistent with the quantitative patterns identified in earlier chapters, particularly in Chapters 6 and 7 with regard to the level of Chinese foreign spending directed to specific political regimes in the context of the BRI. How BRI projects are implemented also reflects a political regime’s demand for clientelist resources combined with the public-private orientation of the corporate sector. Weak electoral autocracies in which political incumbents face elevated threats to their hold on power combined with their centralized control of state resources leads to a heavier reliance on state-led partnerships with Chinese SOEs in order to extract a larger volume of BRI funds that can be used as clientelist benefits. These features also contribute to a larger number of projects, and especially megaprojects. These features also suggest strong political influence on the implementation of these projects, which is reflected not only by federal government involvement in these projects but also via the weak PPP system. And as a consequence of the political pressure to skirt due diligence processes, numerous problems are more likely to arise as occurred with several of the BRI projects (e.g., the ECRL, the two gas pipelines). We can see the difference due to political regime by comparing Malaysia with Greece. In both instances, Chinese SOE-led megaprojects were implemented under circumstances of intense electoral pressure. However, Greece remained committed to due diligence processes and ensured local workers, businesses, environmental, and local stakeholder concerns were adequately considered and addressed. Consequently, the Piraeus Port project yielded very positive results for the Greek economy.

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

Case Studies of Political Regimes and the BRI

285

It is clear from a comparison of these two cases that political regimes play a decisive role in how BRI projects are implemented. The implementation of BRI projects in the context of closed autocracies displays a lower reliance on host country SOEs and PPPs, reflecting the greater prevalence of private firms owned by the ruling elite to collaborate with Chinese SOEs and the reduced incentives to leverage existing financial resources via PPPs. Electoral democracies, by contrast, display the greatest similarities to weak electoral autocracies with regard to the incentives to use PPPs, but with a stronger reliance on private firms. Overall, the evidence from the case studies offers detail consistent with the quantitative findings regarding BRI project characteristics presented in Chapters 6 and 7. A final point is that not all Chinese SOE projects are problematic. Indeed, Piraeus Port illustrates that Chinese SOE-led projects can be hugely beneficial to the local economy. Problems are most likely to occur when incumbents face strong political opposition coupled with weak institutions to ensure due diligence processes are adhered to as witnessed with the Jakarta-Bandung HSR in Indonesia and the East Coast Rail Line in Malaysia. Of course, objections to Chinese SOE-led foreign spending do not focus solely on the economic merits of a project, but also extend to concerns about influence on a host government’s policies as well as worries regarding the acquisition of sensitive technologies. Appendix Table 8A.1  Chinese foreign spending in the UAE, 2005–2019 (Millions of USD) Year

Investment

Construction

Combined

Projects initiated

2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Total

0 0 0 0 0 0 0 0 450 0 0 1,360 3,130 2,230 610 7,780

0 300 650 3,550 670 130 450 1,150 280 640 760 2,790 4,660 5,780 3,710 25,520

0 300 650 3550 670 130 450 1,150 730 640 760 4,150 7,790 8,010 4,320 33,300

0 0 8 0 1 6 2 2 1 7 2 10 11 3 1 54

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

286

China’s Chance to Lead

Table 8A.2  Chinese foreign spending in Djibouti, 2005–2019 (Millions of USD) Year

Investment

Construction

Combined

Projects initiated

2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Total

0 0 0 0 0 0 0 190 0 0 0 0 0 0 0 190

0 0 0 0 0 0 0 510 0 0 1,020 0 0 0 0 1,530

0 0 0 0 0 0 0 700 0 0 1,020 0 0 0 0 1,720

0 0 3 0 0 0 1 0 2 1 0 1 2 0 1 11

Table 8A.3  Chinese foreign spending in Malaysia, 2005–2019 (Millions of USD) Year

Investment

Construction

Combined

Projects initiated

2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Total

0 0 0 0 0 350 1,830 460 3,200 400 7,240 2,370 930 1,280 780 18,840

0 0 0 1,830 0 1,810 1,260 3,100 1,930 2,870 1,260 6,210 4,860 490 220 25,840

0 0 0 1,830 0 2,160 3,090 3,560 5,130 3,270 8,500 8,580 5,790 1,770 1,000 44,680

0 1 1 1 1 2 4 0 6 8 6 23 4 3 11 71

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

Case Studies of Political Regimes and the BRI

287

Table 8A.4  Chinese foreign spending in Indonesia, 2005–2019 (Millions of USD) Year

