Regional Development Banks in Comparison: Banking Strategies versus Development Goals 1107163439, 9781107163430

Table of contents :
Cover
Half-title page
Title page
Copyright page
Dedication
Contents
List of Figures
List of Tables
Preface
Acknowledgments
Introduction
1 International Financial Institutions, Development, and Regional Development Banks
1.1 The Research Question in the Larger Context
1.2 The Importance of the Puzzle
1.3 Research Design
1.4 Development Assistance
1.5 Government’s Role
1.6 Cooperation
1.7 Globalization: Adapting to a New Reality
1.8 Conclusion
2 Multilateral Governance: Theoretical and Empirical Underpinnings
2.1 Who Rules?
2.2 Hypotheses and Research Design
2.3 Conclusion
3 Origins, Politics, and Structure of Regional Development Banks
3.1 Founding Principles
3.2 Comparing the RDBs
3.3 RDBs’ Capital Structure
3.4 Geography, States, and People
3.5 Lending Strategies
3.6 Conclusion: What Are They Like, What Do They Do
4 RDB Loans and Developing Countries
4.1 Data: Description and Sources
4.2 Variables and Explanation of Models
4.3 A Historical Look at RDB Loans
4.4 The State of Development
4.5 Selected Countries
4.6 A Comparison to the World Bank
4.7 Conclusion
5 Banks or Development Agencies?
5.1 Analysis
5.2 Findings
5.3 Comparison Summary
5.4 Conclusion
6 Political and Economic Constraints, Principals and Agents, and Prospects for Development
6.1 Political Interests and RDBs
6.2 Why Be a Donor?
6.3 Power to Borrowers?
6.4 Wealthy States, Poor States, and RDB Structures
Conclusion Future Outlook
Appendix 1 Life Expectancy and HDI Rank
Appendix 2 RDB Shareholders
Appendix 3 AsDB Professional Staff
Bibliography
Index

Citation preview

Regional Development Banks in Comparison

In a study that contributes to IR and IPE theory, Ruth Ben-Artzi raises substantive issues relating to aid, development, international relations and globalization. Regional development banks (RDBs), designed by politicians and economists to maneuver through labyrinths of economic, social, and political development, possess the potential to be central players in the long-term planning involved in healing and advancing poverty-plagued regions. However, RDBs in particular have received little attention. With a systematic analysis comparing four central regional development banks, this book explores why there’s a variation in strategy despite similar institutional design. The formal arrangements and raison d’être of RDBs are to assist developing countries in the process of poverty alleviation – a task that is often a risky investment. Focusing on the dichotomy between their banking and development roles, Ben-Artzi demonstrates that RDBs are potentially critical catalysts in the fight against poverty, even with their institutional limitations. Ruth Ben-Artzi is Associate Professor of Political Science at Providence College. Professor Ben-Artzi has been a post-doctoral fellow at the University of Pennsylvania’s Browne Center for International Politics and a visiting fellow at Sciences-Po in Paris, France. She holds a PhD from Columbia University and an AB from the University of Haifa.

Regional Development Banks in Comparison Banking Strategies versus Development Goals

RUTH BEN-ARTZI Providence College, Providence, Rhode Island

University Printing House, Cambridge cb2 8bs, United Kingdom One Liberty Plaza, 20th Floor, New York, ny 10006, USA 477 Williamstown Road, Port Melbourne, vic 3207, Australia 4843/24, 2nd Floor, Ansari Road, Daryaganj, Delhi – 110002, India 79 Anson Road, #06–04/06, Singapore 079906 Cambridge University Press is part of the University of Cambridge. It furthers the University’s mission by disseminating knowledge in the pursuit of education, learning, and research at the highest international levels of excellence. www.cambridge.org Information on this title: www.cambridge.org/9781107163430 © Ruth Ben-Artzi 2016 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2016 A catalogue record for this publication is available from the British Library. isbn 978-1-107-16343-0 Hardback Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party Internet Web sites referred to in this publication and does not guarantee that any content on such Web sites is, or will remain, accurate or appropriate.

To Andrew

Contents

List of Figures List of Tables

page ix xi

Preface Acknowledgments

1

2

3

xiii xv

Introduction International Financial Institutions, Development, and Regional Development Banks 1.1 The Research Question in the Larger Context 1.2 The Importance of the Puzzle 1.3 Research Design 1.4 Development Assistance 1.5 Government’s Role 1.6 Cooperation 1.7 Globalization: Adapting to a New Reality 1.8 Conclusion

7 10 12 15 20 25 31 35 39

Multilateral Governance: Theoretical and Empirical Underpinnings 2.1 Who Rules? 2.2 Hypotheses and Research Design 2.3 Conclusion

41 43 55 61

Origins, Politics, and Structure of Regional Development Banks 3.1 Founding Principles 3.2 Comparing the RDBs 3.3 RDBs’ Capital Structure

63 64 69 87

vii

1

viii

Contents 3.4 3.5 3.6

4

5

6

Geography, States, and People Lending Strategies Conclusion: What Are They Like, What Do They Do

RDB Loans and Developing Countries 4.1 Data: Description and Sources 4.2 Variables and Explanation of Models 4.3 A Historical Look at RDB Loans 4.4 The State of Development 4.5 Selected Countries 4.6 A Comparison to the World Bank 4.7 Conclusion Banks or Development Agencies? 5.1 Analysis 5.2 Findings 5.3 Comparison Summary 5.4 Conclusion Political and Economic Constraints, Principals and Agents, and Prospects for Development 6.1 Political Interests and RDBs 6.2 Why Be a Donor? 6.3 Power to Borrowers? 6.4 Wealthy States, Poor States, and RDB Structures Conclusion: Future Outlook

90 98 107 109 110 117 121 131 140 146 149 151 153 157 182 192 194 196 219 226 232 236

Appendix 1: Life Expectancy and HDI Rank

245

Appendix 2: RDB Shareholders

248

Appendix 3: AsDB Professional Staff Bibliography

257 259

Index

269

Figures

3.1 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 4.10 4.11 4.12 4.13 4.14 4.15 4.16 4.17 4.18 4.19 4.20 4.21 4.22 5.1

The capital structure of RDBs IDB loans, 1961–2009 AsDB loans, 1966–2009 AfDB (gross) loans, 1967–2009 AfDB (net) loans, 1966–2000 EBRD loans, 1991–2009 FDI to developing countries, 1970–2000 FDI to Africa, 1970–2002 FDI to Latin America and the Caribbean, 1970–2009 FDI to Asia, 1970–2009 FDI to developing Europe, 1970–2009 IDB net loans to Argentina, 1961–2009 IDB net Loans to Haiti, 1961–2009 AsDB net loans to Singapore, 1967–2009 AsDB net loans to Bangladesh, 1967–2009 AfDB net loans to Tunisia, 1967–2009 AfDB net loans to Sierra Leone, 1967–2009 EBRD net loans to Slovenia, 1991–2009 EBRD net loans to Uzbekistan, 1991–2009 IDB and WB gross loan disbursements to the Americas, 1961–2000 AfDB and WB gross loan disbursements to Africa, 1969–2000 AsDB and WB gross loan disbursements to Asia, 1968–2000 EBRD and WB gross loan disbursements to developing Europe, 1994–2000 The effect of FDI on loan amount (US $) for the IDB ix

page 88 124 126 128 128 130 133 134 135 136 137 141 141 142 142 143 144 145 145 147 148 148 149 160

x

List of Figures

5.2 The effect of infant mortality rate on loan amount (US $) for the IDB 5.3 The effect of US affiliation during and after the Cold War on loan amount (US $) for the IDB 5.4 The effect of FDI on loan amount (US $) for the AsDB 5.5 The effect of infant mortality rate on loan amount (US $) for the AsDB 5.6 The effect of Japan affiliation during and after the Cold War on loan amount (US $) for the AsDB 5.7 The effect of infant mortality rate on loan amount (US $) for the AfDB 5.8 The effect of Cold War on loan amount (US $) at different levels of polity for the AfDB 5.9 The effect of US affinity on loan amount (US $) for the AfDB 5.10 The effect of FDI on loan amount (US $) for the EBRD 5.11 The effect of primary education on loan amount (US $) for the EBRD 5.12 The effect of infant mortality rate on loan amount (US $) for the EBRD 5.13 The effect of polity score on loan amount (US $) for the EBRD 5.14 The effect of hegemon affiliations on predicted loan amount (US $) for the EBRD 5.15 The effect of infant mortality rate on loan amount (US $) at high (90th percentile) and low (10th percentile) levels of foreign direct investment for the IDB, AsDB, AfDB, and EBRD 5.16 The effect of primary education rate on predicted loan amount (US $) at high (90th percentile), median and low (10th percentile) levels of foreign direct investment for the EBRD 5.17 The effect of US affinity on loan amount (US $) at high (90th percentile) and low (10th percentile) of infant mortality rate and during and after the cold war for the IDB 5.18 The effect of hegemon affiliations on loan amount (US $) during and after the Cold War for the IDB, AsDB, and AfDB

161 164 167 168 169 173 174 175 179 180 180 181 183

185

187

189

190

Tables

2.1 The voting power of largest shareholders in the MDBs (2002) 3.1 The MDBs in comparison (2009) 3.2 IDB borrowers by group 3.3 AfDB country classification 3.4 AsDB member countries (2002) 3.5 Presidents, staff, and shareholders of the RDBs 3.6 Net capital flows to developing countries, 1986–1995 (at constant 1995 prices) 4.1 Top ten recipients of concessional and nonconcessional loans, 2010 4.2 IDB authorized and disbursed funds, 1961–1970 4.3 GDP per capita and infant mortality rate, by region 5.1 Variables 5.2 IDB loans per capita 5.3 AsDB loans per capita 5.4 AfDB loans per capita 5.5 EBRD loans per capita 5.6 Results (summary) 6.1 Hegemony at the four RDBs 6.2 US abstentions and negative notes on loans approved by the IDB Appendix 1: Life Expectancy and HDI Rank Appendix 2: RDB Shareholders Appendix 3: AsDB Professional Staff

xi

page 45 70 73 81 82 91 99 113 122 139 156 158 165 171 177 184 199 204 245 248 257

Preface

My interest in international organizations, development, and aid began long before graduate school. While an undergraduate student at the University of Haifa in Israel, I joined the Society for International Development (SID), an international NGO (nongovernmental organization). I quickly became very involved in the Israeli chapter of the organization, founded its local Youth Chapter, attended international conferences on its behalf, and, with the organization’s support, helped draft Israel’s report on the status of women for the UN’s Fourth World Conference on Women. At the time, I found the subfield of international political economy a refreshing revelation: growing up in Israel, where conflict is a constant presence, and having served in the Israeli Defense Forces (IDF) during the first Intifada and the first Gulf War, I was enamored with approaching the study of international relations from an angle of cooperation. I became interested in interactions between states and nonstate global actors that did not constitute war or conflict. It struck me that in the post–Cold War world that was just emerging, understanding global relations between countries and societies through the lens of inequality and cooperation could provide more enduring policy solutions for conflict. I further developed this interest after graduation, when I took a research assistantship at the Development Centre of the Organization for Economic Cooperation and Development (OECD) in Paris. In that capacity, I helped create a directory of European NGOs that work in the field of development, attended Inter-American Development Bank and Asian Development Bank conventions, Development Assistance Committee (DAC) deliberations, and many lectures on sustainable xiii

xiv

Preface

development. Although I did not know then that my dissertation research and subsequent book would focus on the Regional Development Banks, I was certain that in my graduate work I would explore the role of international institutions in remedying global inequalities. During the same period, the Oslo Accords between Israel, the Palestinian Authority, and the US were signed and their implementation begun. With the cautious optimism that followed this agreement, many thought the Israeli-Palestinian conflict might be heading for a resolution, and I learned that a “blue print” was being developed for a Middle East Development Bank. As the only developing region without a dedicated development bank, I questioned whether a development bank could help foster cooperation among Middle Eastern countries, and who would benefit from such a bank. By the time I had to articulate a dissertation topic in graduate school, it was clear to me that RDBs warranted further examination: A detailed comparative analysis of these banks would provide necessary insight into the expectations we should have from development banks in general, and a potential Middle Eastern Development Bank, more specifically. At this writing, a Middle East Development Bank seems more unlikely than ever, but other development banks – such as the New Development Bank (NDB) and the Asian Infrastructure Investment Bank (AIIB) – have emerged, or are in the process of being created. In fact, the model of “development banking” has expanded rather than contracted, despite widespread anti-IFI (international financial institution) protests in the early 1990s (e.g., the “Fifty Years Is Enough” campaign). In this respect, I hope that a deeper understanding of these institutions would facilitate more careful planning and realistic objectives when new development banks emerge.

Acknowledgments

As a SID member-activist, during my year at the OECD Development Centre, and later in graduate school, while I was conducting research for my dissertation, I met many individuals who work in international and development institutions and affiliated agencies. Most of them are passionate about the work they do. Some are truly idealistic, while others are “recovering idealists” – those who had high hopes but have come to terms with the limitations of international institutions. All were extremely knowledgeable about their respective organizations, development and aid, and the policy preferences of their home states. They generously shared their experiences and insights. My deepest gratitude goes to these practitioners from whom I have learned a great deal and without whom I would not understand how development institutions work. I am especially indebted to Henny Helmich whom I first met during my SID days, and who was later my supervisor at the OECD Development Centre. Henny’s mentorship during my year at the OECD and beyond helped shape my grasp and views on development aid. He facilitated my attendance at special meetings for the IDB and AsDB, peer reviews of the Development Assistance Committee, and made possible my appointment as a Development Centre UN observer while I was in graduate school, which exposed me to UN meetings and briefings. Later, when I was doing research for my dissertation, Leo Harrari, at the IDB Special Office in Paris, provided access to data not publicly available, as well as many great stories. A few years later, Carlos Jarque filled in some of the missing pieces and also recounted some important examples from his days in the Mexican government. Many other IDB staff in the Washington, DC, headquarters and the Paris office patiently xv

xvi

Acknowledgments

answered my questions with important insights. At the EBRD, I received a similar warm welcome during my research visits, including a peek into the Penthouse designed by Jacques Attali. Scholars at the OECD Development Center helped me understand the AsDB and AfDB, especially Jean Claude Berthelemy; AsDB and AfDB staff members visiting the OECD were generous with their time, and officials at both development banks provided detailed answers to my written questions. But I would not have considered pursuing a PhD and going into academia without the guidance of Zeev Maoz, who was an exceptional mentor when I was an undergraduate student. He is responsible for my introduction to the study of international relations and with his encouragement and steadfast support I went to graduate school. As his research assistant, first at the University of Haifa and then at the Jaffe Center for Strategic Studies in Tel Aviv University, I learned more about how to conduct research and how to work diligently, than from any instructor in any class I have ever taken. Zeev’s dedication to high quality scholarship and to mentoring students is inspiring. Naomi Chazan was also an early mentor, setting an example – and a high bar – as a scholar and teacher, an activist, a politician, and, above all, a strong woman. In graduate school at Columbia, I was fortunate to have fantastic dissertation advisers. Helen Milner, especially, pushed me to ask tougher questions and skillfully guided me through the dissertation research and writing process to its completion. Her professionalism and prompt responses to inquiries as well as extremely quick turnaround of drafts of chapters I gave her kept me on my toes. Charlie Calomiris, challenged me to examine some of the questions I was asking in a more complex, refreshing way, and even when we didn’t agree, our conversations challenged my thinking and improved my clarity. David Epstein helped identify data and guided me through its analysis. David Baldwin gave me some of the soundest advice, early on, which remains relevant: he said that at certain junctures, time spent stepping back and assessing the “big picture” – what I am actually trying to say – is more productive than racing with data collection, analysis, and writing. I was also lucky to have a wonderful cohort of friends in graduate school. Never having lived in NYC and arriving alone from Israel only two weeks before the start of classes, not knowing anyone in the city, I would have been completely lost without these friendships. Not all were in my field of IPE and some were ready to graduate when I arrived, while others started their doctorate with me or after me. Our group at Columbia

Acknowledgments

xvii

was smart, fun, and supportive. Many of these friends have published books that I proudly display; they remind me of the intellectual impact of my peers on my own thinking. These friends are many more than I can mention, but most notably, Shylashry Shankar, Shanker Satyanath, Samir Awad, Amitabh Dubey, Tim Buthe, Stacie Goddard, Ron Krebs, Allison Kingsley, Evan Reznick, Eleonora Passoti, Megumi Naoi, Roger Schoenman, and Dawn Brancati. This work has received institutional support, without which I would not have been able to carry out the research. Columbia University’s Presidential Fellowship and the Political Science Department, The Chazen Center for Business Education at Columbia, and Sciences Po in Paris, all provided funding during the dissertation-research phase of this project. The University of North Carolina (Wilmington) graciously hosted me as I was completing the writing of the dissertation, and the Browne Center for International Relations at the University of Pennsylvania generously supported me while I prepared my dissertation for deposit and transitioned to a post-Doctoral life. At the Browne Center, Ed Mansfield was a guide, a friend, and studious critic of my work – at a time I needed it most. Others who read earlier versions of parts of this book, were generous with their time to discuss theory, methodology, findings, development banks, book-writing strategy, and facilitated my ability to complete the book, include Joseph Stieglitz, William Easterly, Mark Blyth, Mike Tierney, Christopher Kilby, Tanisha Fazal, Barbara Upton, Dan Miodownik, Lilach Nir, Shirli Kopleman, Pauline Luong-Jones, Rebecca Weitz-Shapiro, Tulia Falleti, Melani Cammett, Doug Blum, Susan McCarthy, Bill Hudson, and Carly Iafrate. Also, participants at the IGERT symposium at Columbia (2004), fellow panelists and discussants at the 2005 ISA and APSA annual meetings, and attendees at my talk at the University of Pennsylvania’s Browne Center that same year, all provided important insights, critiques, and asked questions that helped shape this book. I am particularly indebted to Dawn Brancati and Lisa Minca for their expert instruction in statistical analysis. My research assistants at Providence College – Cara Bragg and Shannon Hulst – diligently examined annual reports of RDBs and helped me update some of the data for this book. For support during the long process of transforming my dissertation to this book, including conducting substantial additional research, I am thankful to Providence College’s institutional backing, and particularly to my colleagues at the political science department who patiently waited

xviii

Acknowledgments

for this book to come to fruition. Lewis Bateman at Cambridge University Press has been extremely supportive over these years and brought this book to the production phase. Robert Judkins and Claudia Bona-Cohen of CUP guided me through the book’s production, and the two anonymous reviewers who read this manuscript provided immense guidance: their suggestions have notably improved this book and for that I am extremely grateful. I hope this final product lives up to their expectations and the expectations of colleagues who advised, read, and remarked over the years; all errors, of course, are my own. Finally, institutional and peer support alone would not have been sufficient resources for completing this book. As a mother of three whose children were all born during the time I was working on this project while at Providence College, I would not have succeeded in combining family and work if it were not for a phenomenal network of parents, babysitters, and teachers who are my community. To all my friends and child-care providers who, on many occasions, took my kids to day care and school, picked them up, fed them, played with them, and have generally been my compass – pointing me to what is most valuable – I am enormously appreciative. My family in Israel has always, lovingly, motivated me to push a little more. My grandparents, Moshe and Dvora Kestler, knew I was working on a book before they passed away, and were so proud and encouraging. They never questioned or doubted and were my safety net. My father, whose academic work ethic is inspirational, and my mother, who is not only sincerely interested in my research, but also spends time searching for news reports and clipping them for me, making sure I don’t miss any commentary on development banks, have both been a source of encouragement throughout these years. They have also been dedicated grandparents, by far their most important role. My brothers and sister-in-law have all, at various times and in various cities, provided company, laughs, meals, and overall intelligence – and kept me grounded. I could not have married into a more loving and supportive family. My father-in-law and his wife and my sister-in-law have always taken an interest in my work and knew when not to ask if the book was finished. My husband’s extended family welcomed me into their homes for holidays, weekends, and many festive celebrations. They have supported our family in innumerable ways, have been the most awesome aunts, uncles, and cousins to our kids, and have made it easy for me to adopt America as my new home.