Investment

Construction

Combined

Projects initiated

2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Total

0 100 0 130 230 830 1,870 2,100 420 2,520 3,760 570 2,550 5,000 3,480 23,560

1,060 1,210 2,480 1,130 0 0 1,830 1,650 1,560 2,450 2,300 3,200 1,040 1,090 4,700 25,700

1,060 1,310 2,480 1,260 230 830 3,700 3,750 1,980 4,970 6,060 3,770 3,590 6,090 8,180 49,260

0 0 0 5 6 0 3 6 4 4 12 13 10 9 19 91

Table 8A.5  Chinese foreign spending in Greece, 2005–2019 (Millions of USD) Year

Investment

Construction

Combined

Projects initiated

2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Total

0 0 0 5,790 0 0 120 0 0 0 0 970 2,200 0 670 9,750

0 0 0 0 0 130 0 0 0 0 0 0 0 340 320 790

0 0 0 5,790 0 130 120 0 0 0 0 970 2,200 340 990 10,540

0 0 0 0 0 0 0 13 0 1 0 0 2 1 0 17

https://doi.org/10.1017/9781009385879.008 Published online by Cambridge University Press

9

Chinese Exports of Digital Technologies and Standards

The previous empirical chapters have focused primarily on Chinese spending related to infrastructure development in the context of the BRI. The evidence has firmly shown that electoral autocracies have been the major recipients of BRI spending. I further theorize infrastructure spending offers a platform for Chinese digital technologies and standards to be introduced and thereby promote de facto standardization. This chapter examines whether the patterns observed for political regimes and Chinese infrastructure spending are also manifested with regard to Chinese digital technologies exports. To assess this claim, I first begin with an overview of China’s formal standardization efforts. Thus far, China’s role in standards development organizations has rapidly increased. However, established players have taken notice and are pushing back. Collectively, they retain dominant representation in these organizations and it is increasingly likely that they will coordinate their efforts to limit China from overtaking them in establishing formal standards. China is likely to have more success in promoting the de facto standardization of its digital technologies by exporting them to developing countries. The subsequent section discusses how China has engaged in this effort. I next turn to an analysis of a novel dataset regarding the export of Chinese smart cities technologies to identify whether there is a discernible pattern according to the political regimes of host countries. The findings reveal that electoral autocracies are the predominant destination for these technologies among developing countries. I then examine whether Chinese foreign aid in the information and communications technology (ICT) displays a similar bias toward electoral autocracies since the launch of the Digital Silk Road in 2015. I use a dataset constructed by AidData.org which includes 555 ICT aid projects between 2000 and 2017. I analyze both the number of commitments as well as the total dollar amount going to each type of political regime for each year. The findings indicate there is a pronounced increase in both aid indicators going to electoral autocracies from 2015 onward, and a growing divergence between electoral autocracies and electoral democracies. 288

https://doi.org/10.1017/9781009385879.009 Published online by Cambridge University Press

Chinese Exports of Digital Technologies and Standards

289

The tests establish a strong correlation between Chinese exports of digital technologies and standards that varies by political regime. To establish whether there is a link between the construction of BRI projects which then contributes to the export and adoption of Chinese digital technologies, I conduct focused case studies of two host countries representing political regimes with the largest share of Chinese smart cities technologies exports, Malaysia (electoral autocracy) and Greece (liberal democracy). An important difference that emerges from these cases regards central government involvement in promoting Chinese DSR projects in Malaysia in comparison to Greece where market mechanisms independent of government involvement are more influential. Finally, I focus on Huawei’s new cloud business which the company has aggressively promoted after the United States blocked sales of mobile chips and related technologies to the firm in 2019. This dealt a crippling blow to Huawei’s consumer products business which accounted for nearly half of the company’s revenue. The development of Huawei’s cloud business thus became a strategic priority vital to the firm’s survival. Huawei received support from Beijing through public cloud contracts, which form the foundation on which subsequent deals have been struck between Huawei and foreign governments or state-owned enterprises for cloud infrastructure and e-government services. I use a dataset developed by the CSIS Reconnecting Asia project that identifies deals through April 2021 to examine whether electoral autocracies are more frequent customers than other regime types. The findings are consistent with prior evidence indicating they are again the major recipients, accounting for about half of all deals. The chapter begins with a discussion about the limits of formal standardization for China in the context of standards development organizations. I then discuss the promise of de facto standardization for Chinese technologies. Next, I analyze the pattern of Chinese smart cities technologies exports across political regimes, followed by ICT aid. The case studies about Malaysia and Greece provide context for the findings in these two datasets. Finally, I examine the latest moves by Huawei with its cloud business which is being aggressively promoted in developing countries, and which has benefited from heightened demand for digital and cloud-related services during the pandemic. 9.1

The Limits of Formal Standardization

China, led by Huawei, has gained increasing influence in standards development organizations (SDOs). SDOs are the international bodies that ensure technology products can connect and communicate with

https://doi.org/10.1017/9781009385879.009 Published online by Cambridge University Press

290

China’s Chance to Lead

each other. At SDOs, stakeholders from around the world convene to decide which technological specifications should become the international standard. The outcome of this process can profoundly impact the fortunes of technologies, companies, and nations. Historically, SDOs have been dominated by the United States and Europe, but China has rapidly increased its participation and leadership in these organizations, especially in new technologies that are yet to be standardized such as fifth-generation mobile technology (5G),1 artificial intelligence,2 and data security.3 A variety of indicators reveal China’s growing clout, including the number of leadership positions it holds in international SDOs which help China shape the agenda, or by influencing the standard-setting process (Seaman 2020; Ruhlig 2021; Russel and Berger 2021).4 China is also rapidly increasing its active membership in standard developing committees which is where actors can submit proposals and comments to the standardization process. Since 2005, its participation on such committees has overtaken France, Japan, and the United States. Additionally, China has rapidly increased its contributions to standardization committees. For example, China has made a larger share of standardization contributions to 4G and 5G technologies compared to any other country.5 Finally, China is ahead of other country actors with regard to 5G standard essential patent declarations with 33 percent (Malkin 2020; Russel and Berger 2021).6 Standard essential patents are patents that could become essential for the standard and that participating actors are willing to license the patent. China also appears to be outcompeting other countries in terms of international patenting related to other

1

The U.S., China and Others Race to Develop 5G Mobile Networks, Forbes, 3rd April 2018, The U.S., China and Others Race to Develop 5G Mobile Networks (forbes.com) (accessed March 17, 2022). 2 Guidelines for the Construction of the National New Generation Artificial Intelligence Standard System, SAC, 4th August 2020. www.sac.gov.cn/sxxgk/zcwj/202101/ P020210122407767317794.pdf 3 Guidelines for the Construction of the Online Data Security Standards System « China Copyright and Media (wordpress.com) (accessed March 17 2022). 4 ISO and NIST in US-China Business Council, “China in International Standards Setting: USCBC Recommendations for Constructive Participation,” February 2020. International Organization for Standardization. www.iso.org/home.html 5 IPlytics in Valentina Pop, Sha Hua and Daniel Michaels, “From Lightbulbs to 5G, China Battles West for Control of Vital Technology Standards,” The Wall Street Journal, 8 February 2021. 6 Pohlmann, Tim et al., Study on the investigation and analysis of the patent situation in the standardisation of 5G (commissioned by the Federal Ministry for Economic Affairs and Energy) (in German), IPlytics, Berlin, 2020; 5G-patent-study_TU-Berlin_IPlytics-2020. pdf (accessed March 17 2022).

https://doi.org/10.1017/9781009385879.009 Published online by Cambridge University Press