Acknowledgments

xix

My kids – Tomer, who is now old enough to flip through the pages of my manuscript and ask questions; Sivan, who is so excited that I have been working on a book that she decided to write one herself; and Oren, whose first birthday we are about to celebrate as I write these words – always come first. They have changed my life and showed me what is most precious. I love you and nothing is more important to me than your happiness. But it would not be possible for me to enjoy my family and complete this book without the unswerving love and support of my husband, Andrew, to whom this book is dedicated. He has made more sacrifices with his time and career than I can count, and he accompanied me in this journey from the beginning – from my early days in graduate school, when we first met, through the tribulations of writing a dissertation – and finishing it, multiple moves, job searches, pregnancies, infants, family deaths, happy and difficult times. Andrew read so many drafts of this book that he could probably recite it in his sleep. He’s been my biggest advocate and toughest critic. Not having written a paper in English until I arrived in graduate school, he taught me how to become a better writer and more precise thinker. And he did all that and sustained the home front while holding a full time job. Balancing family with an academic career would be impossible without this genuine partnership that we have. I owe Andrew my love and thanks – for making it all possible.

Introduction

In his renowned novel Great Expectations, Charles Dickens explores class differences and struggles in nineteenth-century England through the story of a young boy named Pip. Living in a poor fishing town, Pip one day receives a huge sum of money from an anonymous source. The money is kept in a trust, intended to support Pip’s transformation into a gentleman. Accepting the generous offer and the conditions that come along with it, Pip moves to London. He quickly becomes well-versed in the ways of the aristocracy, and more educated and sophisticated than he would ever have otherwise been. One day, the generous donor appears. To Pip’s surprise, it is not the wealthy old lady whose niece he used to play with (whom he always suspected was responsible for his fortune), but a convicted runaway criminal whom he helped to hide when he was very young. This revelation poses some serious moral dilemmas: Is there an importance to the nature of the source of the money that meant to benefit Pip? Did it really benefit him, though it uprooted him from his home under the assumption that a privileged life among the nobility is superior? What power does the source of his finances have over his life, and what kind of demands can his benefactor make of him? Finally, given these circumstances and dilemmas, would Pip have been better off not accepting the offer? The moral questions that are raised by Dickens’ novel on an individual level with respect to Pip, his personal destiny, and the choices he is given may also capture the complex relationship between donor and borrowing states within the arenas of international financial institutions and their policies.1 1

These can obviously also be posed with regard to bilateral aid, but that is not the central matter of this study.

1

2

Introduction

Ultimately, the concerns preoccupying the critics and supporters of these institutions are very similar to those presented by Dickens: In what way can or should donors impose their policies on those receiving aid? Are these institutions doing more harm than good? Would developing countries be better off integrating into the world economy at their own pace, with no outside influence? And, finally, would it be realistic to expect donor countries to provide aid but forgo imposing conditions on the developing countries receiving the aid? While this research cannot begin to answer all of these complicated questions, it provides a first and important step in the study of regional development banks, which belong to the family of international financial institutions (IFIs). It explores the internal dynamics of these institutions, analyzes how they approve loans, and places the disbursed loans in the context of the socioeconomic and financial status of the recipient countries. The donor countries that support these institutions, although not anonymous, have certain agendas that they wish to advance. If it were not for their own self-interest, these wealthy nations might not engage in development efforts in the first place. Yet while the borrowing members of the institutions are eager for the assistance, critics of these institutions often claim that IFIs contribute to developing countries’ debt in a way that exacerbates these countries’ problems rather than constructively advancing their development. Naturally, donors wish to retain control wherever possible, while borrowers strive for more independence and power over their own destinies. RDBs have been a central vehicle for multilateral assistance since the 1960s. While the Inter-American, Asian, and African Development Banks (IDB, AsDB, and AfDB) were created as a result of Cold War geopolitics – in an effort to expand American influence – the official justification for their creation was to provide poor countries with a bigger say in multilateral banks that affect their development. It would follow, then, that the RDBs should be spared the harsh criticism generally directed at IFIs, particularly with regard to donor control. But whether RDBs serve developing countries better than global IFIs by virtue of their membership composition and the voting shares that are allocated to borrowing (developing) members is a question that still lingers. To operate properly, it may prove advantageous (or even imperative) that multilateral banks have a nonregional, nonborrowing hegemon; however, this would then undermine the stated rationale for the RDBs existence, that is, that aid institutions managed and run by the recipients of the aid are in a better position to contribute to the goals of poverty alleviation and development.

Introduction

3

Moreover, the fact that RDBs, despite greater stakeholder involvement, fall into the same policy related “traps” that the World Bank and IMF are so heavily criticized for might indicate that institutional mechanisms have a greater impact than the preferences of the institutions’ member states. The research presented here has implications for numerous theoretical and policy debates associated with multilateral financial institutions and development aid. These include whether and how cooperation works, and the power politics and policy agendas engendered in multilateral organizations. Furthermore, it considers questions about the merits of aid and multilateralism, including the way multilateral aid/lending institutions work; the effect aid may have on developing countries; and the fundamental question of whether developed countries use multilateral institutions to advance their own interests. The theoretical foundations for this book, rooted in the ongoing realismliberalism (neoliberal institutionalism) debate, as well as more recent constructivist approaches to cooperation, provide a useful focal point for the study of RDBs. The comparison of four institutions allows for an assessment that goes beyond interests, preferences, and outcomes. It provides a platform from which differences in norms, ideas, and culture, as well as hegemonic configuration, can be analyzed. The synthesis of these issues requires careful analysis. Although these underlying themes may be theoretical in nature, the ideological positions that they generate have a strong bearing on the “real world.” An understanding of the theoretical premise of development can shed light on the approaches, critiques, and possibilities of development assistance and poverty alleviation policies. The subject of assisting developing countries has become central to academic and policy debates since the end of the Cold War. Bilateral assistance has always been considered a useful foreign policy tool. But even the bilateral avenue is scrutinized, to a certain extent, by the international community: in 1970 the UN General Assembly passed a resolution that the world’s rich countries contribute 0.7 percent of their gross national product (GNP) to aid. Over the years, this pledged target has been affirmed in many international commitments.2 Furthermore, the Organization for Economic Cooperation and Development (OECD) publicizes the actual amount allocated to aid by OECD countries and

2

Examples include the 2002 International Conference on Financing for Development in Monterey (Mexico), the World Summit on Sustainable Development in Johannesburg held the same year and more recently, the Paris Declaration on Aid Effectiveness in 2005, and the forums on Aid Effectiveness in Accra (2008) and Busan (2011).

4

Introduction

examines the actions of individual members through the Development Assistance Committee (DAC) in an effort to hold rich countries accountable. Multilateral assistance, one would think, should be less prone to political influences because members must compromise. However, in recent years there has been a surge of activists from nongovernmental organizations (NGOs) and various grassroots movements who are waging a “war” against multilateral financial institutions (namely the World Bank, IMF, and WTO) and other international financial entities that are perceived to be dominated by wealthy countries or large corporations (such as the G7 and the annual International Economic Forum at Davos). For example, the “Fifty Years Is Enough” NGO, one of the leading organizers of the campaign against multilateral financial institutions, describes the World Bank and the IMF in the following way: The IMF and World Bank have been empowered by the governments which control it (led by the US, the UK, Japan, Germany, France, Canada, and Italy – the “Group of 7,” which holds over 40% of the votes on their boards) with imposing economic austerity policies in the countries of the so-called “Third World” or “global South.” Once Southern countries build up large external debts, as most have, they cannot get credit or cash anywhere else and are forced to go to these international institutions and accept whatever conditions are demanded of them. None of the countries has emerged from their debt problems; indeed most countries now have much higher levels of debt than when they first accepted IMF/World Bank “assistance.” (www.50years.org)

Sights of protesters waving banners, chanting, and at times clashing with police forces, whose ranks are heavily reinforced at locations of meetings and conferences of the international financial institutions, have become all too familiar. These events warrant careful investigation and analysis. This book and its conclusions provide a platform for understanding the complex decision-making processes and internal political makeups of multilateral financial institutions. By analyzing the history of loans these institutions made in the context of the socioeconomic status of the countries that borrow from them, this study lays out unique and systematic insights into RDB activity patterns over time. Comparing the four development banks yields important conclusions about the role of principalagent relations and institutional culture in generating aid policies. This, in turn, provides a better understanding of their impact and effectiveness. Initially, I thought this project would provide clear-cut answers: The strong and vocal ideological stance of all sides of the debate concerning multilateral financial institutions could easily lead one to believe that one must be either “for or against.” However, as I accumulated more

Introduction

5

data, interviewed more RDB officials, and visited the headquarters of two of the four RDBs in this study, I realized that just as the problems faced by poor countries and their relations with wealthy countries are complex, so are the strategies undertaken by RDBs (and, by extension, other multilateral financial institutions). Although a central characteristic of the RDBs is that they are different in membership from their international counterparts (the World Bank and the IMF), they are related in that they are lenders to developing countries. The people who work in these institutions are not all cold-hearted bankers or calculating economists. Apart from some obvious misguided policies, some of the traps they fall into are bureaucratic, some are political, and some are simply based on lack of knowledge and processes of trial-and-error. Since these institutions are not likely to be eliminated in the foreseeable future, we should strive to understand how they really work. Fortunately, an emphasis on reforming IFIs has been a major concern of government officials, scholars and think tanks, and even IFI’s own leadership and staff. For example, the RDBs have been studying (and to a varied and limited extent, implementing) productive avenues for reform; the World Bank has conducted numerous internal studies on its own reform needs; and in 2000, the US government commissioned a bipartisan assessment on the merit and future of these institutions (the International Financial Institution Advisory Commission). Those who advocate that we reconsider how wealthy countries assist poor ones serve us well when they present alternatives and critique the current policies of these institutions. At the outset of the twenty-first century, and with the changed nature of global politics and finance since the end of the Cold War, understanding the creation and endurance of IFIs is crucial if we want to engage in useful debates concerning their policies, effectiveness, and necessity. The interaction between developed and developing countries has become the focus of global political and economic agendas, and development banks are key players in these relationships. They are central to the study of international political economy and security studies alike. As developed and developing countries are encountering more security-related friction, some of which can be traced to poverty in the developing world, it becomes apparent that securing cooperation through multilateral institutions can serve both the international economy as well as global security. But these institutions must work well for this task to succeed. If they generate angry protests and cynicism then their mere existence may be counterproductive to the desired pacifying impact of cooperation.

6

Introduction

The four RDBs analyzed in this book are representative of the multitude of multilateral financial arrangements that exist today, and this study therefore brings us a step closer to an improved theoretical analysis of this phenomenon, as well as a basis for better policies for sustainable development. By examining the four RDBs – comparing their structures and policies – this book offers a novel analysis of IFIs. First, it demonstrates the tension between the development mission of these institutions and their bank-like features. Second, it shows that the way in which the institution is governed impacts its ability to operate more like a development agency. Overall, despite the difference in configurations of hegemons, all the RDBs struggle to generate more “financially risky,” albeit mission-focused, policies. Further, while it may be expected that an institution with an active and influential hegemon would advance financially prudent policies, the AfDB, an institution governed mostly by borrowers, without an external, active hegemon, also appears to prefer financially cautious policies. Understandably, Multilateral Development Banks (MDBs) are concerned with their financing, which provides their borrowers the support they seek. However, the way in which the institutions are structured and governed determines their ability to truly be development institutions. Finally, the more recently created European Bank for Reconstruction and Development (EBRD) experimented with an entirely different mission – one that does not shy away from political goals (democratization and privatization), and does not try to hide the interests and status of its non-borrowing members. In that sense, it is more like a bank with political goals rather than a political institution with financial goals. The rest of this book carefully lays out the histories of the RDBs, maps their lending over the years, and places the data analysis in the context of the stories of each of these institutions. Following a detailed introduction to multilateral development banks (Chapter 1), Chapter 2 presents the analytical and theoretical issues at stake when it comes to regional cooperation and economic development. Chapter 3 presents the history and the background of each of the four major RDBs. The data employed in this study are presented in Chapter 4, along with examples of specific countries, while Chapter 5 applies rigorous statistical analysis. The conclusions of this analysis are supplemented in Chapter 6 with qualitative data, primarily from numerous interviews and secondary sources. A brief conclusion summarizes the book, suggests new avenues of research, and provides policy advice along with speculation as to the role of RDBs in the future.

1 International Financial Institutions, Development, and Regional Development Banks

This association of poverty with progress is the great enigma of our times. It is the central fact from which springs industrial, social, and political difficulties that perplex the world, and with which statesmanship and philanthropy and education grapple in vain. From it come the clouds that overhang the future of the most progressive and self-reliant nations. It is the riddle which the Sphinx of Fate puts to our civilization, and which not to answer is to be destroyed. So long as all the increased wealth which modern progress brings goes but to build up great fortunes, to increase luxury and make sharper the contrast between the House of Have and the House of Want, progress is not real and cannot be permanent. – Henry George, Progress and Poverty, 1879

In this era of globalization – where international financial markets affect developed and developing countries alike, and where financial flows as well as economic crises spread across borders and oceans with remarkable ease – the role of multilateral financial institutions in the international financial scheme is crucial. These public institutions, designed by politicians and economists to navigate the labyrinth of economic, social, and political development, possess the potential to be central players in the long-term planning involved in healing poverty-plagued regions and advancing their economies. This formidable task offers the invaluable payoff of furthering global economic and financial stability and improving human welfare. Consequently, the issues of aid and development that are related to globalization have been at the center of academic debates. Multilateral financial institutions, designed to implement global financial policies, have been an integral part of these debates and have come under 7

8

IFIs, Development, and Regional Development Banks

increasing scrutiny. Critics vary from those who believe that these institutions do more harm than good by contributing to the “moral hazard” problem1 and should therefore not exist, to those who believe that they can play an important role in development, but need to be reformed (Stiglitz, 2002, and Sachs, 2005). But even if one begins in the second, more positive, camp, there is still substantial disagreement as to how these institutions should function so that they can be both constructive and efficient. Much of this controversy stems from limited knowledge about how multilateral financial institutions work and what they are designed to do. Additionally, the financial crises of recent decades (the Mexican Peso Crisis in 1995, the Asian Financial Crisis in 1997, the Russian crisis in 1998, the crisis in Argentina in 2001, and the 2008 US credit crisis) have not only highlighted the potential of international financial institutions to be central actors in efforts to correct market failure, but also signified the fluid changes in the international economic system and the importance of institutional flexibility in addressing new challenges effectively. Since the late 1980s, RDB lending actually exceeded that of the World Bank. Still, RDBs have received little academic attention compared to other multilateral financial institutions.2 Evidently, with loan commitments ranging from US $4 billion to over US $12 billion in 2011 for each of the four banks compared to the World Bank’s (IBRD) US $26.7 billion loan commitments that same year, the RDBs are certainly central actors in the regions that borrow from them. Further, in a volatile global economy, lenders of last resort could conceivably be essential for economic recovery. For example, the recent credit crisis and its spread to the developing world, makes the presence of IFIs even more crucial: as private banks scale back their lending to poor countries, these public institutions will surely bear the brunt of the burden of assisting developing countries – not only to continue the efforts already made, but also to prevent a large scale credit crisis, given these countries’ accrued debt. The central puzzle of this study stems from the conflict between the development function of RDBs and their bank function.3 As this research 1

2

3

That is to say, developing countries follow inefficient and, sometimes destructive, financial policies knowing that if a crisis occurs they will be bailed out (see Milton Friedman, 1992; also, for the debate on the need for aid in general, see Moyo, 2008 and Easterly, 2001). See Culpeper (1997), Krasner (1981), Neumayer (2003), Babb (2009), and Strand (2010) for the few comparative academic discussion of the regional development banks. By “conflict” I mean that these two functions often pull the banks in contradictory directions, forcing them to prioritize one over the other.

IFIs, Development, and Regional Development Banks

9

shows, these institutions embody structures and operational policies that are similar to those of private banks and corporations. This means that considerations of loan repayment by borrowers (developing countries) are central to the RDBs’ decision-making. And by making financial considerations a priority, development needs can be marginalized. Moreover, as their membership is comprised of states, decisions can often be the result of foreign policy preferences and, therefore, RDBs’ policies can potentially become extensions of member states’ policies, compromising their central role of agents of development. Nevertheless, RDBs’ mission is to fill a certain development gap and (multilaterally) assist developing countries in the process of poverty alleviation and development. This task frequently requires formulating and following policy that is financially risky or that does not pay off in the short run. These conflicting identities are set against the backdrop of the changing paradigms dominating the field of development finance. Specifically, since the central operation of these institutions is loan-making, the line between aid (to developing countries) and financial and political motives (of developed countries) can sometimes be blurred.4 Further, the bank financing, that is, the credit, that is available in developed countries is not available to the same extent in developing countries. Development-focused assistance should therefore be designed to supplement financial markets, overcome market failure, and provide credit to support projects that developing countries’ governments cannot afford, thereby aiding in processes of development that are often not lucrative for private sector investors.5 Multilateral channels of development assistance permit many different potential donors to coordinate their efforts, and yet allow developing countries to avoid direct dependence on one single donor. However, this institutionalization of development is problematic because financial markets and private financial flows are dynamic in nature, making development lending a moving target. Thus, the environment in which RDBs operate is continuously changing. It is at these junctions of change that development banks need to continually reinvent themselves in order to remain on the path to fulfill the goals they were set up to achieve.6 4

5

6

This also contributes to multilateral development banks’ problematic positioning in the court of public opinion. The recent banking crisis in the US has demonstrated that private credit institutions often operate with a preference for short-term profits and little, if any, long-term planning. In this study, I mostly refer to “poverty alleviation” and “development” collectively. However, it should be emphasized that by no means do I regard them as synonymous. In

IFIs, Development, and Regional Development Banks

10

Thus, the research question guiding this study is: How do the RDBs balance their bank features with their development mission? To explore this puzzle this book first analyzes the history and foundation of the RDBs, and then explores the way they operate and how – and whether – they have changed over time.

1.1

the research question in the larger context

Recently, with accelerated changes in the global financial markets, the role of multilateral financial institutions has been scrutinized and reconsidered. Many of these institutions were established by Western states in the aftermath of WWII and during the Cold War, with the goal of helping advance less developed economies. These institutions, such as the World Bank (WB) and the International Monetary Fund (IMF), were fundamentally a political initiative and are still a political enterprise. However, since the loan mechanism, policies, and the financial assistance they provide are frequently the targets of controversy, IFIs often try to elude their political makeup and portray themselves as economic-driven organizations, characterized by professionalism and immunity to the “political game.”7 The goals of these institutions are noble: help developing countries overcome the vicious cycle of poverty and eventually become part of the “club” of developed nations. However, it is impossible to imagine that such efforts can be carried out in a nonpolitical way. First, setting up a financial institution in which states are members and shareholders8, and monetary contributions are reviewed and decided upon by legislators, is political in nature. Second, the threshold as to who is categorized as developing and to what degree, and who can become a member, is prone to political influences. Third, since resources are limited while demand is immense, setting priorities among so many deserving requests for loans embodies, in part, a political preference.

7

8

fact, many students of development stress that they are not: poverty alleviation does not necessarily mean that development has occurred, and successful development does not imply the alleviation of poverty. The idea of overcoming market failures embodies both political and economic attributes: It is pursued by states and is thus a political decision by nature, while it involves financial and economic means, resources, and strategy. According to Barry Metzger, former General Counsel to the Asian Development Bank (interview May 2001), shareholders (states) play a very intrusive role in the activities of the RDBs.