Chinese Exports of Digital Technologies and Standards

291

digital technologies, including AI. In 2018, China filed more than 2.5 times as many AI patents as US actors.7 China’s rapid rise in the number of leadership positions and participation in international SDOs reflects the government’s financial backing and efforts to coordinate among digital technology actors (Ruhlig and ten Brink 2021). Programs at the local, provincial, and national levels provide financial subsidies to Chinese companies for submitting standards applications to international SDOs.8 In the 5G domain, Huawei stands out not only for the size of its R&D investments, but also illustrates the government’s willingness to commit significant financial backing to private companies involved in the development of technologies it considers of strategic importance. In 2017, Huawei spent more than the combined R&D investments of its two main 5G competitors, Ericsson and Nokia (US$ 14.3 billion compared to a combined total of US$ 10.6 billion) (Morris, 2018). Over the course of twenty-five years, Huawei has received nearly USD75 billion in tax breaks and cheap loans (Ruhlig and ten Brink 2021). Significant R&D investments are a prerequisite for standardization, enabling Huawei to submit the largest number of 5G standards of any company in 2018.9 Although China is rapidly increasing its participation in and leadership of international SDOs, China’s influence remains far outweighed by the collective contributions of the more established countries (US, UK, Japan, Germany, France, etc.). And there is a very real possibility that pushback will occur. This has already happened. For example, the US government added Huawei to its “entity list” in 2019 which prohibits US companies from engaging in trade with the firm10 as part of a wider effort to curtail its ability from participating in global markets and global standard setting bodies. This led to major Western markets restricting the use of Huawei technologies, including the UK, Germany, and Japan.11 It also contributed to US-based SDOs, such as the Institute of Electrical and Electronics Engineers, banning Huawei from its meetings. However, this decision was later reversed due to concerns over Huawei’s inalienable role in standardizing key elements 7 Shi-Kupfer, Kristin and Ohlberg, Maraike, China’s Digital Rise: Challenges for Europe, MERICS, Berlin, 2019 8 2019年全国各地区标准化制修订补贴政策 (gieha.org) (accessed March 17, 2022). 9 IPlytics in Dan Strumpf, “Where China Dominates in 5G Technology,” The Wall Street Journal, February 26, 2019. 10 US Entity List? The Impacts of Huawei Sanctions on Global Traders (aeb.com) (accessed March 17, 2022). 11 Huawei: Banned and Permitted in Which Countries? List and FAQ – ChannelE2E: Technology News for MSPs & Channel Partners (accessed March 17, 2022).

https://doi.org/10.1017/9781009385879.009 Published online by Cambridge University Press

292

China’s Chance to Lead

of 5G telecommunications equipment technology. This in turn led some US companies to withdraw from international SDOs altogether in order to avoid potential legal consequences, causing US influence in technological standardization to decline. This forced the US government to remedy this issue by allowing technology sharing for a limited set of standards development activities to Huawei and its affiliates.12 While this initial effort backfired in achieving its intended objective, the United States and other advanced economies are reconsidering how to bolster their influence in SDOs. The EU, for example, has moved toward stronger top-down strategic coordination with a high-level strategic forum facilitating exchange between private standardizers, the European Commission, and EU member states.13 Similar suggestions have been proposed for the United States.14 Despite significant differences in the approaches to standard setting of the United States, Europe, Japan, and other like-minded states, there exists tremendous opportunity for cooperation and collaboration. For example, proposals are being considered that would enable the United States to work with allies and partners to lower global barriers to industry participation in international SDOs. For China, there are two main options for increasing the adoption of its technological standards. First, China can continue its efforts to expand its influence in international SDOs, but it seems increasingly likely that resistance will grow among the established players. Second, China can pursue the de facto adoption of its standards by selling its products directly to customers and establishing a dominant position in key product categories and markets that make it too costly for users to switch to a different standard. This has the dual benefit of directly promoting the adoption of its standards regardless of the standards endorsed by international SDOs and secondarily bolstering the argument for international SDOs to endorse Chinese standards given their widespread adoption. Regardless of how China’s role in SDOs evolves, establishing de facto dominance of standards offers the greatest certainty and the path of least resistance to China achieving its ambition of becoming a standards-maker for a growing share of the global marketplace.