1.1 The Research Question in the Larger Context

11

In 1999, the US Congress commissioned a report by experts and political appointees from each side of the political aisle designed to assess the role of multilateral financial institutions (namely, the IMF, the WB, the RDBs, and the World Trade Organization). In March 2000, the International Financial Institution Advisory Commission (IFIAC) concluded in its report that: “A large gap remains between promise and achievement” of the international development banks (p. 1). It also recommended changing the names of the World Bank and the regional development banks to “global development agency” and “regional development agencies.”9 Although seemingly semantic, the proposed change of name underscores the conundrum that is central to this book: What are development banks? Can an institution whose main concerns are growth and poverty alleviation in developing countries be called a bank? The puzzle is thus inherent in the name of these institutions: development banks. Do regional development banks operate in a way that is similar to that of commercial banks? Or, are RDBs agencies that seek political, economic, and social development in their regions?10 More importantly, under which conditions can we expect these institutions to operate like banks, and under which conditions to act like development agencies?11 I suggest that the name currently used by these institutions is part of the design to avoid the implication of politics. After all, most people do not think of a bank as a political institution. The word “bank” connotes objectivity, efficiency and factual, data-dependent decision-making.12 But over time, the organization calling itself a “bank” can become confused as to its identity: if it is expected to operate as a bank (prone to auditing firms’ ratings of financial standing, structured like a corporation, 9

10

11

12

“To underscore the shift in emphasis from lending to development, the name of the World Bank would be changed to World Development Agency. Similar changes should be made to the regional development banks” (IFIAC Report, p. 9). In the context of this question, one must ask, what is development? This can be examined both by analyzing the meaning the banks themselves give to this term, and by the analysis of academic studies on development (such as Przeworski, 1991; Evans, 1995; and Haggard and Kaufman, 1992). In the midst of the current financial crisis, the difference between public/governmental financial strategy and that of private institutions is striking: Public agencies with government backing have a preference for providing welfare in order to stabilize the economy, even if the credit extended is long-term and the prospects for financial returns remain unclear. This does not mean that banking policies are always sound: it simply denotes that private banks are institutions that make decisions based on projections of profit rather than political benefits.

12

IFIs, Development, and Regional Development Banks

and making decisions based on financial returns), then how does development assistance, especially to the most needy in the highest-risk areas, come into the picture? As Culpeper noted in his summary of the NorthSouth Institute study, the RDBs initially served as liaisons between capital and developing countries.13 At that time, they essentially provided the sole source of capital financing. However, as private capital flows to developing countries increased, the RDBs began struggling with their role and mission.14 It is therefore apparent that as international conditions change, RDB lending programs that were at one point essential for development are no longer the only source of financing for developing countries. Thus, even if the programs of the institutions, deemed development-oriented, have not changed over the years, their impact and importance is altered since the environment in which they operate has experienced vast changes. To summarize, multilateral financial institutions have lately faced criticism not only from long-time objectors in the US Congress and/or government, one of their most powerful shareholders, but also from some traditional strong supporters – in governments and NGOs alike.15 The debate over the necessity and role of these institutions is a result of the conflicting interests or goals of the development banks themselves, which causes confusion and misperception among both supporters and critics as to what to expect of them and how, possibly, to change them.

1.2

the importance of the puzzle

The puzzle is, therefore, directly related to the nature of multilateral financial institutions, the goals they profess, and the kind of development assistance they provide. The questions remain paramount: Are these banks political agents of the countries that control them? Do they operate 13

14

15

The North South Institute conducted a study on the RDB in the mid-1990s (see Culpeper, 1997). “Whereas originally the MDBs [multilateral development banks] were all established as banks in the sense of acting as intermediaries between the capital markets and developing country borrowers, they all – with the exception of the EBRD – eventually acquired a quasi–grant-making capacity by attracting funds from bilateral donors. This capacity enabled the MDBs to become multilateral aid agencies as well as banks, thus giving them something of a dual personality (a suggestion they all stoutly deny)” (Culpeper, 1997, p. 12). “In addition to a lack of support from their most powerful shareholder, many of the multilateral economic institutions born after World War II or developed during the Cold War have encountered other problems. In the past several years a number of studies conducted by these institutions and their supporters have sharply criticized their management practices and ability to define and accomplish objectives” (Upton, 2000, p. 2).

1.2 The Importance of the Puzzle

13

as independent institutions, more concerned with a balanced budget and strong financial support than with the impact of their policies on development? Or are they poverty-alleviating development institutions? Solving this puzzle and shedding light on some of the problems and critiques that these multilateral financial institutions face is important for the following reasons: (1) The countries that support these institutions are not willing to eliminate IFIs, even if they are flawed. As the 2000 IFIAC Report, commissioned by the US Congress, suggests that development assistance and the way IFIs operate are becoming central policy outlets for OECD (Organization for Economic Co-operation and Development) countries, given their extensive involvement in these institutions and the financial support they provide them. (2) Globalization has created new challenges for multilateral financial institutions: as capital and financial flows grow in volume, developing countries face new opportunities and a different financial terrain. Multilateral financial institutions must not only adapt to this new reality and reassess their goals and strategies, but also seize the opportunity to become as effective as possible where most needed. (3) RDBs are well known in the developing countries to which they cater; a project or loan that may seem insignificant by Western standards may be regarded as considerable in some of the poorest countries in the world. This point highlights the fact that the disbursement amounts provided by the RDBs should be weighed against the credit standards of the borrowing countries rather than those of the West.16 (4) Recent controversy over the effectiveness, necessity, and future of these institutions has spurred fervent debates among policy makers and academics alike. These controversies have generated numerous arguments suggesting reform or even elimination. Positions that do

16

For example, from 2001 to 2006 the AsDB made close to $985 million in loans to the Philippines. During that period, the Philippines’ external debt fluctuated between $50 billion and $65 billion. As a percentage of debt, the AsDB loans, on average during that period, made up more than 1 percent of external debt. However, more importantly, since the vast majority of RDB loans are in the form of project or program loans, their visibility lays in the actual product that is attached to them. For example, the IDB recently made a $75 million loan to Honduras to benefit 100,000 poor households in rural areas. While in absolute numbers this may seem like a small amount, its direct impact cannot be measured monetarily, since in addition to the prospect that this program will succeed, the beneficiaries of this loan are now familiar with the IDB and understand the possibility of its impact.

14

IFIs, Development, and Regional Development Banks

not advocate some change are in the minority. Although not as wellknown and researched as the World Bank or IMF, RDBs are an integral part of these debates. In fact, the IFIAC report advises that they should become the primary development agencies in their respective regions. (5) With the apparent volatility of the private financial sector, governmental institutions are more reliable in offering credit that extends support for welfare and development, and therefore IFIs occupy a unique role that needs to be more clearly and empirically evaluated. Examining the RDBs’ decision-making structures for their lending – their primary operational apparatus – sheds light on the puzzle, highlighting the economic and political forces that influence the institutions’ work and helps to determine whether they are compatible. Thus, this book focuses on the institutional structure – the way these banks were formed, how they function, and how they evolved – and its effects on the development strategies they advance (epitomized by the actual disbursement of loans). More specifically, I examine why these banks were created and how they developed over time via relations to the countries that formed them and the countries that became members over the years (both donor and borrowing). These relations are indicative of the principal-agent relations of the RDBs, their preferences, and their strategies. Through systematic analysis of the loans made by the banks since their inception, this research shows the lending strategies/patterns employed by these institutions. It is the combination of these two phases of the study that demonstrates the intricacies of the work of RDBs. This analysis, then, provides a theoretical basis for the more general questions concerning public financial institutions and their independence from political influence, the assumption that political influence is necessarily negative, the constraints (both political and economic) that such institutions face daily, and the way stakeholders or shareholders delegate powers to the institutions’ bureaucrats. In sum, this research includes two main features: (1) A comparison of the RDBs. (2) An analysis of the loans they each made over enumerate years. These two components contribute to answering the question of whether and how the RDBs’ development goals are reconciled with their banking features, and what are the various institutional characteristics that make each of them unique (in structure, in operation, and in effectiveness).

1.3 Research Design

1.3

15

research design

As stated above, most studies of multilateral financial institutions focus on one institution, usually either the World Bank or the IMF. However, I have chosen to focus here on all four RDBs for a number of reasons. First, these four institutions are comparable (to each other) in their scope and capacity.17 Second, a comparison serves to highlight differences among the institutions, allowing for more accurate conclusions. Thus, unlike a case study of the World Bank or the IMF, this analysis is based on a larger sample study that can yield more generalizable results. Third, the distinct differences between the banks – in their institutional structure and culture as well as in their mandates – can further advance the understanding of the role and limitations of multilateral financial institutions. Finally, this study helps both supporters and critics engage in a more educated discussion on the merits and shortcomings of IFIs. By analyzing the way member countries approve loans, the size of loans, their countries of destination, and the way members interact with the management and staff of the RDBs, I generate an empirical and theoretical understanding of RDBs and more generally, multilateral financial institutions. Hence, this study shows that RDBs are complex institutions with a multilayered rationale: They serve both as a political arena and as a forum for collaboration on development. They suffer from political dependence on powerful members, political agenda setting, “strategic lending”18 and bureaucratic stiffness. At the same time, in some countries that experienced growth and a decline in poverty indicators, RDBs have undoubtedly contributed to development over the years.19 Hence, I explore the bank versus development agency identity of multilateral financial institutions by comparing the four cases, and highlighting variations among them. This analysis is then used to determine whether RDBs engage in bank-like operations 17

18

19

The IDB lends to almost all of Latin America, the AsDB lends to almost all Asian developing countries, the AfDB to practically all African countries, and the EBRD lends to almost all of developing Eastern and Central Europe. Thus, save the Middle East, where no RDB exists, these banks cover essentially the entire developing world. It is important to emphasize, that the numerous development banks that are not part of this study embody different sizes and scopes and would not make for a revealing comparison. Based on the term coined by Thomas Schelling (The Strategy of Conflict, 1960; Strategy and Arms Control, 1961; and Arms and Influence, 1966), “strategic lending” in this research refers to the RDBs’ decision to engage in lending that serves the political and financial standing of the banks – namely, the guarantee of the repayment of loans and the servicing of political interests of influential donors – rather than lending that follows the banks’ stated goals of poverty alleviation. For more on strategic lending see Baldwin, 1965.

16

IFIs, Development, and Regional Development Banks

or advance a development-oriented strategy, and how their institutional design impacts their place on the bank/development agency spectrum. The conclusions drawn by this study present a multifaceted picture of these institutions. Since there is a fine line between the economic and political spheres, attaching meanings to policies can be difficult. It seems only logical that the very idea of institutions that are banks (with all the associations attached) as well as development agencies (with all the expectations attached) generates criticism. By examining the lending operations of the RDBs, this book closely scrutinizes the actual activities these banks engage in that put them in the line of fire. The research design is based on the premise of the triangular relationship between (1) the formal arrangement that serves as the RDBs’ Articles of Agreement, (2) the internal practice of the institutions that is a result of the formed habits, the conventions, and the backgrounds of their employees, and (3) the actual development projects carried out by the RDBs, which are both an outcome of and an impact on the formal and informal arrangements.20 Based on the World Bank model, three of the major RDBs – the IDB, the AsDB, and the AfDB – were created in the late 1950s and 1960s after developing countries expressed dissatisfaction with the WB. The regional institutions have smaller budgets than the WB, but due to their geographic proximity and smaller scope, were expected to be a superior mechanism for carrying out certain development projects in their regions. In addition, borrowing countries appear to be larger stakeholders in the RDBs with greater involvement in the decision-making process and governance of the banks. The comparison herein is twofold. First, between the IDB, AsDB, and AfDB, which were created roughly at the same time and under similar circumstances and mandates. Second, between those three banks and the EBRD, a “generational comparison” can be conducted, since the EBRD is analyzed in light of its foundation in the early 1990s at the end of the Cold War and was therefore initiated under very different circumstances and with a different mandate. The formation of EBRD thus serves to highlight the changes and conclusions (if any) that the international community incorporated into the establishment of a new development bank. In

20

The loans disbursed by the banks have an impact on the institutions as far as they provide the premise for the RDBs’ reputation, the criticism directed at them, and the credit rating they receive. Furthermore, interest and loan repayments account for a substantial portion of the RDBs’ budget.

1.3 Research Design

17

addition, this comparison emphasizes the effects of a certain time period’s international political and economic conditions on setting the tone for financial institutions’ creation and goal making.21 The initial variables are therefore as follows: IV Institution

DV Outcome

The outcome is the product of the institution. It is the actual operations it carries out and their manifestation. In this case it is the loans made by the RDBs – their central activity – and the impact that the loans may or may not have. The institution is comprised of the countries that created it (donors and borrowers), the people who work in it, the Articles of Agreement that set its modes of operation, and the way these various elements interact to make policy decisions. For example, whether the principal employees and policy makers of the institution are bankers (as in the EBRD) or economists and development experts (as in the other RDBs) has a considerable effect on policy outcomes. Additionally, the extent and method of delegation of powers from the principal (member countries) to the agent (management and staff) is a key determinant of the institutions’ profiles. More specifically, the reference to institution in this text essentially means structure, and outcome implies the output of this entity. In order to correctly evaluate the output and the outcome of the institution, I analyze the loans the banks make in light of financial, socioeconomic, and political indicators of the countries that receive the loans, so as to place the banks’ activities in context. To summarize, the institution is a complex variable that includes many elements. In this analysis I narrow it down to the following: (1) Members states and their policy preferences; (2) Management, staff, and their mandate; and (3) Decision-making procedure and process. Specifically, the capacity of the banks to support lending activities is analyzed in the context of the following elements: hierarchies; key 21

It is noteworthy at this point that the Middle East is the only developing region for which there is no development bank (Egypt is a member of the AfDB). Although there is an Islamic Development Bank that services some of these countries, it is unique in its role, mandate, and scope. Since it follows very different rules and is not based on the WB and RDBs model, it is not included in this study. However, the political interest in a separate development bank for the Middle East is demonstrated by the numerous blueprints and policy discussions held about the topic, especially in the first half of the 1990s.

18

IFIs, Development, and Regional Development Banks

actors; review process; personnel (politicians/economists/bankers); and resources.22 The outcome is comprised of loans (amount and destination). To assess the outcome or output of these institutions, I collected an original data set that includes yearly loans per country (unit: country-year) of the IDB, AsDB, AfDB, and EBRD from the year of their creation until 2009.23 While the loan amounts (total, yearly) constitute the dependent variable, the independent variables include financial, socioeconomic, and political indicators of the borrowing countries. They facilitate conclusions about the characteristics of the lending process – the banks’ priorities, whether their lending is strategic (to wealthier countries or countries that are politically important to major donor members), and the relative importance of their loans to the borrowers. In this study, strategic lending is defined as a loan policy that strengthens the parameters that assure institutional success rather than one that puts considerations of poverty alleviation at the forefront.24 If the banks must keep up their “bank” profiles, priority will be given to projects with better prospects for “succeeding” (i.e., for making loans that will be repaid). To justify their existence, the banks must continue to make loans and maintain their credit ratings (evaluated by private rating agencies such as Standard & Poor’s and Moody’s). This means that, under parameters of strictly strategic lending, banks would not be able to afford to lend to risky projects or to countries that do not provide adequate guarantees. Furthermore, “strategic lending” also has a political component: it suggests lending that favors countries that are close political allies of major shareholders. Hence, strategic lending can be either economic, 22

23

24

RDBs, like all IFIs, have been struggling with issues of transparency. As public institutions, interested parties, activists, and NGOs often demand greater transparency. Over the years, and especially since the mid-1990s, the RDBs have improved the accessibility to data on their loans, although much information is still difficult (if not impossible) to obtain (such as interest rates). Project evaluation remains an elusive task. While the World Bank attempts to provide such data, the parameters for measuring “success” and “failure” are vague, as many projects are long-term in nature. Data for the AfDB was retrieved from OECD/DAC database and from various annual reviews. “Defensive lending” is another form of strategic lending. This is where the institution loans to a government so that it can continue to make interest payments on outstanding loans. It is not possible to make the determination that the RDBs lend defensively as the bulk of their loans (although government guaranteed) finance specific projects or programs. However, we can speculate that repeated sovereign loans to the same country, even if designated for specific projects, helps the borrowing government finance repayments of past loans.

1.3 Research Design

19

political, or both. Being a central factor of the puzzle, strategic lending (in terms of outcome) is analyzed using quantitative and qualitative methods: the qualitative research conducted for this study consists of numerous interviews with bank officials (current and former), US Treasury officials, IFIAC Report writers, OECD analysts, and other researchers involved in the review of these institutions, while the quantitative research consists of statistical data analysis of loans and other social, economic, and political indicators. The considerations taken into account when approving loans, coupled with the analysis of their destinations, shed light on the dual identity of the multilateral financial institutions. The answer to the question of whether their procedures and output resemble a development agency or are closer in nature to those of a bank will help sort out RDBs’ distinctiveness from private institutions (or lack thereof), and their usefulness as tools for development. Therefore: (1a) If the lending policies of the RDBs are constrained by internal rules and regulations (either internal or donor-inspired) that resemble those of a bank, and the lending patterns indicate preferences that point to “sure success” (strategic) rather than need-based loans, then the institution would be placed closer to “bank” on the bank/ development agency scale.25 (1b) If the RDBs allocate a larger proportion of their loans to countries that are politically aligned with principal donors, lending may be called politically strategic, and the institutions would further itself from the development agency end of the scale (lending is, then, not driven by need). (2) If the banks’ activities show a preference for poorer countries and less lucrative sectors, along with an internal review that encourages such policies, and if members consider this preference a priority, then the RDBs’ work can be considered closer to that of development agency.26 25

26

It should be noted that the RDBs have a concessional lending window designed to serve the poorest countries. This is, however, a small part of their operations and its existence does not necessarily justify prioritizing strategic lending over poverty alleviation and development in the nonconcessional lending window. I elaborate on concessional loans in Chapters 4 and 5. A development agency in this study will refer to an organization whose main objective is to foster development in all sectors. For the purpose of this study, I consider development agency activities to be those that private institutions will not consider undertaking and bank activities to be those that private institutions have a financial incentive to carry out.

20

IFIs, Development, and Regional Development Banks

1.4

development assistance

Before discussing the creation of the RDBs, one needs to look into the reasons for the formation of development financial institutions in general – what was their originally stated raison d’être? In some cases, external international conditions, such as the post–WWII financial crisis and the Cold War, were the pretexts for the establishment of these institutions. But their persistence over time, and the flux of new regional financial arrangements that continue to sprout, suggest the following possible reasons for the existence of multilateral financial institutions: • Their existence is an indication that countries assume that development assistance is needed. • The mere agreement to create a multilateral development financial institution embodies the assumption that market forces are not satisfactory in regulating finance. There is a perceived notion that cooperation among nations is needed to advance development in developing countries and overcome market failure. • Member countries have political and/or economic interests that give them reason to take part in this cooperative effort. • Multilateral (development) financial institutions are Pareto-optimal (in their area of activities) compared to other development possibilities (private sector, bilateral, and so on). Their Articles of Agreement stress the importance of the uniqueness of the lending opportunities provided in the case of the IDB, AsDB, and AfDB (lenders of last resort), and the rule of “additionality”27 in the case of the EBRD. • Their existence is the result of the foreign policies of member states; they are designed to facilitate communication and coordination, and, at times, to disguise foreign policy by carrying it out under the auspices of a multilateral institution. The stated goal of international financial institutions is to enable coordination among donor countries in their efforts to assist development in developing countries. Their main assignment is the provision of aid to the most needy, where no other option for development exists.28 They all

27

28

This will be discussed in depth later. The general idea is that the projects and services that the EBRD offers have an additional element to what other banks might be willing to grant. This is based on the WB’s founding Articles of Agreement, which serve as the RDB model. The EBRD stresses the value of “additionality” which provides a basis for the approval of loans (based on interviews with EBRD officials, Feb. 2002). The AsDB states: “ADB

1.4 Development Assistance

21

stress their role in poverty reduction, development, and social progress (except for the EBRD, which is committed to the development of market economies and democratic principles). Simply setting these ideals as objectives separates multilateral financial institutions from private financial institutions. It establishes a political and humanitarian standard for the activities they undertake: rather than profit, they seek a visionary goal of social and economic progress that is absent from the vocabulary of regular banks. In addition, this essentially political endeavor attempts to appear free from the political interests of the member states that created the institutions. Since the RDBs were preceded by the World Bank and the IMF, an overview of relevant attributes of these global institutions is essential for the comprehension of the context and model that influenced the creation of the regional institutions. Until its collapse, the Bretton Woods system represented the latest international monetary regime. As part of the agreement, signed by forty-four nations in 1944, the International Bank for Reconstruction and Development (World Bank) and the International Monetary Fund were created to provide postwar international monetary and financial arrangements.29 The WB, designated as a development institution, was designed largely to supply the capital needed for postwar reconstruction

29

ventures where others – especially private capital – fear to tread. It provides assurance, security, and guarantees. ADB is a known and trusted development specialist. Private capital follows ADB, not vice versa” (ADB FAQs, www.adb.org). The IDB’s official motto is the following: “The Bank’s Charter states that its principal functions are to utilize its own capital, funds raised by it in financial markets, and other available resources, for financing the development of the borrowing member countries; to supplement private investment when private capital is not available on reasonable terms and conditions; and to provide technical assistance for the preparation, financing, and implementation of development plans and projects” (www.iadb.org). And the AfDB posits that: “The Bank’s mission, therefore, is to assist Regional Member Countries (RMCs) to break the vicious cycle of poverty in which they are entrapped. Working towards this goal, the Bank would endeavor to facilitate and mobilize the flow of external and domestic resources, public and private, promote investment, and provide technical assistance and policy advice to RMCs” (www.afdb.org). “The Bretton Woods agreements . . . were remarkable in a variety of ways. First, they represented an unprecedented experiment in international rule making and institution building – rules and institutions for post-war monetary and financial relations. Second, the Bretton Woods agreements were the decisive step in the reopening of the world economy . . . the world economy would abandon regional currency and trade groupings in favor of a liberal multilateral system. Third, Bretton Woods created an entirely new type of open system – something that the capitalist world had not seen before” (G. John Ikenberry in Bordo and Eichengreen, 1993, p. 155). For more on the history and making of Bretton Woods see Bordo and Eichengreen, 1993.

22

IFIs, Development, and Regional Development Banks

and long-term development projects mainly for Western Europe,30 while the IMF served as an immediate supplier of liquid assistance – the “lender of last resort.” Membership in the IMF was a prerequisite for membership in the WB, and donor countries’ votes were commensurate with their relative contributions.31 Due to its long-term development assistance agenda, the WB was also responsible for the promotion of private foreign investment by guaranteeing and participating in loans by private investors. This symbiotic relationship between the private and public sectors is relevant also to the RDBs. The long-term goals of development require augmenting private sector interest in developing countries. However, if the public institution (WB, RDBs) does not redirect its focus to different venues when private capital is available, it is not fulfilling its mission. Members subscribe to the WB by taking part in shares of capital stocks. At the onset (and still today), 20 percent of the subscribed capital is paidin and 80 percent is subject to call by the Bank “only when required to meet its own obligations on its borrowing or guarantees” (ibid., p. 368).32 Therefore, the daily operations (including the budget) of the WB utilize only 20 percent of the subscribed capital to which members commit. The Bank’s goal is to avoid having to summon funds from the callable capital pool, as this would jeopardize its relationship with investors and most likely negatively affect its credit rating. The Charter of the WB was drafted under the shadow of the perceived capital market failures of the period between the wars. To avoid repetition of these failures, the Bank created strict protective provisions to govern loans. In short, the borrower government must guarantee the loan and be in a position to repay it. In addition, the Bank would make loans only to projects that it was convinced would not be able to obtain loans from private sources under reasonable conditions.33 Although the division of labor between the IMF and the World Bank now seems blurred,34 John Maynard Keynes, one of the early architects of the Bretton Woods system, was quoted saying “I should like to see the Board of the Fund composed of 30 31

32

33

34

Dominguez, in Bordo and Eichengreen, 1993, p. 366. For example, from the onset, in the IMF each member has a quota that determines its financial contribution. Voting power consists of one vote for each $100,000 in quota. Voting shares and their significance at the RDBs are discussed in detail in Chapter 2. This is what is known as “callable” capital. This same model was used in the creation of the RDBs, and this structure and its effects will be discussed later. “The Bank must be satisfied, before making or guaranteeing any loan, that in the prevailing market condition the borrower would be unable to obtain the loan from private sources under reasonable conditions” (IBRD, 1954, p. 7). See IFIAC Report, 2000; Feinberg, 1988.

1.4 Development Assistance

23

cautious bankers, and the Board of the Bank of imaginative expansionists.”35 The purpose of this distinction was to separate the task of development, which is long-term and involves mostly project-based lending, from emergency assistance, which is short-term and often involves the need for immediate liquidity and large transfers to governments during moments of crisis. The WB and other Bretton Woods institutions would not exist if it were not for US support. Despite the war and the depression that preceded it, the US economy was clearly in better shape than that of any of the other countries participating in the creation of these institutions. Thus, American support was needed for any new system to emerge. And the United States, being in a better bargaining position vis-à-vis other countries, was able to exert its influence and hold the largest portion of the voting shares (at the inception of the WB, the United States had 35.07 percent of the votes, followed by the UK with 14.52 percent).36 This system is very different than the one-vote-per-country method of the United Nations and, more recently, the World Trade Organization (WTO). It formed the foundation for the creation of the RDBs, in which a corporate structure based on relative shareholder power was applied, albeit with the charter-based regulation that regional members (developing countries) hold the majority of those “shares” (membership at the RDBs consists of “donor,” or nonregional, members and “borrowing,” or regional, members).37 Although borrowers contribute to the banks, the donors obviously contribute the bulk of the RDB capital. The borrowing members, however, retain the contractual right to limit the voting power of the nonregional donors.38 The three multilateral institutions created by the Bretton Woods Agreement – IMF, IBRD (WB), and GATT39 – were able to survive the collapse of the Bretton Woods system “by evolving with the changing economic environment.”40 But if the WB “evolved” and fulfilled the 35 36 37

38

39

40

Moggridge, 1980, cited in Bordo and Eichengreen, 1993, p. 226. See Culpeper, 1997, pp. 23–27. With the exception of the AsDB, where three donor members are also counted as regional (Japan, Australia, and New Zealand). The membership rules and shareholders’ powers are further discussed in the next chapter, and are followed in Chapters 5 and 6 by empirical analysis and findings regarding shareholders’ roles, rights, and activities. For more on developing countries as World Bank shareholders, see Helleiner, 1996, pp. 293–323. The ITO was planned at Bretton Woods – and failed. However, it provided the footprint for the GATT that succeeded. Thus, I consider it as one of the three institutions created by the Bretton Woods Agreement, even though technically, it is not. Dominguez in Bordo and Eichengreen, 1993, p. 358.

24

IFIs, Development, and Regional Development Banks

mission envisioned by its creators, why were the RDBs formed approximately fifteen years later? Was the global institution not sufficient? Or was there a notion that regional institutions could do a better or different job? The existence of IFIs, and RDBs among them, has been marred with controversy.41 In addition to the possibility that member countries promote their selfish interests through these institutions, some argue that IFIs have become bureaucratically cumbersome, unable or unwilling to evolve or adjust to new financial situations while lacking insight and judgment as to what constitutes “development” and what qualifies as strategic lending.42 Furthermore, some scholars and observers of multilateral institutions posit that, once created, institutions can take on “a life of their own,” incorporating a newly fashioned “institutional” culture into their operational ideas.43 Therefore, for MDBs, this means that certain habits and rules become part of the common practice and norm, creating a well-oiled machine that operates systematically according to certain conventions that grow out of a newly evolved institutional culture. Frequently, this means that while formal rules and practices are being followed, management and staff (as well as shareholders) lose sight of the goals that motivated the initial creation of the institutions, leading to questionable outcomes.44 Table 1.1 suggests that more than half of the projects sponsored by the WB in the 1990s failed to produce satisfactory results (57 percent). These figures were obtained from the WB’s own data: dividing its project assessment mechanism into three categories (outcome, sustainability and institutional development), the WB traditionally rates itself (using internal “independent” evaluations) only on the basis of “outcome,” concluding that over 90 percent of its projects are satisfactory. This means that a loan that was granted and used (for example, a school was built in Indonesia) is categorized as “satisfactory” in the “outcome” column (whether the school continues to be used, regardless of how it is being used, etc.) The table above is comprised of statistics of the next stage of project assessment: the “sustainability” category documented by the WB itself.45

41

42 43 44 45

For more on aid policies and the record of aid agencies in developing countries see Easterly, 2001. Easterly, 2001; Vreeland, 2003, 2007; Stone, 2002; Thacker, 1999. Barnett and Finnemore (1999), and various other works by Finnemore. See Meyer and Rowan, 1977. Based on an interview with Dr. Adam Lerrick (February 19, 2001), one of the authors of the IFIAC Report.

1.5 Government’s Role

25

table 1.1 Performance of World Bank projects Failure Rate of Projects to Achieve Satisfactory Results (%) 1990–93

1994–97

1998–99

1990–99

Adjustment lending Investment lending Africa South Asia Latin America East Asia

55 60 75 66 51 38

45 59 74 56 50 36

37 56 68 60 37 48

47 59 73 61 48 39

Low income Lower middle income Upper middle income High income

73 48 45 27

69 50 36 30

66 46 31 28

70 49 39 28

Total

59

56

53

57

Source: IFIAC Report (2000).

Although this research does not measure RDBs’ success rate per se, by comparing the lending practices of the RDBs and analyzing how the various banks fare on their development profile, this study is a necessary step in determining “success,” because “success” must be analyzed in the context of what the institutions set out to do. Whether or not RDBs suffer from the same shortcomings as the WB is analyzed with the help of the following research questions: What considerations are taken into account when approving loans? What is the mechanism used to approve loans and what factors are the most important in the process? Are RDBs a product of political agendas used to benefit certain countries or policies? And, finally, are the loans made by the RDBs “strategic” (as defined earlier)? Therefore, although “success” or “failure” is difficult to measure, the analysis and configuration of the institutional mechanism responsible for supporting projects in developing countries is critical to the study of development assistance.

1.5

government’s role

What – if anything – should be the government’s role in administering aid? Disputes over the role of public institutions in social and economic affairs are controversial in international affairs as much as they are in domestic

26

IFIs, Development, and Regional Development Banks

ones. The notion of “development” and the need to assist less developed countries in the process of growth and poverty reduction has been a central pillar in the grand debate between socialism and capitalism. Admitting that interference by the state is crucial to help advance the economies of some countries contradicts the laissez-faire claim that market forces create the desired equilibrium on their own, without government assistance in the form of redistribution. Accordingly, there are different views among capitalists about the issue of development. There are those who maintain that the market left alone will operate best, and that governments that meddle in the economy tend to cause more harm than good,46 and there are those who encourage government intervention where the market needs “correction.”47 Some of these arguments rest on the notion that more developed economies in emerging markets will make for better trading partners and investment potential, boosting the global economy as a whole and contributing to local interests in developed nations. Classical economists such as John Stuart Mill discussed issues concerning government intervention in the market. This history is not only relevant, but also important in understanding today’s neoclassical economic perspectives. Critics of capitalism, then and now, claim that capitalist states’ interventions in the economic affairs of developing countries benefit their own selfish imperialist interests, and are not driven by benevolence or the desire to assist developing countries in joining the developed world. Lenin wrote in 1916 that “[t]he uneven and spasmodic character of the development of individual enterprises, of individual branches of industry and individual countries, is inevitable under the capitalist system.”48 The collapse of the Soviet Union symbolized the decline of revolutionary socialism, but it does not necessarily point to the superiority of the capitalist system, as we know it. However, as John Stuart Mill suggested, within the capitalist enterprise there is a certain spectrum; capitalism does not have to reflect the division of labor and property that was so criticized by communism.49 Moreover, he stresses that in a society that is “founded on private property and individual competition” human improvement and progress should be a chief concern, as it will benefit the community 46 47

48 49

For example, Milton Friedman, 1982. Even Adam Smith said as much in “The Wealth of Nations” when discussing market failure. Lenin, 1993 (reprinted in Frieden and Lake, 1995, p. 110). “But it is not by comparison with the present bad state of society that the claims of Communism can be estimated . . . ” (J.S. Mill, 1871, reprinted 1994, p. 17).

1.5 Government’s Role

27

as a whole (J.S. Mill, 1994, p. 17). Over the years this debate has evolved as different political systems have emerged. The institutions examined in this book are distinctly phenomena of the twentieth century. Since 1945, which began an era of unprecedented levels of national and global growth, the variations between developed and developing countries have become more striking than ever.50 It is also important to note that concerns about “Third World” poverty and growth are recent historical developments. Only after WWII, when poor countries became independent, and therefore “important,” did their problems come to seem more urgent to the developed world (Easterly, 2001, p. 30). For example, as Easterly (2001) demonstrates, the wealth gap between the fifty-eight countries with the highest per capita income and the fifty-eight countries with the lowest per capita income between 1950 and 2000 increases dramatically. These growing gaps are the basis for debates among scholars and practitioners over which strategies employed by multilateral financial institutions best serve more equitable growth and development. However, though multilateral financial institutions can be looked upon cynically, and although wealthy countries might have acted in self-interest (Cold War alliances, market expansion, etc.), the mere creation of these institutions acknowledges that market forces alone do not sufficiently create a desirable economic balance, and that eradicating poverty in developing regions requires governmental intervention. Since the inception of the WB in the mid-twentieth century, political and economic approaches to development strategies and economic growth have changed several times.51 These shifts were the outcome both of disappointing results and of the changing dominant theories adopted by policy makers and analysts. International financial institutions may also simply have needed time to develop from their infant stages. Moreover, shifts in the international environment and the graduation of developing countries into new stages of development required a continuous, rigorous examination of the dominant paradigms of the field. 50

51

“The richest fifth of the world’s population currently enjoys 82.7 percent of global product . . . while the world’s poorest fifth enjoys only 1.4 percent. And while the ratio between the income of the richest 20 percent was 30:1 in 1960, it grew to 32:1 in 1970, 45:1 in 1980, and 59:1 in 1989” (Freiden and Lake, 1995, p. 417). Also, see Easterly, 2001, chapter 3: “For the whole period 1960–1999, the poorest countries did significantly worse than the rich countries, with the poorest two-fifths barely mastering positive growth” (p. 60). See “Perspectives on Aid and Development,” 1997, Gwin and Nelson eds.

28

IFIs, Development, and Regional Development Banks

The prevailing view in the 1950s and 1960s was that growth, through the accumulation of capital, would be the key to poverty reduction.52 Accordingly, the WB (and the RDBs, once they were created) focused efforts on infrastructure, as the general view of the Import Substitution Industrialization (ISI) growth strategy was prevalent. This strategy was a result of the economic isolation period following the Great Depression in 1929. After WWII, many developing countries institutionalized this policy by instituting high tariffs and promoting government policies that encouraged the development of local industry.53 However, the oil crisis of the 1970s created a wave of outrage in developing countries, giving rise to theories of dependency that blamed the biased rich-countries-controlled multinational corporations and international banks for acting as central obstacles to development. Extrapolating from Marxism and nationalism, the dependency school argues that the periphery (or dependent countries) is unequal to the center (or dominant, rich countries), and that the latter dominates the market of commodities, capital, and labor, ensuring the persistence of inequality and dependency. This theory of an asymmetrical power relationship brings political and social factors into the otherwise economic-driven theories of development, and posits that without such factors, an honest analysis of poor countries cannot be undertaken.54 As a result, in the mid-1970s the World Bank and the RDBs began to devote additional resources to poverty alleviation projects; but the conflict over how best to promote sustainable growth persisted and prompted calls for a New International Economic Order (NIEO). However, in the meantime, analysts and development specialists in the United States and Europe became increasingly critical of the governments of developing countries, putting much of the blame for stagnation on their domestic economic policies.55 This was in stark contrast to the former, structuralist, position. 52

53 54

55

The first generation of postwar development economists based their theories on experience from the Great Depression and the USSR. They encouraged increased investment to spur growth (see Easterly, 2001, pp. 30–42 for more on “Financing the Gap” and the Harrod-Domar model). Frieden and Lake, 1995, p. 418. See also Summers and Thomas, 1993. For more on Dependency Theory see Meier, 1995, pp. 107–111; Cardoso and Faletto, 1979. The World Bank Report in 1981 “Accelerated Development in Sub-Saharan Africa” argued strongly that stagnation was the result of local governments’ policies rather than adverse international price trends or other external forces (discussion in Gwin and Nelson eds., 1997, p. 4).

1.5 Government’s Role

29

The debt crisis of the 1980s saw an augmentation of these new ideas: it was now believed that growth could be achieved only through open markets, fiscal discipline, and the promotion of exports (export-led growth) rather than import substitution policies. The inefficiencies that resulted from protectionist policies were now recognized by many developing countries. Although this new approach appeared to be successful in East Asia (recall the East Asian “tigers”),56 its results were negative in Latin America, where high inflation and populist governments dominated in the 1980s, as well as in sub-Saharan Africa where stagnation became unbearable. However, by the early 1990s, after the fall of the Soviet Union and with the economic success of Chile and the resumed growth of Brazil and Peru, the crucial importance of fiscal discipline resonated as a winning paradigm. There was now a more widely held consensus on issues such as market-driven prices, interest rates and exchange rates, and the benefits of trade liberalization, issues about which there had previously been profound disagreement in academic and policy-making circles. This was also the theoretical current prevalent at the time that influenced the creation of the EBRD. The result was an institution with a different mandate and different goals than the more senior RDBs, which have since been pressed to shift paradigms, leading their policies closer to those guiding the EBRD. Therefore, as we shall see in the coming chapters, the question posed here is not only whether institutional change is needed (or not), but also whether new approaches, such as those exemplified in the charter of the EBRD, will in fact help achieve the original goals set out by the older RDBs. Answering these questions becomes complicated, because instilling the strict market regulations that are part of the conservative fiscal approach requires a period of adjustment. In developing countries, where the economic situation is dire in the first place, these new policies create an even more impoverished society. Social sectors such as health and education have been severely hurt by the austerity measures of the market-oriented fiscal policies. By the mid-1990s these concerns finally surfaced with an explosion of criticism against the structural adjustment programs of the international financial institutions.57 With soaring rates of poverty in some regions and increasingly higher debt, along with the 1997 Asian Financial Crisis, which disenchanted the proponents of the “Asian success story,” international financial institutions were now rethinking their roles 56 57

See Krugman (1994), Radelet and Sachs (1997) for more on the “Asian Miracle.” See Easterly (2001, chapter 6) on adjustment loans and their shortcomings.

30

IFIs, Development, and Regional Development Banks

and contemplating new ways to incorporate a quest for growth and poverty alleviation (or at least reduction) into their aid and lending strategies. Given the new reality of today’s global economy, reassessment is necessary; the combination of increasing interdependency where finance is concerned, along with a growing number of markets, investment opportunities, and options available to at least some developing countries, has changed the landscape for multilateral financial institutions. Multilateral institutions are now not the only players in the international arena dedicated to working in developing countries. As some developing countries become more developed, private investors are more easily attracted, seeing not only the opportunity of a new market, but also, potentially, high-yield investments. This constitutes a new challenge for multilateral development banks (MDBs): they now need to differentiate between projects that could be financed by private actors who are new to the scene and projects that are worthy of public financing as part of longterm development goals. Many nongovernmental organizations, academics, and policy makers are promoting the cancellation of debt for some of the poorest countries – the Highly Indebted Poor Countries (HIPCs).58 This initiative (1996) is lacking according to some critics (Sachs et al., 1999). But it is a beginning for a new approach to tackle the escalation of the debt problem in some of the world’s poorest nations. As Sachs et al. argue, international organizations should play a major role in administrating the debt relief programs, making sure the increased funds are directed toward basic needs to improve human welfare. In HIPCs, they posit, “ . . . most basic human needs are jeopardized by the continuation of contractual debt servicing” (Sachs et al., 1999, p. 4). In line with the IFIAC Report, their analysis emphasizes an increase in grants and flows to the poorest countries along with a sensible social program. One of the problems they raise is that although the transfer of resources to nongovernmental agencies may increase inflows on the accounting sheets, it does not necessarily help governments avoid default or increase social spending. This is because the money is spent on specific projects rather than on relief of the debt that can free resources for health and education. Most important, they

58

The HIPCs are a group of countries designated by the IMF and World Bank to be poor or highly indebted. There are 41 countries on the original HIPC list; 34 of the 41 countries including 77 percent of the HIPC population (712 million) are in sub-Saharan Africa (Sachs et al., 1999, p. 3).

1.6 Cooperation

31

conclude, debt relief is necessary but not sufficient to help these countries meet the goals of improving health and education in the coming years. Thus, aside from writing off the outstanding debt of the neediest countries in the world, international agencies and institutions could engage in longterm planning that combines poverty alleviation through the improvement of health and education standards alongside the promotion of economic growth.59 It is important to emphasize that debt relief is a formula that has been studied and practiced by the World Bank and the G7 (beginning with the June 1987 meeting) for over two decades. However, treating the phenomenon of debt as an external factor unjustly represents reality. The cause of the dire debt problem in poor countries is not entirely the lending policies of the donors. Moreover, by canceling the debt, developing countries will not miraculously recover from their endemic problems, shortcomings, and misfortunes.60 While the promotion of debt relief advanced by such an unlikely coalition as Sachs, Bono from U2, the Dalai Lama, and the Pope (all part of the debt forgiveness campaign “Jubilee 2000”) may generate skepticism due to its celebrity/polemic nature, it highlights a solution to which IFIs could contribute: RDBs still have a major role to play in the future of development assistance for poor countries. Moreover, continued cooperation between developing and developed countries through joint financial institutions remains indispensable, since it provides the only formal arrangement that allows for an active exchange of ideas and collaboration between donors and borrowers.