12

In Rare Move, US Clears Limited Cooperation Between US Firms, Huawei (voanews​ .com); Why the US Campaign Against Huawei Backfired – The Diplomat (accessed March 17 2022). 13 Will China Set Global Tech Standards? | ChinaFile (accessed March 17 2022). 14 www.chinafile.com/conversation/will-china-set-global-tech-standards (accessed March 17 2022).

https://doi.org/10.1017/9781009385879.009 Published online by Cambridge University Press

Chinese Exports of Digital Technologies and Standards

9.2

293

The Promise of De Facto Standardization

In contrast to formal standardization, de facto standards emerge as a result of widespread adoption rather than regulatory or industry decisions. Once infrastructure is built and accompanying digital technologies and software installed, host countries become “stuck” with the standards embedded in them. Subsequently, it is very costly for customers to switch to a different system that utilizes different, sometimes incompatible standards even if they are endorsed by SDOs. This allows Chinese companies to lock-in foreign countries to their technologies and products and exclude foreign competitors. The Chinese party-state plays an important and active role in promoting the export and adoption of Chinese technologies and standards. There are several mechanisms by which this is pursued. First, the government has established policies and action plans that require Chinese companies to include Chinese standards in their infrastructure exports and foreign projects. Second, the government engages in bilateral agreements to secure the adoption of its standards in partner countries. Third, general commitments to adopt Chinese standards are included in regional agreements. In 2015, the State Council issued a “Plan to Deepen Reform of Standardization Work,” which directs Chinese companies to integrate Chinese standards into foreign projects and equipment and infrastructure exports. To this end, Beijing published a 2015 action plan, “Connect One Belt, One Road Through Standardization,” which commissions a comparative analysis of standards across BRI countries, calls for 500 priority Chinese standards to be translated into foreign languages, and proposes the establishment of standardization centers in BRI host countries (Russel and Berger 2021). In 2019, Chinese agencies established a BRI National Standard Information Platform in collaboration with five international SDOS (including the International Organization for Standardization, the International Electrotechnical Commission, and the International Telecommunication Union) to provide updates on standards development, country standardization overviews, and classification and translation capabilities. Additionally, a translation cloud platform was introduced that is devoted to the translation of Chinese standards into English. Bilateral agreements offer an additional mechanism to promote Chinese standards. By 2019, China had signed ninety such agreements with fifty-two countries. Although these agreements are not legally binding, they promote cooperation on standards by establishing an exchange between the two countries’ standardization bodies and provide a first step toward more substantive and focused agreements in the future.

https://doi.org/10.1017/9781009385879.009 Published online by Cambridge University Press

294

China’s Chance to Lead

A third way by which China promotes the adoption of its standards outside of formal SDO proceedings is through regional engagements. For example, the 2021 China-ASEAN International Standardization Forum promoted standards compatibility with China, introduced a new China-ASEAN Standards Cloud Platform, and launched an international standardization training center in Nanning, China. A new Chinese SOE was also launched, China-ASEAN Information Harbor, which seeks to promote Chinese standards in the region through projects such as smart ports, cloud communications, and big data information platforms. Finally, cooperation and coordination on standardization was also included in the Asia-Pacific free trade agreement, the RCEP, which includes fifteen Asia Pacific member countries representing 30 percent of global GDP. It was formally signed in 2020. Rail offers a clear illustration of how China has pushed ahead with de facto standardization despite a lack of formal recognition of its standards. Since 2009, China has been promoting its high-speed rail (HSR) standards by applying for foreign patents and by working to promote the adoption its standards by international SDOs. Despite strong backing by the Chinese government, Chinese HSR standards only accounted for 0.7 percent of all international standards by the end of 2016 (Yan 2020). China has nevertheless pushed ahead with promoting its railway technologies in BRI countries. China offers funding, mostly loans, for the development of railways if they are constructed by the China Railway Corporation or other Chinese manufacturers based on Chinese standards (Cai 2017). The CRRC (China’s largest railway company and an SOE) has also sought to establish overseas manufacturing plants in countries where China has won projects worth more than one billion CNY (Hu 2014). Chinese railway projects and standards are more likely to be implemented in countries with a higher volume of Chinese investments (Yan 2022). By the end of 2017, China had seven overseas HSR projects (Yan 2022). This is too few to draw confident inferences relating to different political regimes, but the individual cases do illuminate the role of the state in promoting its HSRrelated standards, as with the Indonesian Jakarta-Bandung (Yan 2020, 2022) and the Ethio-Djibouti railway project (Chen 2021) discussed in Chapter 8. Other projects being built with Chinese standards include the China-Laos Railway, Nepal’s Gyirong-Kathmandu railway, the MombasaNairobi railway in Kenya (Sun 2017),15 Malaysia’s East Coast Rail Link (Earley 2021),16 and the Abuja-Kaduna railway in Nigeria (Ruhlig 2021). 15

www.brookings.edu/blog/africa-in-focus/2017/07/06/china-and-the-east-africa-rail​ ways-beyond-full-industry-chain-export/ (accessed March 17, 2022). 16 https://saiia.org.za/research/chinese-infrastructure-provision-to-the-global-south-andthe-setting-of-rail-standards/ (accessed March 17, 2022).