1.6

cooperation

Over the years, states have developed various regimes that seek to pool resources in assisting poor countries. These different institutions are indicative of the numerous methods employed by wealthy nations to assist developing ones. In the most general sense, they can be divided into two categories: bilateral and multilateral. Bilateral aid, both in the form of grants or transfers and loans, is part of a country’s foreign policy agenda. The political nature of this aid is evident. Since numerous countries can vie 59

60

Note that sectors that grow and have positive net flows are not necessarily public and therefore do not contribute to a country’s balance sheet or to the improvement of social programs. This is part of the critique of the 1996 HIPC initiative, which measured poverty in accordance with exports, when debts are owed by governments and not exporters (Sachs et al., 1999, p. 6). See Easterly 2001, chapter 7.

32

IFIs, Development, and Regional Development Banks

to assist the same developing country, there have been efforts to try to coordinate these various activities with the hope of improving the overall results. The Development Assistance Committee (DAC), of the Organization for Economic Cooperation and Development (OECD) is the central mechanism, created by developed countries in 1961,61 to coordinate their development strategies vis-à-vis developing countries.62 One of the goals of the DAC is to make sure that even bilateral assistance is coordinated, that there is minimal duplication and overlap, that countries are aware of what other countries are doing, and that development activities are being examined and reviewed.63 In addition, the OECD/ DAC monitors and sets the threshold for member-state contributions (percentage of GDP) for development assistance. Although it has no legal enforcement mechanism, making these contributions public information can put noncompliant countries in the spotlight, pressuring them to amend their development assistance policies.64 Multilateral financial institutions, such as RDBs, can be placed further along the “cooperation axis”: their job includes more than transparency of policy, provision of information, and coordination. They are actively involved in making loans and setting policy for financial assistance to developing nations. Their contributions set standards for investment in developing countries, facilitate financial flows by providing guarantees where risk would otherwise pose a problem, and help attract other, private sources of investment. As mentioned earlier, the original purpose of multilateral financial institutions was to overcome market failure. Especially in the aftermath of WWII, it was understood that the market should be assisted in helping less developed and war-affected countries reach an economic equilibrium that would enable them to participate as partners in the international market. These institutions strive to function

61

62 63

64

The OECD took over from the OEEC (Organization for European Economic Cooperation), which was formed to administer American and Canadian Aid under the Marshall Plan. This role has been gradually added to the scope of activities of the OECD. The DAC meets regularly to review the development activities of member countries. In each meeting, one member is reviewed by two other members. These reviews consist of confidential information regarding financial transfers and assessment of the projects by the reviewing parties (who have traveled to the locations where assistance was given and examined its uses). These meetings are not accessible to the public and are attended by the countries’ ambassadors to the OECD, the highest officials. For more on the DAC’s Peer Reviews see Ben-Artzi, 2016. The set OECD goal for commitments to aid is 0.7% of GDP (the US has never met this goal).

1.6 Cooperation

33

as an independent entity and avoid becoming a political arena for the interaction of member states. However, since member states are shareholders and ultimate decision-makers, the relative power of these states is significant and their interaction is prone to political considerations dictated by the governments at home. The policies of the banks, even if criticized on merit, are still a product of negotiation and cooperation among the members. Multilateral financial institutions make a strong case for cooperation – institutions, not only set up to coordinate and review, but also to provide a forum for joint decision and policymaking. Although it is rational for states to resist relinquishing power or to put themselves in a situation where they do not exert absolute control in foreign policy decisions, RDBs demonstrate that under certain circumstances, states prefer to become members of such institutions when they perceive that their gains will most likely surpass their losses.65 The history of the RDBs’ inception demonstrates that timing – both political and economic – is essential for the creation of multilateral financial institutions. The combination of an appeal voiced by developing countries for a regional financial institution, coupled with a political environment conducive to hegemonic interest in such a project (Cold War), were the foundations for the RDBs’ existence. Their creation, whether superimposed by a hegemon (as was the case with the AsDB and the EBRD), or a consequence of the lobbying of developing countries (IDB and AfDB), was, in the end, a product of the joint interests of all parties involved. Whether a hegemon exists or not, multilateral financial institutions are an example of an arena in which countries cooperate, and they continue to do so as global political and economic circumstances change. Even at the extreme end, where findings indicate that RDBs lend only where financial assistance from the private sector is available, or that their effectiveness is questionable, they are nevertheless an example of cooperation. Still, this does not preclude “power politics.” In fact, as Thacker claims, multilateral organizations are “in a sense a difficult or crucial case for political theories of international finance. It is easy to see how bilateral capital flows could be subjected to the push and pull of international and domestic politics, but on the executive board of the IMF any single country’s power is diluted by the presence of other principals within the decision-making body” (Thacker, 1999, p. 71). 65

And gains, in this case, may not always be financial: they include reputation, spillover effect, bargaining tools, influence, and so on.

34

IFIs, Development, and Regional Development Banks

Similar to the IMF, RDBs are a tough case for proponents of statist theories. Much of the “push and pull” politicking is done behind the scenes, in informal ways. The observable outcomes are mostly a result of official unified board decisions, endorsed by the majority (and usually all) of the member countries. Throughout the years, a certain institutional discourse has developed which enables the various members to sense conflict before it arises, and informal institutional mechanisms solve these problems or eliminate them before they reach the formal stages of negotiation. For example, knowledge of the US stance on most issues enables the banks’ staff to strategize their proposals to fit US requirements for support. If there is an anticipated conflict, discussions will take place before proposals are brought to the board for a vote. At the IDB for instance, a proposal that is not already (unofficially) “preapproved” by the United States will most likely not reach the voting phase.66 Criticisms of MDBs, suggesting that they are extensions of American foreign policy preferences, are abundant.67 Even within the United States, many refer to MDBs’ activities in the context of American foreign policy. But is this impression based on reality? How many particular American (or other hegemon/s) interests are actually carried out as part of the official RDB policy? By an examination of RDBs’ history and interviews with former and current officials and US Treasury Department representatives, in addition to analysis of actual shares and votes held by each country, this study demonstrates that, although hegemonic involvement is crucial to institutional creation and sustainability, the theoretical framework most useful for the analysis of RDBs daily activities is within the nexus of cooperation. 66

67

Section 4 (b) of the Articles of Agreement of the IDB state that no changes in voting shares can reduce the 30 percent voting share of the largest member – the US. Together with the stipulations that all decisions must be approved by an absolute majority of over threequarters of the votes, the US holds de facto veto power (“No increase in the subscription of any member to the ordinary capital stock shall become effective, and any right to subscribe thereto is hereby waived, which would have the effect of reducing the voting power (i) of the regional developing members below 50.005 per cent of the total voting power of the member countries; (ii) of the member having the largest number of shares below 30 per cent of such total voting power; or (iii) of Canada below 4 per cent of such total voting power.”) Copelovitch (2010) demonstrates that major shareholders are central to the World Bank; Thacker (1999) shows how the US influences the IMF; Stone (2002) argues that the IMF’s deliberations are subject to ratification by major donor countries; and Weaver (2008) examines the problematic governance structure of the World Bank that results in organizational hypocrisy.

1.7 Globalization: Adapting to a New Reality

1.7

35

globalization: adapting to a new reality

Any analysis of IFIs in the post–Cold War global economy must be conducted along with serious discussion of the effects of globalization. The end of the Cold War brought a change in the world order. Although globalization is not a new phenomenon, even in the twentieth century, its current form offers different challenges.68 Deregulation, new technologies, global financial flows, and the opening of markets in non-OECD countries characterize the contemporary form of globalization, which began in the 1980s. In addition, the end of the bipolar Cold War era increased public concern over such issues as the environment, cultural heritage, and migration. Finally, the globalization of corporate activities has been a major contributor to multilateralism since the 1990s. Thus, globalization is affecting all aspects of economic life, whether local or international.69 The RDBs that started operating during the Cold War and, to a certain extent, as a result of it, have continued to operate after the Cold War ended. In fact, the EBRD was established during the dwindling days of the Soviet Union. It seems that such a big structural change in world politics and world finance should have a paramount effect on these institutions. And indeed, I argue that this new reality presented new opportunities and new constraints. Multilateral financial institutions, although not obsolete, have been struggling to adapt to the post–Cold War world and the increasingly global economy in trying to achieve their goals. Since globalization has altered the economic opportunities available to developing countries, the value and measurement of the RDBs’ operations may have changed even if they did not adjust their policies. More than ever, the main goal of the RDBs, combating poverty and promoting development in developing countries, is timely. However, the situation of developing countries has changed. Some have become a bigger

68

69

Previous evidence of globalization includes the Gold Standard period preceding WWI (1870–1914), and the Bretton Woods agreement in the 1950s and 1960s (for more on this point see Oman, 1996). Globalization is defined as: “the growth, or more precisely the accelerated growth, of economic activity across national and regional political boundaries. It finds expression in the increased movement of tangible and intangible goods and services, including ownership rights, via trade and investment, and often of people, via migration. It can be and often is, facilitated by a lowering of government impediments to that movement, and/or by technological progress, notably in transportation and communications. The actions of individual economic actors, firms, banks, people, drive it, usually in the pursuit of profit, often spurred by the pressures of competition. Globalization is thus a centrifugal process, a process of economic outreach, and a microeconomic phenomenon” (Oman, 1996, p. 6).

36

IFIs, Development, and Regional Development Banks

part of the global economy, attracting investment and private finance (e.g., Chile, Brazil, and China). At the same time, others, left behind, are infected by growing rates of poverty, malnutrition, illiteracy, and AIDS (e.g., Haiti and Zimbabwe). The RDBs are at a crossroads: should they continue to sponsor projects that now have higher prospects of being financed by other means, or should they shift their focus to the most socially and economically needy projects, where results might not materialize immediately, and where no other institution is likely to invest?70 Hence, reiterating the role of globalization, it is important to keep in mind that certain projects that RDBs funded in the early years were deemed crucial for development and would not have otherwise been funded. Today, when they support similar projects, their policy may put them closer to a “bank” identity on the bank/ development agency scale simply because the realm of possibilities for aiding developing countries has been expanded. In light of these developments in the international economic and political environment, RDBs’ policies warrant a reexamination. In the first three decades of the RDBs’ existence, some of these questions might have been moot: for certain aspects of financing, they were the only game in town. Thus, questions on the nature of loans were overlooked or not a central issue to policy makers during the Cold War years. However, since the beginning of the 1990s, questioning the necessity and work of multilateral development banks has become a popular sport.71 When analyzing RDBs since their inception in the 1960s, it is important to set apart the effects of the last two decades. Recent changes serve to highlight some of the considerations that took place during the Cold War years that may not have been clear at the time. The creation of the EBRD is a perfect example of the changing times and the desire to find a new formula for development banking (the EBRD has a clear political-economic agenda to create market economies). Recent criticism of multilateral financial institutions has highlighted the problems they now face. Some claim that development banks are competing with the private sector, rendering their existence questionable. After all, the World Bank and its regional counterparts all vow to make 70

71

And these considerations are set against the constraints imposed by shareholders and the critique advocated by NGOs, politicians, and so on. Private sector actors now more visible in developing countries include portfolio and equity investors, multinational corporations, and private banks that provide microfinance loans, to name a few. Furthermore, some of the developing countries that have fared well financially, such as China and Brazil, are now taking part in developing efforts in the global south.

1.7 Globalization: Adapting to a New Reality

37

loans where no other possibility exists, but if their “bank” identity leads them to seek profit, then they may be stepping over this boundary. The puzzle concerning the nature and activities of RDBs and how they impact development cannot be examined without setting it in the context of globalization. It is clear that the end of the Cold War and an expediently growing international capital and finance market mark a turning point for multilateral financial institutions.72 What is the effect of this new global environment on RDBs? Have they made the necessary changes to adapt to it? See Figure 1.1.

Outcome

Globalization

Institution

figure 1.1

Globalization affects the institutions and the outcome, or the effectiveness of the institutions’ lending activities. The institution is affected because the relations among nations are evolving, the shareholders change their preferences, and the possibilities and needs of development change. The effectiveness of the institutions’ lending activities is altered because new players are introduced into the financial, economic, and political realms guiding developing countries’ development opportunities (e.g., private investors, commercial banks, and nongovernmental organizations). Some of the critiques and efforts to rethink the role, resources, and structure of multilateral financial institutions have been directed at the inability of these institutions to adapt to this new international environment. Data shows that from 1993–1999 approximately 70 percent of World Bank loans were given to eleven developing countries: China, 12 percent; Argentina, 10 percent; Russia, 9 percent; Mexico, 7 percent;

72

“Changing global economic and political conditions – including, most prominently, the elimination of the Cold War rationale for aid and the ‘globalization’ of the world economy have raised major questions about the need and purpose of future aid” (Eccles and Gwin, 1999, p. 12).

38

IFIs, Development, and Regional Development Banks

Indonesia, 7 percent; Brazil, 7 percent; Korea, 6 percent; India, 4 percent; Thailand, 3 percent; Turkey, 3 percent; Philippines, 2 percent. The recipients of the remaining 30 percent are 145 other developing members – mostly with lower growth rates and lower levels of development than the aforementioned 11 countries.73 Even though most of the countries receiving the bulk of the loans are some of the most populous in their regions, disproportion is still evident. To collect further evidence, this study analyzes the distribution of RDBs’ resources; whether they are significant to the countries that receive them, whether they provide assistance where no other option exists, or conversely, whether private sector flows are substantially larger in volume. Moreover, to present a more accurate picture of the activities of the banks, the data is assessed relative to GDP per capita and the population of the countries that are the destination of loans. This type of analysis provides more information than the one done by the IFIAC on the WB, since my study examines the loans disbursed by the RDBs in the context of numerous indicators pertaining to the destination country, not just its per capita income. Moreover, since this research covers the complete span of years of the RDBs’ existence, it furthers the analysis to include change over time. As mentioned earlier, the RDBs are designed in a way that regional (borrowing) members are shareholders: they have a certain amount of votes, depending on their relative contribution and provisions in the Articles of Agreement.74 Both the regional and nonregional members comprise the Board of Directors of the bank. These shareholders are responsible for all major decision-making, including approval of loans. This type of membership, and the role that states play in the institutions, further highlights the importance of globalization in the analysis and the work of the RDBs: if globalization affects states, then it necessarily affects the institutions of which these states are members. In sum, the process of globalization therefore directly affects RDBs: their possibilities and

73

74

Based on World Bank Data, cited in the IFIAC Report. In addition, the report shows that in these same 11 countries 98.6 percent of the resources (flows) come from the private sector, suggesting that the large percent of the World Bank’s resources directed at these countries is relatively small compared to the total flows into these countries. As we shall see, unlike the WB’s membership structure that is also based on the corporate shareholder model, most of the RDBs set provisions in their Articles of Agreement that allocate developing countries more votes (shares) than their actual contribution, while setting a limit on the number of shares available to nonregional members.

1.8 Conclusion

39

constraints are continuously altered by the global economy, creating an ever-growing challenge for these institutions.

1.8

conclusion

In this chapter, I discussed in detail the central question guiding this study: under which conditions do RDBs operate in a manner more like private banks than like development agencies? Solving this puzzle becomes particularly important in view of the ongoing debate on the merits of international financial institutions and the role they play in a changing political and economic global environment. Most importantly, this chapter explained how development assistance came to be a concern for developed countries, and the evolution of theories of development that guide the policies of institutions that administer foreign aid. Overall, the importance of RDBs in this scheme was established and the primary pillars of this book were laid out in order to further the analysis. Placing the RDBs in the context of international financial institutions (IFIs), this chapter provided an overview of the more general questions concerning development assistance, the role of multilateral financial institutions, and the factors that lead to their creation. The rich analysis of the World Bank and the IMF, offers a setting both for comparison with the RDBs (how are RDBs different from the global institutions?) and for a better-informed discussion of the role of IFIs in development, including the political motivations of their member states (what state interests lead to the creation of IFIs?). Furthermore, this chapter identified the central theoretical perspectives relevant to the RDBs and their functioning. By doing so, it provided the foundation for the rest of the book; namely, addressing the question about the RDBs’ ability to truly function as development institutions. This included a discussion of cooperation and globalization and how they provide both useful analytical tools and a constructive context for the study of RDBs. Finally, by placing the puzzle in the framework of globalization – how it affects the institutions and how it changed the environment in which they work – a dynamic dimension was added to this analysis, suggesting the changes that occur over time. The following chapters begin to answer many of these questions. The next chapter brings into the picture the theoretical underpinnings of this research and shows how the analysis of the RDBs relates to established schools of thought. It then presents the book’s hypotheses and data in detail. Chapter 3 provides an overview of the RDBs history, evolution,

40

IFIs, Development, and Regional Development Banks

and structure. Chapter 4 begins the quantitative analysis of the RDBs and the independent variables examined in this study. It provides a timeline account of the loans made by the RDBs, the impact of several variables on each of the regions, and an analysis of select countries and the amounts of loans they received over the years. Chapter 5 furthers the quantitative analysis and presents a crosssectional time-series analysis of the loans made by the various RDBs. In this chapter, I examine the interactions between the variables and determine the significance (or lack thereof) of the various economic, social, and political indicators. In addition, Chapters 4 and 5 examine whether there has been a shift in the resources available to developing countries and whether the RDBs are adapting to it.75 They also highlight the differences between the institutions, explaining possible causes for variation and the way institutional structure affects institutional results. Finally, Chapter 6 complements the previous quantitative chapters with a more in-depth look into the operations of the RDBs. Based on interviews with officials at the RDBs, the US Treasury, the OECD Development Center, and independent specialists, this final chapter analyzes the complex influences of politics, institutional design and culture, and human nature on the policies of the RDBs.

75

This is done by comparing FDI and RDB loans to developing countries over the years, and by looking at the socioeconomic indicators of developing countries to determine the division of loans between the most needy and the wealthiest.

2 Multilateral Governance: Theoretical and Empirical Underpinnings

There are limits to “imbalance” in economic development, to “lack of integration” in research and development, to “fragmentation” in policy making which would be dangerous to pass. And it is clearly impossible to specify in advance the optimal doses of these various policies under different circumstances. The art of promoting economic development, research and development, and constructive policy making in general consists, then, in acquiring a feeling for those doses. This art . . .. will be mastered far better once the false ideals of “balance,” “coordination,” and “comprehensive overview” have lost our total and unquestioning intellectual allegiance. – (Albert O. Hirschman and C. E. Lindbolm, Economic Development, Research and Development, Policy Making: Some Converging Views, 1962. Reprinted in Albert O. Hirschman, Rival Views of Market Society, 1986, p. 10)

How is policy made at multilateral institutions? More specifically, how does the diverse membership of development banks’ governance bodies conduct its decision-making? RDBs display a complexity that is fundamental to understanding not only the way international institutions work, but also why they sometimes do not work so well. As political organizations with governments as members, as institutions with a corporate-like structure and economic goals, and as agents located in different regions with different members and therefore susceptible to different cultural influences, RDBs present a compelling object of research. Moreover, the theoretical findings generated here carry significant contributions to the study of cooperation and international institutions, given the comparative component and institutions’ endurance over time. 41

42

Multilateral Governance: Theoretical Underpinnings

Students of international relations have extensively researched the phenomenon of cooperation. Whether it is possible to overcome the noncooperative outcome of the Prisoner’s Dilemma Game, for example, has been examined in many research projects.1 Researchers have investigated the best ways to achieve cooperation, overcome the collective action problem, and sustain cooperative endeavors. Scholars have also developed numerous hypotheses seeking to explain the existence of cooperation. Some of the secondary issues studied have been: Do countries cooperate as equal partners? Under what circumstances are they prone to cooperate? And, what sorts of resources are needed to achieve cooperation? International institutions (such as the WB and IMF) represent a strong case of international cooperation that has been examined extensively. They are instances of institutionalized cooperation on a multilateral level, where a country’s defection or nonmembership can be costly. This study transposes the various long-running arguments as to how institutions become established and how their continued existence becomes assured to the arena of the regional development banks. Theories such as Hegemonic Stability Theory2 and Principal-Agent Theory3, usually used to explain the origins and maintenance of the WB and the IMF, are employed here similarly with the RDBs. While Hegemonic Stability Theory posits that international institutions are maintained through the direct and active involvement of a hegemon, Principal-Agent theory suggests that under various circumstances principals delegate authority to agents. If one considers that an international institution with its entire bureaucratic infrastructure is the agent, then the principals, or member states, get involved or exert control to a different extent depending on their preferences and the payoff structures of different policy decisions. Theories on organizational culture also offer relevant insights into the evolution of the institution. With time, such theories assert, institutions acquire certain habits, fixed reputations, and external and internal relationships that dictate outputs. Corporate governance literature, too, offers perspectives useful to this study, since the management 1

2

3

See, for example, Keohane, 1984; Axelrod, 1984; Oye, 1986; Stein, 1990; and Martin, 1992. Gilpin, 1981; Grieco, 1990; Keohane, 1984; Gowa, 1984; Keohane and Martin, 1995; Fearon, 1998; and Snidal, 1985. Pollack, 1997; Epstein and O’Halloran, 1999; Hawkins et al., 2006; and Nielson and Tierney, 2003.

2.1 Who Rules?

43

structures of the RDBs resemble that of corporations.4 Finally, the abundant literature on foreign aid can also be drawn upon in investigating RDBs.5 All of these theoretical and analytical approaches offer compelling ways to study IFIs. At times we might look at these theories as presenting alternative explanations, while at other times they complement each other. This chapter introduces these theoretical mechanisms and ties them to the central questions regarding the operations of RDBs. Hypotheses and data are then discussed in detail, drawing on the theoretical frameworks mentioned above, and thereby setting the stage for the empirical analysis and findings to follow.

2.1

who rules?

Theories abound regarding international cooperation, as many seek to explain how cooperation is achieved, how it endures, and what it signifies. The following section elucidates some of the main views that are relevant to the institutions in question. Although these frameworks offer compelling alternative explanations for the analysis and findings of this study, I maintain that no particular theory “wins”; rather, the relative primacy of any theory depends on specific circumstances and the perspective used to examine the RDBs. 2.1.1

Hegemonic Stability Theory

Derived from Realist theories of international relations, Hegemonic Stability Theory (HST) posits that “only a hegemon has both the interests and the resources necessary to provide the collective good of world economic stability” (Gowa, 1984, p. 680). HST, however, can hardly be characterized as a uniform analytical approach. Theorists such as Krasner, Gilpin, and Kindleberger hold different views as to the systemic primacy of hegemons in international relations. They also hold different views as to the level of involvement needed from a hegemon to maintain stability. Though, as Gowa (1984) observes, “Despite their differences, then, Kindleberger, Krasner, and Gilpin concur that world economic stability is likely to deteriorate in the absence of a hegemonic power” (ibid.). 4 5

Beim and Calomiris, 2001; Aoki and Kim, 1995. Bueno de Mesquita and Smith, 2007; Alesina and Dolla, 2000; Dunning, 2004.

44

Multilateral Governance: Theoretical Underpinnings

In one view, HST claims that hegemons maintain economic stability (e.g., Kindleberger). Other scholars point to a relation between hegemons and the liberal economic order (Krasner and Gilpin).6 The subject of multinational financial institutions is a case of the latter. Any explanation of the foundations, evolution and activities of the RDBs would be incomplete without addressing whether there has been a hegemon involved, whether it has been needed, and what has been its function.7 After all, a hegemon can make considerable impact on the way the RDBs operate and where they are located on the bank/development agency continuum. Since we are concerned with the outputs of the development banks as well as with how the RDBs make the policy decisions that determine these outputs, HST may have significant relevance to this study. For example, Table 2.1 presents the percentage of votes of the biggest shareholders of the RDBs. It is clear that the United States plays a substantial role in most of these institutions. But despite the variation in voting shares – some countries holding more shares than others – no country in any of the RDBs holds a majority. Moreover, it is important to point out that since the United States holds a prominent place in the world economy, more voting power does not necessarily represent hegemony, but rather the manifestation of relative economic strength. In the chart below, the AfDB stands out as the only RDB whose borrowing members (regional countries) represent the majority of shareholders among the top five voting powers.8 However, as we shall see later, the AfDB suffers from the most severe structural and functional problems, almost to the point of its dissolution. Not having an economically powerful shareholder (a Western hegemon) may be a contributing factor. In this case, HST holds explanatory power to the extent that it indicates that with the absence of a hegemon an institution is less likely to operate effectively. Another observation that stands out in the chart is the makeup of the largest shareholders of the EBRD: they are all developed countries. This point shall be further explored as well. Besides the fact that these are obviously the countries that have the capital necessary for the bank to function, it also points to a difference in philosophy – a parting from the 6 7

8

Gilpin, 1987. “For example, whereas Kindleberger tends to view the hegemon as motivated by cosmopolitan economic goals, I believe that the United States has been motivated more by enlightened self-interest and security objectives” (Gilpin, 1987, p. 88). Borrowing member countries of both the AfDB and the IDB hold a majority of the shares. However, the US is the single largest shareholder at the IDB, its shares – and veto power – having been established in the bank’s Articles of Agreement.

2.1 Who Rules?

45

table 2.1 The voting power of largest shareholders in the MDBs (2002) Total Vote (%)

Country Inter-American Development Bank United States Argentina Brazil Mexico Venezuela Asian Development Bank

United States Japan China India Australia

31.11 10.95 10.95 7.04 5.87

13.20 13.20 5.66 5.57 5.12

Country African Development Bank Nigeria Egypt United States Cote d’Ivoire Japan European Bank for Reconstruction and Development United States Japan Germany United Kingdom France

Country

Percent of Total Vote

World Bank United States Japan Germany France United Kingdom

17.03 6.04 4.67 4.47 4.47

Total Vote (%)

10.00 5.77 5.61 5.02 4.63

10.91 9.32 9.32 9.32 9.32

Source: 2000, 2001, 2002 Annual Reports of the RDBs and the World Bank.

structure of the three other RDBs, which were created in the 1960s, while the EBRD was created in the early 1990s. Therefore, although it is clear that the RDBs were all created with the necessary support of wealthy donors (hegemons), it is not entirely clear how those donors continue to be involved in the banks’ operations as emerging markets shift alongside vast global political changes. Explanations of hegemons can be too simplistic: a malady often ascribed to Hegemonic Stability Theory. As Keohane (1984) observes, both realists and institutionalists found ways to claim the victory of their predictions during the first two decades following WWII (p. 8). For institutionalists, economic cooperation was a result of shared interests

46

Multilateral Governance: Theoretical Underpinnings

and interdependence. Realists argue that early postwar economic regimes rested on American political hegemony. Since this debate cannot be resolved in the method of Solomon’s trial, perhaps the correct way to address it is not by picking one approach over the other. Arguably, under certain circumstances leadership and power are needed to “make rules” or create regimes in the international economic arena. On the other hand, since most of the time we are not living under extreme situations, even the “all powerful” hegemon does not (and cannot) necessarily get involved in all aspects of the international economy. Thus, while a hegemon might be needed for the institution to exist and survive, and while it may be needed to weigh in during emergencies, the daily decisions and operations of international financial institutions are not necessarily determined by the hegemon. Overall then, the specific loans made by the RDBs and the lending policies the development banks pursue are often not significantly guided by hegemon(s). 2.1.2

Corporate Governance

If hegemons’ interests are not driving RDB policies, then the banks’ governance entity may be. The governance body of the RDBs consists of the management (staff) and the Board of Directors. As the RDBs struggle to define their identity, their proximity to the world of finance (e.g., banks, the private sector, global capital) seems to give at least their organizational structure corporate-like attributes. Theories of corporate governance may then provide a useful insight into the mechanism by which these governmental institutions operate. There are four primary models of corporate governance:9 state ownership and control, family ownership and control, bank-centered control systems, and control by dispersed shareholders. Control by dispersed shareholders is arguably the most modern; it is prevalent in developed countries, but as Beim and Calomiris (2001) claim, it works only if supported by legal and information structures. In developing countries, family ownership is more abundant, creating instability over time as families grow and change. The most important element of an effective governance mechanism, they posit, is information and control, and finding ways to align the incentives of the management with value creation (pp. 208–218).

9

Beim and Calomoris, 2001, pp. 208–218.

2.1 Who Rules?

47

This is easily applicable to RDBs: the structure of “shareholders” (member states) is similar to the model of market-based control by dispersed shareholders.10 It is considered the most flexible and responsive governance system. But even according to Beim and Calomiris, what seems to be most important are (1) the presence of at least one large shareholder who will have enough at stake to pay attention and lead decisions (resembling the idea of a hegemon), and (2) the existence of legal protection for shareholders (which is dependent on the availability of information and transparency). Both of these conditions, I will argue, need to be present for an RDB to function effectively. It can be said, then, that RDBs operate like corporations, where member states are shareholders while the president of the RDB essentially holds the position of a CEO. RDBs and similar organizations may therefore be usefully analyzed in light of the corporate governance literature that is becoming central to the economic analysis of developing nations. A “corporate-like” structure, however, does not mean one where considerations are purely financial and induced by profit-making exigencies. After all, even firms and commercial banks possess political elements.11 As for RDBs, while at various instances one state, or the president of the bank (CEO), may take a leadership role, often policy-making seems to be derived from coalitions. The disappointment over the failure of programs meant to confront poverty has led major multilateral institutions, such as the OECD, in a quest to find the reasons and possible solutions for the stagnation in developing countries.12 Studies have recently found that the role of governance is important in achieving good results (1996–1998 Program on Good Governance and Poverty Reduction carried out by the OECD Development Center). Hartmunt Schneider, in an OECD Development Center policy brief, advances the idea of Participatory Governance. This 10 11

12

Where the president of the bank and his or her staff act as management. See Baldwin, 1965. At the RDBs, member states, even if concerned with their ability to wield power or influence, give significant weight to the institutions’ ability not to extract from the callable capital they have committed to; thus, their financial interests are driven by maximizing access (procurement) and political capital while minimizing financial contributions – or at least limiting them to include only what is paid-in. In order to sustain this, the banks must guard their portfolio and ensure that the callable capital remains precisely that – callable. This is reflected in the 1990 World Development Report, the 1995 “Social Summit” in Copenhagen, the strategic document of the OECD Development Assistance Committee “Shaping the 21st Century” (1996), and in the UNDP’s “Overcoming Human Poverty” (1998).

48

Multilateral Governance: Theoretical Underpinnings

strategy is designed to complement new approaches for the reduction of poverty with a political-economy angle.13 The conventional definition of governance, as the “exercise of authority and control in a society in relation to the management of its resources for social and economic development” (p. 7), Shneider says, does not address the problem of information and agency: it assumes decision-makers have perfect (or at least sufficient) information about resources and needs (OECD Development Center, 1999). Participatory Governance aims at overcoming the problem of asymmetric information by introducing maximum transparency and sharing of information by all stakeholders (i.e., those involved in the process of development, including the beneficiaries). Fundamentally, this idea is parallel to Corporate Governance, where legally binding processes, transparency, and accountability are highly valued. Ultimately, Participatory Governance extends the idea that RDBs are modeled after corporations and uses approaches present in the Corporate Governance literature; when applied to RDBs, the theory suggests that the entire process of lending – loan application, design, process, administration, development, and implementation – should be conducted in a “shareholder” manner. All parties to the process should be stakeholders, not just the bank, making it more likely that information and problems will be processed and addressed in a timely and effective fashion. Using this framework we can gain further insight into the decision-making process taking place in the board rooms of RDBs and how it shapes their policies of loan-making. 2.1.3

Principal-Agent Theory and Delegation

The shareholders of the banks obviously cannot review every proposal or make daily decisions. As in most corporations (and government agencies), the shareholders must delegate authority to the management of the development banks.14 The management follows guidelines that are set in Articles of Agreement that were signed by the member countries (shareholders). In addition, it operates within the financial limitations given by the members in periodic replenishments (most RDBs have had seven or eight replenishments since their inception). These replenishments set the budget within which the bank can operate. 13 14

OECD Development Center, Policy Brief no. 17, 1999. For more on delegation see Pollack, 1997; Epstein and O’Halloran, 1999; Vaubel, 2006; and Hawkins et al., 2006.

2.1 Who Rules?

49

Principal-agent theory analyzes the degree to which principals delegate authority to their agents. This is a form of new institutionalism in rational choice theory that combines intergovernmentalism with a supranational approach: “The primary virtue of the new institutionalism in rational choice theory . . . is that it allows us to transcend the intergovernmentalist-neofunctionalist debate by acknowledging the initial primacy of the member states and, proceeding from this point, to generate a series of hypotheses about supranational autonomy and influence more precise[ly]” (Pollack, 1997, p. 101). There is no doubt that member states often delegate certain functions to the management and staff of the RDBs. This is a case of supranational autonomy. The daily activities of the development banks provide their officers with firsthand information and control over the way it is reported. The principal members do not have the mechanisms or the tools to get involved in every step of proposal preparation and decision-making. Where member states exert their power is mostly in instances of large political stakes (such as halting loans to a country that is not supportive of the United States during the Cold War, or, conversely, increasing the amount of loans to an important ally) or during economic crises. Thus, the development banks are often able to carry out their functions independent of member states due to RDBs direct access to information and to the control mechanisms set up by members. As Pollack argues: “. . . the ‘agency’ or autonomy of a given supranational institution depends crucially on the efficacy and credibility of control mechanisms established by member state principals, and that these vary from institution to institution – as well as from issue-area to issue-area and over time – leading to varying patterns of supranational autonomy” (ibid.). The analysis of the four RDBs is instructive in that it highlights the variations in delegation not just over time and in various issue areas of the institution, but also between institutions of similar scope and objectives. The variation in the configuration of principals among the RDBs impacts the principal-agent relationship even if the agents (institutions) are all similarly designed.15 Member states delegate authority to institutions for numerous reasons. An obvious reason is the cost and time associated with monitoring: principals are often reluctant or incapable of making the investment 15

Vaubel (2006) argues that the chain of delegation in international organizations is longer than other public or private organizations, making them more susceptible to principalagent problems.

50

Multilateral Governance: Theoretical Underpinnings

required for close monitoring. Second, some of the issues the institutions are responsible for are too complicated, technical, or detailed for the principals to address specifically. RDBs in particular, work in domains that can be highly specialized and require specific expertise. After all, it is for these reasons that states decide to create or join international institutions in the first place – to relieve the need to develop their own expertise and its costly mechanisms. Moreover, when institutions perform these tasks as agents, it may be technically impossible, and it would be certainly highly inefficient, for principals to control policy without delegation. This is because the information principals receive travels through numerous filters: first, experts in the field offices analyze proposed projects; next, management in field offices assess these analyses and pass them to the relevant division at the headquarters; and, finally, the RDB upper management reviews what is presented to them by their divisional managers. This all occurs before any information reaches the shareholders (member states). Thus, there is much room for the banks (agents) to maneuver, to create an organizational culture and set an agenda that is autonomous from the principals. This mechanism also affords the agent the opportunity to collect, control, and frame information presented to principals, granting it power to portray issues in certain ways and to insert its own agenda into proposals. Moreover, the professional background, nationality, and experience of an RDB’s president and staff impact the relationship between its principals and its agents. My examination of the relationship between the principal(s) and the agents reveals much about the institutions and their inner workings. Going back to the central research question – whether RDBs are bank-like institutions or development agencies, and how the various powers and structures within the institutions impact the way they pursue their lending activities – it becomes clear that unraveling the interaction between principals and agents is essential if we are interested in discovering the root of the development banks’ policies, and what is behind decisions to approve loans. In summary, principalagent analysis provides a substantive theoretical pillar for examining the RDBs and their positions on the bank/development-agency continuum.16 16

It should be noted that this Principal-Agent perspective is not necessarily contradictory to certain elements of hegemonic stability theory, where some power is delegated by hegemons while some is retained.

2.1 Who Rules? 2.1.4

51

Organizational Culture

One of the primary criticisms of the realist and institutionalist analyses of international organizations is the constructivist approach. This position holds that organizations have a power that is independent of the states that created them.17 Having much in common with some of the themes of the principal-agent arguments (e.g., the categorization of institutions as “agents”), constructivists emphasize the dynamic quality of international organizations. With the help of Weberian bureaucratic theory, proponents of this view maintain that the rational-legal authority creates normative power (Barnett and Finnemore, 1999, p. 700). This does not always mean that this power or independence is “positive.” As is the case of many international organizations (and bureaucracies), getting caught up in the “rules” can sometimes mean inflexibility and “bad” decision-making. Within RDBs, for example, even the noble goal of providing loans for development can create situations where institutional functionaries overlook important details concerning the projects in order to process the loans in a timely fashion (what critics call “getting loans out”) to meet budgetary demands. This is due to a sense of urgency to display loan-making, which is, after all, the development banks’ main operational tool. But the rush to get loans out the door can lead the banks to forgo some projects that are more urgent for development but less likely to be financed by any other financial institution because they are not deemed “bankable.” It also means that when new or unexpected situations arise, the rules cannot be easily bent or changed. For example, during financial crises or when financing needs to be expedited, the multiple bureaucratic steps required within the agency prevent flexibility. And though a manager at an RDB might have power in terms of bringing forth information and putting proposals on the table, the manager is not always capable of dealing with the specific issues surrounding each loan: he or she depends on information supplied by experts in local offices. Thus, there is a process of multiple delegations, making tight control by the principal even more complicated. Research into organizational culture from the field of sociology offers another insight into the operations of agents, labeled here as “groups.”18

17 18

See Barnett and Finnemore, 1999; Finnemore and Sikkink, 1998. See Buell, 1994; DiMaggio, 1997.

52

Multilateral Governance: Theoretical Underpinnings

“The existence of group level cultures (shared understanding partly independent of individual beliefs) is also suggested by the tendency of groups to adopt public positions more extreme than the preferences of their members, especially when acting with reference to a contrasting group” (DiMaggio, 1997, p. 273). This analysis supports the argument that institutions form a culture autonomous of the individual member-states. In addition, the people who operate as agents in the RDBs can be bound by a sense of “comradeship” and loyalty from belonging to the same organization and being familiar with the same terminology and practices. As DiMaggio explains: “ . . . one of the more notable characteristics of modern societies is the existence of a cultural division of labor in which intellectual producers intentionally create and diffuse myths, images, and idea systems” (ibid.). In reference to the RDBs, the staff shares their organization’s culture. In many ways, this explanation is consistent with hegemonic stability theory: by internalizing or anticipating the preferences of the principal, representatives of member states support policies that are advanced by the hegemon(s). Some other studies of organizations and firms describe an integrationlike procedure and see the organization as parallel to a family.19 This position holds that the closeness of the employees in the organization inspires an organization-wide consensus that is often described in highly emotional terms, and “[t]his emotional language leaves no room for dissent” (Martin, 1992, p. 47). Since consensus can be a form of conformity, this type of analysis raises interesting points for the study of RDBs, the central ones being: (1) Is there an organizational culture developed throughout the years? And (2) if so, do staff of the RDBs feel a greater loyalty to the institution than to their country of origin (i.e., are the nationalities of the staff significant, or does the institution produce a sense of belonging that transcends national borders?) Or, (3), is there no conflict at all between the nationalities and the organization? In addition, cultural explanations illustrate why institutions that are designed in a similar fashion operate differently. 2.1.5

Foreign Aid Theories

John White (1974) claims that “Aid” and “Development” are interpreted according to two opposing views: one, the cynical view, postulates that aid is merely a disguise or tool used by the countries that provide the aid in 19

Martin, 1992, especially pp. 45–70.