https://doi.org/10.1017/9781009385879.009 Published online by Cambridge University Press

Chinese Exports of Digital Technologies and Standards

295

Although the export of Chinese railway technologies and standards are prioritized as part of the BRI, digital technologies and standards that are central to achieving China’s economic growth objectives are of even greater importance (see Chapter 4). The export of these new and emerging technologies and standards is commonly referred to as the Digital Silk Road. 9.3

The Digital Silk Road and Smart Cities Technologies

The Digital Silk Road (DSR) refers to an increasing number of digital and telecommunications infrastructure projects that Chinese firms are executing in partner nations across the BRI. The DSR first appeared in a government white paper on the BRI in 2015 (Hemmings 2020) and has been rapidly integrated into the BRI due to three features of China’s industrial policy initiatives including increasing support for artificial intelligence-related technologies, a reliance on private firms with greater expertise in AI than their state-owned counterparts, and rebuilding Chinese cities supported by digital technologies, or smart cities (Naughton 2020). These new digital technologies get included into conventional BRI infrastructure projects via the integration of sensors and communications channels that connect to servers engaging in AI processing of information in real time. The first step in the rollout of the DSR is the promotion and export of China’s technology companies, products, and standards. Related to this are the uses to which these technologies and products may be used, such as the development of smart cities and smart ports. The second step involves cultivating and coordinating support among China’s network of BRI countries to accept its preferred technological standards to win the wider global standards competition in 5G, the next-generation information and communications technology (ICT), and related digital technologies that aim to revolutionize the capture, analysis, and exploitation of data, including AI, big data analytics, blockchain, cloud computing, quantum computing and networking, financial technology (fintech), industrial automation, and other new technologies. Although smart cities technologies are driven primarily by high-tech companies that are not state-owned, the state has considerable influence on their entry into foreign markets in at least two important ways. First, the integration of digital infrastructure into conventional infrastructure projects requires private companies to follow the lead of SOEs, and work closely with them. Second, private firms depend on communications networks that are typically established and controlled by SOEs and the state, such as underseas cables (e.g., the PEACE cable) and satellite

https://doi.org/10.1017/9781009385879.009 Published online by Cambridge University Press

296

China’s Chance to Lead

networks (e.g., the BeiDou Navigation Satellite System). Consequently, there is relatively greater cooperation between the state and private digital technologies firms than with private firms in other industries. By following the implementation of BRI projects, Chinese technology firms are able to establish a strong first-mover advantage in introducing their technologies and standards into many emerging economies. These technologies are often also introduced as “sticky bundles,” as in the cases of smart cities or smart ports, thus making it costly for recipients to switch standards later and thereby contributing to their path dependence of “stickiness.” Finally, network effects enhance the benefits of adhering to Chinese standards among developing countries, especially if they are widespread in a particular region (e.g., Africa, Central Asia, Southeast Asia). In the context of BRI countries, electoral autocracies are expected to play an outsized role in importing Chinese technologies and adoption Chinese standards given their outsized participation in infrastructure projects. 9.3.1

Construction of the Smart Cities Technologies Dataset

The potential to deploy smart cities technologies applies to nearly every country, making it an appealing way to assess the spread of DSR projects. Moreover, the analysis of smart cities technologies is not restricted exclusively to urban settings, but rather to the bundle of technologies typically available for smart cities projects, yielding a useful guide to their general deployment across countries. To identify and classify digital technologies that are considered part of smart cities systems, I follow the classification provided by Atha et al. (2020). The technologies include the major categories of surveillance, network infrastructure, big data, fintech, integrated platforms, and municipal services. As reported in Table 9.1, each technology category includes a number of different product types and firms. Technologies with established smart cities applications are included regardless of whether or not they are described as part of a “smart cities project.” For example, surveillance systems deployed in public settings are included regardless of whether or not they are part of a large-scale smart cities project. To construct the dataset, Chinese firms selling smart cities technologies and systems were identified according to lists of organizations producing Chinese smart cities industrial standards,17 partners and members 17