2.1 Who Rules?

53

order to dominate developing countries.20 The other, the benevolent theory, claims that the countries providing the development aid recognize an obligation to the less fortunate countries by assisting in welfare provisions (pp. 12–13). But interpreting aid in either of these ways mixes in a subjective message of ideology or morality – a prescriptive message, as White (p. 13) asserts. Rather than examining the institutions that conduct these activities and analyzing their nuances, such moral/prescriptive views paint a black-or-white picture. White suggests four types of theories of aid21 that are predominant:

Domestic External

Economic (transfer)

Political (transaction)

Positive, i.e., supplemental theories Negative, i.e., displacement theories

Recipient-oriented, i.e., comparative politics theories Donor-oriented, i.e., international relations theories

The above categories provide a useful analytical framework for the analysis of the RDBs. Theories produced in all four quadrants of this chart can be valuable when investigating the RDBs. Any analyses of RDBs must cover their dilemmas in maneuvering the demands of donors and recipients; the conflicting assessments over whether they contribute to development or not (depending on whether we employ a “domestic” or “external” approach); and the political actors we deem important in analyzing RDBs. Additionally, the analysis of RDBs must take into account that development may not necessarily be the result of aid; development is a process. And how can we measure the process? One way is to think of development as social change. We could then measure development by looking at changes in socioeconomic indicators. Tying this to the work of the RDBs, we then need to compare the function of the banks (i.e., loans) to these socioeconomic indicators. Of the many discussions and debates about aid, the most striking distinction is between those who assess aid as “interest-based” and those who evaluate it as “need-based.” Interest-based explanations have been dominant in the field (Gould, 2006; Vreeland, 2003; Alesina and Dollar, 2000; Thacker, 1999; Mosely, Harrigan, and Toye, 1991; McKinlay and Little, 1977). This literature postulates that much of the 20 21

For more on the reasons states join IOs see Abbott and Snidal, 1998. White, 1974, 105.

54

Multilateral Governance: Theoretical Underpinnings

aid given by developed countries is strategic and related to the interests of donors, rather than directed at the neediest and poorest developing nations (Alesina and Dollar, 2000). McKinlay and Little (1977) explain that the need-based view of aid is controversial because the aid received by low-income countries bears no linear relation to the level of their need, and major donors have been found to be reluctant to increase their multilateral aid contributions, focusing on bilateral aid over which they exert control (p. 59). It follows from these points of evidence that aid provides donors with leverage over recipients. For example, during the Cold War, aid programs were an integral part of the rivalry between the superpowers. If the donors hold a hegemonic status in RDBs, there is no reason to believe that these findings about bilateral aid are not relevant also for multilateral aid agencies. But, if donors are not leveraging power over IFIs so as to advance policies similar to those they follow bilaterally, then the advent of IFIs since McKinlay and Little’s 1977 study (and donors’ willingness to, in fact, increase their contributions to multilateral aid) may demonstrate that the advantages offered to donors in a multilateral setting outweigh preferences that are satisfied by bilateral aid. Testing multilateral financial institutions in this light is a “hard case” since the interest of a single power is not as easily detectible in a multilateral setting as it is in a bilateral one. The critics of multilateral financial institutions who claim these institutions are used as a political tool in the hands of a hegemon(s) subscribe to what McKinlay and Little (1977) refer to as the foreign policy view of aid. However, in a multilateral setting it might not be as easy for donors to advance their foreign policy goals. Thacker (1999) argues that the United States does employ political consideration from its power position on the Board of the IMF, and such accusations have been directed at the WB and RDBs as well.22 To the extent that concerns of power and security dominate donors’ decision to support multilateral development banks (following an interest in aiding development, with the assumption that more developed countries pose less of a security threat and provide expanded markets) a foreign policy view of aid through multilateral financial institutions (as opposed to a humanitarian one) can be put forth.

22

See, for example, Babb (2009) for an analysis of US foreign policy and the MDBs.

2.2 Hypotheses and Research Design

2.2

55

hypotheses and research design

RDBs were created for political reasons and their founding would not be possible if it were not for the commitments of rich (donor) countries. In fact, the foundation of each development bank came about primarily through the aegis of one dominant (at times regional) hegemonic power. The formation of RDBs has been directly linked to the international political environment. Most distinctly, Cold War and post–Cold War politics have been decisive factors in the willingness of donors to orchestrate the creation of RDBs. But despite their strategic origins, I contend here that these institutions were put in place to fill a gap in the development needs of developing countries and not simply to exert power and mark spheres of influence. The smaller membership and funds of the RDBs, compared to that of the WB, has left them relatively out of the spotlight.23 This reduced visibility allows borrowing countries to be more actively involved in decision-making and to have more influence on the institutions’ policies. There is a “limited hegemony” in these types of regional institutions, allowing regional members to take a more active role. Though the hegemon/s have certain guidelines and “red lines,” within these boundaries RDBs have more options for flexibility than do global institutions.24 Thus, I argue that RDBs are better capable of shifting their strategies in light of global changes, and are better suited to focus on current urgent development needs. It should not be surprising to find that an institution that exclusively deals with one region is better equipped to address the idiosyncratic needs of a particular country in that region, which increases the likelihood that programs and projects are effective.25 Whether the RDBs are primarily driven by profit motives (bank) or humanitarian/development [development] motives, questions remain open about the extent to which hegemons are involved in policy-making decisions about loans. If it is found that powerful Western countries are highly involved in the decisions of RDBs due to political interests, then 23

24

25

While in recent years the RDBs total funds exceed that of the WB, individually, their annual disbursements are still smaller than the WB’s annual disbursements. This is due, to a large extent, to the fact that reduced public visibility of the RDBs decreases the political stakes that hegemons have in “appearance.” As a result, hegemons may be more willing to lower their costs by delegating more responsibilities to the agents. Although, theoretically, additional geographical specificity is always possible (and proof of that is the growing volume of even smaller regional banks), the RDBs as they exist today are big enough institutions that they attract attention and funds, but yet remain more focused and specialized than their global counterparts.

56

Multilateral Governance: Theoretical Underpinnings

loans may be tainted with donors’ political preferences rather than driven by development concerns. Alternatively, such involvement by hegemon(s) may be seen to actually enhance development-driven policy, since a hegemon is more apt to take risks. The hypotheses and research design of this study aim to test whether these critiques – which are essentially a result of the conflict between the bank function and development agency function of the institutions – have merit. This discussion, therefore, leads to the following hypotheses: Hypothesis 1: Consistent with their hybrid institutional structure, the lending of regional development banks is strategic: their primary concern is the repayment of the loans rather than the need of the recipient.26 This hypothesis suggests that If RDBs strategically lend more (more loans and higher amounts) to countries that are better off, then that would demonstrate that they are positioned closer to a “bank” on the “bank/ development agency” continuum. Making loans to countries that have stronger economies is less risky because there is a higher likelihood that when the loans mature, they will be paid off (with interest). Socioeconomic indicators of the recipient countries, such as annual household incomes, private investment, health, and education, measure the level of risk – that their lending is not closely associated with indicators of economic development. Thus, a correlation between the amounts and destination of loans and socioeconomic conditions in the recipient countries suggests that the RDBs are not giving preference to the poorest countries. This would mean that the RDBs lend strategically: their primary concern being the repayment of the loans. This strategy puts them at odds with their mission and their commitment to be lenders of last resort (i.e., not to compete with other potential lenders). In testing this hypothesis, I compare the relative loan disbursements of RDBs (dependent variable)27 with economic, social, and financial attributes of each of the countries in their regions that are eligible for loans. Testing regression models include independent variables such as GDP, foreign direct investment (FDI), population, infant mortality, and primary education rates. The results account for the differences in wealth and size of recipient countries, presenting a proportional analysis that takes into account the value of loans to countries that receive them, given their level 26

27

As discussed in Section 1.3, the RDBs interest in ensuring that borrowers do not default on their loans can steer its policy toward less than purely developmental priorities. This is further elaborated on in later chapters. The loans are factored with GDP and population.

2.2 Hypotheses and Research Design

57

of development. The unit of analysis in the regressions is country-year in all the years of each RDB’s existence. Strategic lending will, then, translate into consistently more loans to countries that enjoy more FDI, higher GDP and GDP per capita; lower infant mortality rates and a higher literacy rate. This hypothesis addresses some of the harshest criticism of international financial institutions – those that argue that IFIs provide no basic added value compared with other financial means that might be available to emerging markets. Examining this hypothesis, therefore, will help to answer those charges and determine whether the development banks strategically choose certain borrowers over others, and how these decisions are related to global financial markets. I expect the testing of this hypothesis to show that the strategy of lending changes over the years, especially since the end of the 1980s, as political considerations at the end of the Cold War made way for financial considerations thereafter (financial considerations, in essence, are different than political considerations as we shall see in the next three chapters). Hypothesis 2: The lending of RDBs is influenced by ideological proximity to the hegemon. This hypothesis suggests that a hegemon(s) is needed not only for the creation and continued existence of the RDBs, but also for them to follow policies that are more in line with a development agency (this is despite the fact that the structure of RDBs enables a more balanced involvement of other member countries compared with the WB, including the developing members, in the governing of the bank).28 It not only questions the degree to which hegemons ensure RDBs existence, but also suggests that the power vested in hegemons can be used for “good.” To test this, an independent variable representing countries’ mutual interests is added to the data set:29 for the IDB, the United States’s mutual interest with borrowing members is correlated with loans made by the banks. For the AsDB, I analyze a correlation between Japan’s and the United States’s shared interest with borrowing members and bank loans. For the AfDB, I test a correlation between French and American shared interest with African borrowing members. Since they have fewer shares than the 28

29

This also speaks to literature on the WB and IMF, which mostly suggests that hegemons show a preference to countries with which they are politically aligned (see, for example, Thacker, 1999). Data originally coded by Erik Gartzke and updated and maintain by Strezhnev and Voeten (2008–2012), based on UN General Assembly voting (see Chapter 4 for more details).

58

Multilateral Governance: Theoretical Underpinnings

borrowing African states, the United States and France are less influential at the AfDB. And, for the EBRD, I test a correlation between shared interests of the UK, Germany, France, and the United States with borrowing members and the EBRD loan distribution. This hypothesis is a product of the mounting critique directed at international financial institutions for being arenas that represent a balance-ofpower scheme, and that hegemon(s) on the Board essentially use them as extensions of their foreign policy – a critique that is especially striking within studies on the World Bank and IMF.30 The RDBs are no different in this regard – they also may be potentially fruitful arenas for advancing policies preferred by hegemons. Chapters 4, 5, and 6 demonstrate that the configuration of the Board – and hegemon(s) – impacts the ability of RDBs to pursue policies that are judged on the merit of their development impact rather than on the policy preferences of powerful donors. Surprisingly, where there is a single dominant hegemons – as at the IDB – there seems to be more delegation and less “hegemonic influence.” Moreover, where the donors have fewer shares (the AfDB) policy preferences do not necessarily favor poverty alleviation and development over sound banking. Hypothesis 3: Different membership compositions and institutional structures influence the RDB’s ability to function more like a development institution or more like a bank. Hypothesis 4: The number of hegemons and the degree of their involvement with the RDBs’ operations affects their ability to perform their “development agency” duties. The differences in performance and outcomes among the RDBs can be accounted for by (a) the different level of involvement of a hegemon(s); (b) the different organizational approaches of the founding/principal members; and/or (c) the different modes of operation and organizational culture that result from differences in location and influential members. Thus, institutions that evolve and perform differently, despite similar origins, may do so because of in-house cultures that have been influenced by varied member states, problem-solving styles, and levels of involvement by the principal actors (hegemon[s]).31 Although the Articles of Agreement of the IDB, AsDB, and AfDB were based, in spirit and in practice, on the Articles of Agreement of the WB, 30

31

This is in par with claims that under the auspices of the “Washington Consensus,” IFIs are influenced and controlled – overtly or covertly – by American policies. See Babb, 2009; Weaver, 2008; Stiglitz, 2002; Easterly, 2001; and Thacker, 1999. See Vaubel (1986) for a Public Choice approach.

2.2 Hypotheses and Research Design

59

each of these banks differ markedly when it comes to shareholder composition. The IDB, for example, allocates 30 percent of the shares to the United States making it the single largest shareholder at the bank.32 At the same time it also commits to ensuring that at least 50 percent of its voting shares be assigned to borrowing members. The AsDB guarantees equal (and highest) shares to the United States and Japan, while counting three nonborrowing members in its “regional” pool (Japan, Australia, and New Zealand). In contrast, the AfDB guarantees not only a majority of voting shares to its borrowing members, but also ensures that borrowing member(s) are the largest shareholder(s). Finally, the EBRD is deliberately a “donor” institution where donor members hold the majority of shares and votes. These different compositions affect the hierarchies within the institution and the degree to which powerful donor shareholders delegate policy-making authority to borrowers, who are shareholders as well. Since shareholder composition is set in the Articles of Agreement, there is not much room for change. In fact, over the years, the voting shares of member states changed very little, if at all. The only opportunity to alter the allocation of shares is when the Board of Governors approves a capital increase. The IDB, for example, had only nine such General Capital Increases between 1961 and 2012. When the capital of an RDB is increased, the relative contributions of members can be weighed and altered, all within the parameters of the Articles of Agreement. However, in reality, vote shares are unlikely to change unless a new member has been admitted. When that happens, the shares are typically allocated from the peer group of the new member (i.e., donor, borrower, or, in the case of the IDB, the group of countries the new member belongs to). Furthermore, the percent of shares held by nonhegemonic donors is negligible since access to procurement, the central benefit to donors, is granted no matter how small the share.33 If we analyze the political positions of the most powerful shareholders in each RDB, we can determine whether trends in the loans reflect these positions. The Political Proximity Hypothesis, based on Thacker (1999), is useful to this end. Much like at the IMF (the subject of Thacker’s study), the Boards of Directors of the RDBs vote on whether or not to approve loans. Since various members of the Board have different voting shares, their positions affect the decision to lend differently. For example, in the case of 32 33

See Chapter 6 for a discussion on American “de facto” veto power. Based on interview with Carlos Jarque, Director of the IDB’s Europe Office, April 2012. This is further elaborated in Chapter 6.

60

Multilateral Governance: Theoretical Underpinnings

the IDB, Cuba is not a member, although some IDB officials and representatives of certain IDB member countries support lending to Cuba, especially because of its dire economic and social circumstances.34 But despite these voices, the IDB follows US policy regarding the sanctions on Cuba. As for the AsDB, China was permitted to become a member only in 1987, Taiwan (Taipei China) is still a separate member, North Korea is not a member, and Afghanistan’s membership was not operational between 1980 and 2004. Although this only informs us as to who is “in” and who is “out,” it does suggest that politics plays a role in the decision-making of the banks. In accordance with the Political Proximity Hypothesis, this suggests that more political alignment with a hegemon yields more financial support. An in-depth look into the internal processes of RDBs can help shed light on the “politicking” that goes on inside them. Former AsDB officials have explained that requests for loans usually do not reach the voting stage unless preapproved by the United States.35 In addition, there seems to be a clear competition for power in the AsDB between the United States and Japan, but ultimately policy is approved only with the support of both.36 Voting scale impact on loan probability:37 A 0 (−)

B 0.5 (neutral)

C 1 (+)

This schematic representation of the Political Proximity Hypothesis suggests that countries at point A, on the far left-hand side of the voting space are very unlikely to receive loans, while countries near point C, that are the perfectly politically aligned with powerful shareholders have a very good chance of receiving loans. Being “neutral,” at the center of the voting space, does not impact the chances of receiving aid. In the case of RDBs, I examine whether political considerations and the likelihood of borrowers not defaulting on loans are significant contributing factors to the approval of loan disbursements.

34 35 36

37

Based on interviews with IDB officials, Washington, DC, 1999–2000. Barry Metzger, former General Counsel to the AsDB, June 2000. Ibid. Mr. Metzger, in his role as General Counsel, witnessed friction between the US and Japan in setting the AsDB’s policy preferences. He stressed that the US is concerned with ensuring that Japan’s influence in the institution does not surpass that of the US and that the US maintain its position of power. To that extent, decision could not be reached without the agreement of both. Thacker, 1999, 47.

2.3 Conclusion

2.3

61

conclusion

In this chapter we reviewed the theoretical foundations that help us understand the phenomenon of multilateral development banks. In assessing the RDBs activities on the bank–developing agency continuum, we can draw explanations from various theories and approaches, such as Hegemonic Stability Theory, principal-agent theories, cooperation, aid, corporate governance, and organizational culture. These foundations do not only explain the variations between the development banks, but also the changes within the RDBs over the years. For a complete analysis of the loans made by the development banks – why strategy differs from one bank to the other and in the same bank over time – the study of each RDB through this theoretical prism is indispensable. The hypotheses, presented in the last section of this chapter, are explored and tested in the next four chapters. Chapter 3 provides necessary background on the RDBs and their specific histories, while Chapter 4 explains and describes the variables used in the quantitative and qualitative analysis demonstrating, for example, the aggregate amounts of FDI each of the borrowing regions has received in the last five decades, the changes in health and education, in addition to total lending to each region. It further examines the stark differences between the poorest and wealthiest borrowing member of each region, showing that the poorest of the regions are not the recipients of more loans. An analysis of the hypotheses is found in Chapter 5 via a quantitative analysis, followed by a qualitative assessment in Chapter 6. Chapter 5 further examines the correlation between the borrowing members’ socioeconomic indicators, polity, market openness and the likelihood that they receive loans. It also establishes whether a borrowing country’s policy preferences as they relate to each RDB’s hegemon(s), in the context of UN voting, are related to loan allocations. Overall, it concludes that regardless of hegemonic configurations, RDBs do not hold a clear preference in their lending to countries that supposedly need aid the most and have less access to private credit. The Cold War, this analysis shows, makes a difference mostly for the role played by US hegemony but preference for democracy is not substantiated, in addition to a more nuanced overall role for hegemons. Chapter 6 introduces the various institutional histories, practices, and characteristics into the equation, and based on archival research and interviews, it explores the differences between the RDBs. It shows that the conflicting identities of bank and development agency are common to all RDBs, whether they are dominated by hegemons or by borrowing

62

Multilateral Governance: Theoretical Underpinnings

countries. This chapter explains the interests of donors to hold membership in the RDBs – not necessarily to wield influence over decisionmaking, but often to have access to markets and be part of the “club.” However, it also demonstrates that the framework of the RDBs affords borrowing members a certain amount of ownership over development policies that impact their economy. In proposing answers to the central questions of this study, these hypotheses offer a framework to examine why the RDBs approve and disburse loans, how the internal institutional mechanism and external environment might affect these decisions, and whether there is a change in lending strategies over time. Throughout this book, the interdisciplinary theoretical frameworks presented here help explain the phenomenon we analyze.

3 Origins, Politics, and Structure of Regional Development Banks

Neither a borrower, nor a lender be; For loan oft loses both itself and friend, And borrowing dulls the edge of husbandry. – Hamlet, I, iii, 75

Have RDBs evolved in a changing global environment? What were the reasons for their creation – political, economic, and social – and are they still relevant? The twentieth-century motivations for creating international institutions such as the UN (WWII), the Bretton Woods Institutions (the Great Depression and WWII), and NATO and the RDBs (the Cold War) have given way to an increasingly globalized world economy. IFIs share credit for these global changes, but it also means that they might have to reinvent themselves so that their contributions remain effective. This chapter traces the historical developments of the RDBs. An examination of the development banks’ origins, their evolution as institutions, and their structural components is a valuable basis for discerning the lending policies of the RDBs over the years in order to determine whether they are motivated by political, economic (bankingcentered goals), or social factors (developmental goals). The reasons (both declared and implicit) for the RDBs’ creation shed light on their development as institutions and the political relationships between the member states. Understanding the RDBs’ institutional design – including membership, decision-making and voting, the assignment of veto power, staff, Articles of Agreement, budget, and credit ratings – is essential for evaluating the banks’ activities. Both the political 63

64

Origins, Politics, and Structure of RDBs

environment that paved the way for the establishment of the RDBs and the institutional model upon which they were created, constitute the foundation for political developments within the institutions and between them and their members. Moreover, global political and economic developments, often informed by evolving norms and ideas, help shape states’ attitudes toward multilateral institutions, and enable some of these institutions to outlive the political reasons that contributed to their creation. In sum, by presenting an overview of the RDBs’ history and the interests – internal and external – that shape their institutional design, this chapter sheds light on the controversy surrounding their role as creditors.