Standardization Administration of the People’s Republic of China (2018).

https://doi.org/10.1017/9781009385879.009 Published online by Cambridge University Press

https://doi.org/10.1017/9781009385879.009 Published online by Cambridge University Press

Big Data

Network Infrastructure

Surveillance

Technology Category Chinese Companies Involved

IP cameras (internet protocol cameras), CCTV (Closed Huawei, Hikvision, Dahua, Megvii, Circuit Television), DVRs (Digital Video Recorder JD, Kedacom, Cloudwalk, System), NVRs (Network Video Recorder System), SenseTime, CIGIS, Uniview, Yitu, HDCVI (High Definition Composite Video Interface), Transinfo driver LCD screens, police body cameras, facial recognition, image recognition, smart video cloud, smart sensor products, traffic surveillance systems, public space surveillance systems, license plate recognition, coastal surveillance, UAV (Unmanned Aerial Vehicle) aerial photography, three-dimensional scanning 4G OBD (omnibearing-distance system) Terminal, Advantech, CASC, China Telecom, 5G infrastructure, Backhaul Network Infrastructure, China Soft, CITC, COSCO, Backbone networks, Broadband network, GPON Gosuncn, H3C, Huawei, Sangfor, (Gigabit Passive Optical Network), FTTH (Fiber To ZTE The Home) fiber, GSM-R (Global System for Mobile Communications-Railway) wireless communication, GSM-R (Global System for Mobile CommunicationsRailway) Internet, Optical Transmission Network, RFID (radio frequency identification devices), underseas cables, Super Fusion SSD (Solid State Drive) Cache Technology and Striping Technology Cloud networks, cloud data centers, high bandwidth Alibaba, Bingo Cloud, CASC, Digital interconnections, big data management China, Horizon Robotics, Huawei, Infervision, Inspur, iSoftStone, Megvii, Ping’an, SenseTime, Speechocean, Sugon, Tencent, ZTE

Product Type

Table 9.1  Summary of Chinese smart cities technology exports

Tencent (14), Huawei (13) Total projects: 56

Huawei (44), China Telecom (21) Total projects: 100

Hikvision (40), Uniview (13) Total projects: 80

Top Two Companies by Number of Projects

https://doi.org/10.1017/9781009385879.009 Published online by Cambridge University Press

Blockchain-based ledgers, ATM Dynamic Currency Conversion, E-commerce platform, Mobile payment applications Smart meters, smart grid, advanced metering infrastructure

Product Type Alibaba, ChinaSoft, CUP DATA, Huawei, JD, Ping’an

Chinese Companies Involved

ChinaSoft (6), Ping’an (5) Total projects: 17 Huawei (3), all others are 1 Total projects: 11 Huawei (16), CFLD (10) Total projects: 46

Top Two Companies by Number of Projects

CETC, SCGID, GCL, Hanergy, Huawei, NewEra, Shenhua, SkySolar, YSPI Integrated Platforms “Safe City” Solutions, Unified digital platform, IoT Alibaba, Bingo Cloud, CFLD, (Internet of Things) and autonomous machines, Smart E-Hualu, HiKvision, Huawei, City Data Center, ITS (Intelligent Transportation Megvii, Merchants, Ping’an, System), urban operation system platforms, dispatching Sangfor, Terminus, TouchRoad, systems, call centers, emergency response systems Transinfo, ZTE Municipal Services Integrated government services app, AI Automatic con- Advantech, BAOYE, Carsmart, KEYTOP (14), trol signal lamp, building energy saving solutions, intel- CIMCP, E-Hualu, ENJOYOR, TelChina (7) ligent transportation, intelligent traffic management H3C, Hikvision, Hualu, Huawei, Total projects: 59 solution, Smart Parking System, Smart Streetlamps, Inspur, iSoftStone, IZP, KEYTOP, Smart Waste Management, driverless Metro Line Merchants, Ping’an, SIAT, TelChina, ZTE

Energy

Fintech

Technology Category

Table 9.1  (cont.)

Chinese Exports of Digital Technologies and Standards

299

of Chinese smart cities industry alliances and research centers,18 entities listed in securities research reports on Chinese smart cities industry development,19 and through interviews with executives involved in smart cities-related technologies and projects. The total sample of Chinese companies involved in smart cities products totals 326. Searches were conducted of the official websites of each of these firms for reports of smart cities projects in other countries and supplemented by English language searches of foreign news reports conducted via Factiva which offers comprehensive coverage of global English-language news publications. Altogether, these searches yielded 317 reported cases of 58 different Chinese firms exporting smart cities technologies to a total of 88 countries. Table 9A.1 displays the number of projects for each technology category by country. The wide range of countries with Chinese smart cities technologies from all corners of the world illustrates the vast breadth of demand. 9.3.2 Findings The findings displayed in Table 9.2 indicate emerging economies host a larger number of projects than advanced economies. This is consistent with expectations that development policy, commercial, and political considerations drive demand for digital technologies in emerging economies, whereas the primary driver in advanced economies is commercial since they are predominantly liberal democracies. Examination of technology categories shows the biggest difference between emerging and advanced economies is with surveillance, an especially useful technology for political reasons, and network infrastructure (such as 5G networks) which forms the backbone of Chinese digital technologies. Network infrastructure and surveillance technologies have been especially successful exports due, in part, to the vast resources that the Chinese government has put into developing these industries domestically. China’s industrial policies to promote the development of AI-related technologies, reliance on private firms with greater expertise in AI than SOEs, and the experience and knowledge gained from 18

Enterprise Alliance Group, China Smarter City Development and Research Center, http://scdrc.sic.gov.cn/Column/342/0.htm/ (accessed March 17 2022); 60 Company Introductions, New Smart City Construction Enterprise Alliance, www.eascity.org/​ alliance-partners/alliance-partners (accessed March 17 2022); Alliance Members, Z-Park Strategic Alliance of Smart City Industrial Technological Innovation, www​.smartcityunion.cn/index.php?m=content&c=index&a=lists&catid=16 (accessed March 17, 2022). 19 Ping An Securities (2018).

https://doi.org/10.1017/9781009385879.009 Published online by Cambridge University Press

https://doi.org/10.1017/9781009385879.009 Published online by Cambridge University Press

80 25.2 104 32.8 57 18.0 17 5.4 11 3.5 49 15.5 63 19.9

Surveillance, Number of Projects Surveillance, Proportion of Total Projects

Network Infrastructure, Number of Projects Network Infrastructure, Proportion of Total Projects

Big Data, Number of Projects Big Data, Proportion of Total Projects

Fintech, Number of Projects Fintech, Proportion of Total Projects

Energy, Number of Projects Energy, Proportion of Total Projects

Integrated Platforms, Number of Projects Integrated Platforms, Proportion of Total Projects

Municipal Services, Number of Projects Municipal Services, Proportion of Total Projects

24 7.6

13 4.1

7 2.2

3 0.9

27 8.5

40 12.6

27 8.5

123 38.8

Adv. Econs.

39 12.3

36 11.4

4 1.3

14 4.4

30 9.5

64 20.2

53 16.7

194 61.2

Emerg. Econs

7 2.2**

10 3.2

0 0.0

6 1.9*

11 3.5**

9 2.8***

7 2.2***

38 12.0***

Closed Auts

25 7.9

18 5.7

3 0.9

6 1.9*

14 4.4*

35 11.0

28 8.8

109 34.4

Electoral Auts

14 4.4

10 3.2

2 0.6

4 1.3

7 2.3***

22 6.9*

22 6.9

63 19.9***

Electoral Dems

17 5.4

11 3.5

6 1.9

1 0.3

25 7.9

38 12.0

23 7.3

107 33.7

Liberal Dems

Note: statistical significance is reported for the difference between the proportion of projects (out of the total number of observations) for a particular regime in relation to that reported for liberal democracies, with *** (z-statistic < 1%), ** (z-statistic < 5%), and * (z-statistic