3.1

founding principles

The Inter-American Development Bank, African Development Bank, and Asian Development Bank were created in the 1960s to supplement and improve upon the work of the World Bank, after developing countries voiced disappointment with the aid and development policies of the global institution. Set up after WWII, the WB’s main focus was the reconstruction of Europe. But for many years prior, starting in 1890, poor Latin American countries had been pressing for support to create a regional institution to respond to their unique problems.1 A convention for the establishment of an Inter-American Bank was signed by the United States in 1940, but then abandoned during WWII. Following the war, the establishment of the World Bank seemingly rendered the creation of a regional bank pointless. It is no coincidence that the RDBs were founded in the 1960s. The timing of their creation reflected not only the growing number of postcolonial underdeveloped nations, but also the international political reality at the time and the needs of the sponsoring nations. Most significantly, at the height of the Cold War the United States reversed its initial opposition to the creation of these institutions. Financial institutions were seen as a tool to keep developing countries in the American sphere of influence, making member countries part of a cooperative effort from which “defecting” would be costly.2 As the Cold War entered its third decade, securing spheres of influence became even more important for America, especially in light of the accelerating arms race and the perceived threat to American security. 1 2

See Tussie, 1995; and White, 1974. See, for example, Babb, 2009 (especially pp. 48–49).

3.1 Founding Principles

65

The establishment of the first RDB, the Inter-American Development Bank (IDB), in 1960 came merely one year after the success of the 1959 Communist Revolution in Cuba led by Fidel Castro. It is no wonder that the United States was eager to boost its presence in Latin America: tumultuous political developments and geographic proximity made aiding Latin American a strategic political move.3 The establishment of a regional financial institution supported by the United States, ensured regional cooperation with the United States and enhanced American alliances against the Soviet Union and its communist stronghold in Cuba. Accordingly, the IDB was headquartered in Washington, DC, close to the White House and Capitol Hill, the pillars of power that frame American foreign policy decisions. And, subsequently, Cuba has never been a member of the IDB, an issue that will be discussed in the following chapters. It is significant that the IDB was a product of regional activism in Latin America: ideologically, the IDB was an offspring of the state-led ideas that bloomed in Latin America in the 1950s (in the form of the Economic Commission for Latin America).4 This economic vision led the financial policies and its vision of the IDB in its first years and was (and still is to a certain extent) stamped into the bank’s mode of operation. The structuralist school of thought that dominated at the time justified a strong state leadership to overcome a weak public sector and insufficient capital. This explains the bank’s close relations with states and state-led policies even to this day.5 Thus, from the onset, the IDB’s policies reflected a merger between the development vision advanced by potential recipients of the loans and political motivations held by the major donor(s) of the institution. Once the IDB was established, its officials actively supported the creation of Asian and African Development Banks.6 The existence of similar regional financial institutions, IDB managers believed, would increase the 3

4 5

6

In the late 1950s there was growing resentment of the US in Latin America (as was demonstrated by the hostile reception of Vice President Nixon in his May 1958 tour of Latin America) (Culpeper, 1997, p. 28). See Tussie, 1995, chapter 1. As noted in Chapter 6, since the end of the Cold War, private international banks and financial actors have become central players in the region and their importance is not overlooked by the IDB (or the other RDBs). However, making the necessary adjustments to allow private actors more access to the banks takes time and requires much diplomatic finesse (see Erica Gould, 2006, for an argument about the effect of the private sector on loans for the IMF and World Bank). Culpeper, 1997, p. 28.

66

Origins, Politics, and Structure of RDBs

RDBs’ value and make them more visible, especially with regard to the WB and the IMF. In 1961, African countries asked the UN Commission for Africa to study the possibility of a regional bank in Africa. The agreement to create the African Development Bank (AfDB) came into effect in September 1964 following almost three years of negotiations primarily within Africa.7 The AfDB began operating in July 1966. However, unlike the IDB that was founded on the premise that the United States would play a central role in the institution, the AfDB was created as a predominantly African institution, setting its headquarters in Africa (Abdijan, Ivory Coast),8 and limiting the power of donor countries on its board. The AfDB’s Articles of Agreement stress: “no member country, or group of countries has veto power,” (Articles of Agreement, role of Board of Governors). Given its establishment soon after the IDB (where the United States has veto power), this assertion holds significance – it points to the already different emerging role of nonborrowing members at the AfDB. The AfDB was created in Sudan and its 23 founding members were newly independent African states, eager to establish an institution that was their own. To that extent, until 1982, only independent African countries were eligible to be shareholders and provide the capital for the development bank. In 1982, nonregional members were granted shareholder status thereby enhancing the AfDB’s capital and investment portfolios. Even after this adjustment, the AfDB still boasts its unique African-regional character by emphasizing its location (always in Africa), the nationality of its President (always African), and its ownership structure (majority shares held by borrowing African members).9 Furthermore, the Articles of Agreement establish that decisions, in general, would not be made by voting but, rather, through discussion and consensus. This is confirmed in conversations with policy makers who 7

8

9

These negotiations largely took place among newly independent African states that were eager to take political and economic matters into their own hands. It was an ambitious venture, especially as they had little experience running their own affairs, and the IDB had only been operative for a few years, while the AsDB was not yet established (see English and Mule, pp. 19–20). “The vision behind the establishment of the African Development Bank was clearly one of self-reliance through cooperation among African states, as proclaimed in the preamble to its 1964 agreement: the Governments on whose behalf this agreement is signed; determined to strengthen African solidarity by means of economic co-operation between African States . . .”(ibid., p. 19). African Development Bank Group In Brief (1999, p. 5).

3.1 Founding Principles

67

have worked with the AfDB: it appears that actual voting, or the exercise of one’s voting power is marginal. More often, discussions – both formal and informal – precede any decision, and controversial issues are not usually on the Board’s official agenda.10 The catalyst for the establishment of the Asian Development Bank (AsDB) in 1966 was “the build-up of the Vietnam War and the United States’s desire to strengthen the regional economies as bastions against Communism” (Sherk, 1999, p. 1).11 Although discussions to establish the bank started in 1963 under the auspices of the UN Economic Commission for Asia and the Far East, it was not until April 1965 when “President Lyndon Johnson called for economic development to spearhead a peace initiative in the region and shortly thereafter supported the proposal to establish a regional bank” (Culpeper, 1997, p. 28), that the AsDB became a reality-in-the-making. Thus, although discussions were under way (and the idea was first proposed in the late 1950s) to establish the AsDB, it was not until US support became available that the idea could materialize. The expert group that met in 1965 made the argument for the establishment of the regional institution on three grounds: “. . . that the AsDB should (a) be a conduit for channeling additional resources to the region; (b) finance projects that were not adequately funded by other donor agencies; and (c) act as a focal point for regional activities that would promote economic cooperation” (Kappagoda, 1995, p. 13). In August 1966, the agreement establishing the AsDB became effective and the AsDB’s operations commenced in December 1966. But, despite the crucial involvement by the United States, donor power on the Board of the AsDB includes equal shares to Japan and the United States, making the AsDB subject to dual-hegemonic influence that carries with it policy implications. The composition of the AsDB’s Board of Directors is, then, markedly different than that of the IDB and AfDB: it does not have the single donorhegemon design (IDB), nor does it have a design that allocates the majority of shares to borrowers (AfDB).12 Despite the location of its headquarters in the region (Manila, Philippines), the two most powerful shareholders are both its central donors, one a regional hegemon seeking to guard its influence in Asia (Japan), and the other, the United States, that is concerned with maintaining its superpower status in the region. Furthermore, as the AsDB’s founding Articles stipulate, Japan holds the presidency of the AsDB, 10 11 12

In Chapter 6, I discuss voting shares and their significance in more detail. Donald Sherk, 1999. Both the IDB and AfDB have a majority-recipient shareholder division.

68

Origins, Politics, and Structure of RDBs

while its vice president is always an American national. This power structure is indicative of the two powers’ desire to maintain a close connection to the steering wheel of the institution. The political events that led to the creation of the European Bank for Reconstruction and Development (EBRD) are more recent, and the motivations for its existence appear to differ from those of the other banks. The initial proposal to establish the bank, in 1989, was prompted by French President François Mitterrand. In the waning days of the Cold War, Western Europe was looking at its Eastern neighbors as potential markets and allies. On December 9 of that year, the European Council at Strasbourg strongly endorsed the French initiative. Shortly thereafter, a meeting was convened in Paris (in January 1990) attended by representatives from all twenty-four members of the OECD along with representatives from Malta, Cyprus, eight Central and Eastern European countries, the European Economic Community, and the European Investment Bank (EIB).13 The EBRD was established with a different mandate than the more senior development banks; it specifies that loans are conditional on efforts for democratization and transition to market economy, and it is more flexible in making loans to the private sector.14 To that end, the creation of the EBRD was perhaps the speediest of all other multilateral financial institutions: Negotiations were concluded in May 1990, and “[t]he agreement that established the EBRD entered into force in March 1991” (Culpeper, 1997, p. 29). Overall, then, the EBRD’s raison d’etre appears to deviate from the stated policies of the original three RDBs that were officially designed to assist poor countries and have no formal political agenda: many of the borrowing countries in Latin America, Asia, and Africa were far from adopting democratic measures or market economies throughout the years they received loans. During the Cold War, the major political consideration of Western donor countries (namely the United States) was the support they were receiving from developing countries in their confrontation with the Soviet Union, rather than the domestic political system of these countries. No longer concerned with spheres of influence associated with the Cold War, Western donor countries were able to shift the agenda of the newest RDB and focus on ideology rather than political proximity.

13 14

The discussions were joined subsequently by six other non-European countries. In fact, the EBRD came under severe fire in the mid-1990s, very shortly after it began operating, for making loans to projects that could have been financed by the private sector or commercial banks.

3.2 Comparing the RDBs

3.2

69

comparing the rdb s

A comparison of central parameters of the MDBs – the RDBs and the WB15 – demonstrates some striking differences between the institutions. Important features of the RDBs that lay the foundation for the way they operate include the number of donor member states versus borrowing member states; the portion of member states’ committed capital that is paidin; and the credit rating they receive. Additionally, a comparison of approved and disbursed loans demonstrates the financial scope of the banks. All these are indicative of the institutional pressures guiding RDBs’ policies. Table 3.1 highlights some notable indicators that showcase central differences between the RDBs. First, with 28 percent paid-in portion of subscription, it seems the AfDB requires a higher percentage of paid-in capital to cover the costs of its operations. This leaves less subscribed capital in reserve, which affects the AfDB’s overall financial portfolio – lowering its credit rating, overall financial means, and reducing disbursements (in comparison to committed capital). Using a larger percentage of subscribed capital can also send a negative signal to potential investors: it indicates that since the bank draws more of its “promised” capital, it may have trouble attracting private capital or investing its assets well. Second, it is important to note the ratio of donors to borrowers: while the IDB, AsDB, and AfDB all have more borrowers than donors, the EBRD (and WB) have more donor member states than borrowing ones. It is evident therefore that for the first generation of RDBs, ownership of the banks by their borrowers, the regional members, was imperative. Finally, resolving to value high credit ratings (AAA being the highest), the RDBs are more likely to favor policies that reinforce these ratings. The reasoning behind the preference for high credit ratings is that it sends positive signals to private sector investors and member states about the development bank’s ability to successfully invest and gain access to capital markets. These credit ratings actually reflect the credit ratings of the major donors, making it essential for the RDBs to maintain a strong donor shareholder portfolio.

3.2.1

The Inter-American Development Bank

The IDB was founded in 1960 and is the oldest and largest RDB. Through June 30, 2012, forty-eight members (shareholders) of the IDB supported 15

I included the World Bank in this table to provide the reader with a benchmark when assessing RDBs’ institutional indicators.

table 3.1 The MDBs in comparison (2009)

Number of Shareholders (percent voting shares)

Largest Shareholder (percent voting shares)

Gross Loans, Disbursed (US $ billion)

Paid-In Paid-In Capital Portion of (US $ billion) Subscriptiona

Bank

Year of Foundation

IDB

1960

Donor 22 (49.985%)

Borrowing 28 (50.015%)

US (30.006%)

11.42

5.4

AfDB

1964

24 (39.79%)

53 (60.21%)

Nigeria (8.7%)

3.7

3.7

AsDB

1966

22 (49.96%)

45 (50.04%)

7.9

4.1

EBRD WB (IBRD)

1991 1944

34 (87.65%) 91 (54.42%)

30d (12.35%) 94 (45.78%)

Japan, US (11.657% each) US (10%) US (15.3%)

7.88 18.6

7.17 11.5

5% (10% until 2003) 11% (28% until 2003) 7% (16% until 2003) 25% 25%

Rating & Business Profileb (Standard & Poor’s)

AAA; very strong AAAc; very strong AAA; extremely strong AAA; very strong AAA; extremely strong

Sources: Standard and Poor’s 2012 Supranationals Special Edition, Moody’s 2012, RDBs Annual Reports (various). a b

c d

Paid-in capital reflects actual liquid assets available to the bank; and callable capital represents commitments for credit that are not paid-in. Business profile reflects confidence in governance and management expertise and policy importance (Standard and Poor’s Supranationals Special Edition 2012) The AfDB’s credit-rating was raised to AAA in 2003 by Standard and Poor’s. Until the year 2000 the EBRD had 26 borrowing members.

3.2 Comparing the RDBs

71

the bank through nine capital increases (replenishments). Approved in 2010, the last capital increase raised the IDB’s Ordinary Capital Resources (OCR) to US $170 billion.16 Of this amount, US $10.7 billion is paid-in capital, while the rest constitutes callable capital. Its membership includes twenty-six regional (borrowing) members and twenty-two nonregional (donor) members (thus, the majority of its members are recipients of loans).17 Member states are expected to approve loans that would be consistent with the bank’s mandate “. . . to accelerate economic and social development in Latin America and the Caribbean” (Standard & Poor’s, 1999).18 The vague nature of the mandate and the development bank’s concern with maintaining high credit ratings may encourage managers and staff members to follow profit maximizing policies that could be at odds with the IDB’s ostensible commitment to poverty alleviation and development.19 To that end, the credit rating of the bank has been AAA since 1962 (based on Standard & Poor’s 1999 ratings).20 Since ratings are inextricably linked to donor countries’ financial profiles, the IDB is concerned with maintaining good standing with its chief donors, namely the United States.21 The IDB’s central role is to provide long-term financing for largeand small-scale projects to its regional members – primarily in the public sector – as well as project technical assistance. According to

16 17

18

19

20

21

IDB Annual Report, 2011, 2012; Standard & Poor’s, Supranationals Report, 2012. China is the latest nonborrowing member to join – in 2009; the overall voting shares of borrowers and donors remain the same, and when new donors join, there a shift only in donor shares, save the US, which maintains its 30.01 percent of the shares. “The Bank’s Charter states that its principal functions are to utilize its own capital, funds raised by it in financial markets, and other available resources, for financing the development of the borrowing member countries; to supplement private investment when private capital is not available on reasonable terms and conditions; and to provide technical assistance for the preparation, financing, and implementation of development plans and projects” (http://www.iadb.org/exr/english/ABOUTIDB/about_idb.htm). “. . . the IDB seeks to eliminate poverty and inequality, and promotes sustainable economic growth” (www.iadb.org/about). This mission is repeated in all bank documents and many of its public statements. The rating of all of these multilateral financial institutions is based on the ratings of the callable capital of member countries: they are determined independently of the RDBs’ loan portfolios (based on interview with Dr. Adam Lerrick, February 19, 2001). Numerous officials at the IDB, AsDB, and EBRD explained that the high credit rating is what makes it possible for the banks to borrow, attract investment, and sustain their donor support. They claimed that the credit rating is determined by rating agencies mostly based on the portfolios of donor members and their shares in the bank (interviews 2000, 2001–2002, 2007, and 2012).

72

Origins, Politics, and Structure of RDBs

Standard & Poor’s, in recent years the bank also has financed sector reform loans and debt reduction (ibid.). The IDB’s capital consists of donor’s paid-in capital and revenues from loans and investments: the bank operates under financial principles similar to those of private banks, even though it is not a profit-maximizing institution. “It receives interest income from its loans and investments of funds not immediately needed for disbursements. Income derived from its lending operations and investments covers expenses arising from the bank’s borrowings in capital markets and administrative costs and provisions, and produces net income sufficient to meet financial ratio targets” (http://www .iadb.org/info/bfingles/bank9.htm). Since the paid-in capital of major donors is a fraction of the total capital commitments, the IDB’s investment and loan repayment provide an indispensable source of revenue, placing pressure on lending operations to be financially prudent. In addition, one of the most important factors for credit rating calculation is the bank’s refrain from calling on its callable capital. Therefore, the business side of the development bank’s operations lends its support to a more conservative agenda, relying heavily on the continued support of the donor members, the success of its investments, and the repayment of its loans.22 The IDB divides its borrowing members into four categories.23 These categories are based on the size of the country’s economy that corresponds to voting shares (these also determine the amount of yearly contributions members are expected to make). However, the share of the loans received by a borrowing member is not necessarily linearly related to its voting power (see Table 3.2).24 Although Group A received the highest percentage of resources, it is important to keep in mind that the countries in Group A are bigger and more populated. In addition, Table 3.2 demonstrates that the percentage of loans received by each group of countries corresponds to the groups voting shares: Group D receives a larger percentage of loans than Group C not because the countries in Group D are poorer, but because there are more than twice as many countries in Group D than there are in Group C. The chapters that follow compare lending by the banks activities taking 22

23

24

This institutional design is not unique to the IDB: It is based on the WB’s institutional design and was adopted by all RDBs. The AsDB classifies countries for the Asian Development Fund, but not for loans made by ordinary capital. Ordinary Capital funds of the IDB constitute, as of 2001, 93% of total disbursements. See Appendix 2 for voting power charts. As we shall see in Chapter 6, voting shares are not identical to the amount of contribution.

3.2 Comparing the RDBs

73

table 3.2 IDB borrowers by group % of Loans Voting Share1 1961–1993 Group Aa Group Bb Group Cc Group Dd 1 a b c

d

36.84 7.92 3.18 5.82

55 19 8 18

See Tussie (1995, p. 7). Members: Argentina, Brazil, Mexico, Venezuela. Members: Chile, Colombia, Peru. Members: Bahamas, Barbados, Costa Rica, Jamaica, Trinidad and Tobago, Uruguay. Members: Belize, Bolivia, the Dominican Republic, Ecuador, El Salvador, Guatemala, Guyana, Haiti, Honduras, Nicaragua, Panama, Paraguay, Surinam.

into account loan recipients’ size (population and economy). The IDB’s unique design – the division of borrowing members into groups – creates a hierarchical structure where the wealthier borrowing members hold the majority of the bank’s voting shares (Group A).25 The Articles of Agreement of the IDB spells out some of its founding members voting shares, making it difficult to change their configuration (such as the United States’s shares, and Argentina and Brazil being allocated equal shares).26 However, as will later be explained, shares and voting power are not necessarily directly associated with actual “political” power within the institution. Since bureaucracy can have a substantial effect on policy, the structure of the institution – its hierarchy, departments, etc., contribute to policy outcomes. Therefore, a grasp of the way in which the RDBs are organized is an important stepping-stone to placing them on the Bank/Development Agency continuum. The IDB’s organizational structure highlights the corporate-style framework that the bank follows (see supplemental website for diagram). The Board of Directors, government appointees, regularly represents the 25

26

Group D, for example, includes more borrowing countries, but their aggregate vote shares are smaller than that of Group A. This is the case for many international organizations, where their founding members acquired a role that reflected their relative power at the time the institution was created, and over the years, these structures often attract criticism as they become outdated (e.g., the UN Security Council).

74

Origins, Politics, and Structure of RDBs

Board of Governors, usually high-ranking government officials from member governments, including both the borrowing countries and the donors. The fourteen executive directors, representing forty-eight countries, meet weekly.27 They manage the bank’s daily activities and report back to the Governors, while the latter meet annually to review and assess the bank’s performance.28 More specifically, The IDB’s 14-member Board of Executive Directors is responsible for conducting Bank operations. Board members are elected or appointed to three-year terms by the Governors of the bank. The Board also includes 14 Alternates who have full power to act when their principals are absent. The Executive Directors for the United States and Canada represent their own countries, but all other Executive Directors represent groups of countries. The Board establishes the institution’s operational policies, approves projects proposed by the President of the Bank, sets interest rates for bank loans, authorizes borrowings in the capital markets, and approves the institution’s administrative budget. (http://www.iadb.org/info/bfin gles/bank5.htm)

Looking at this structure, it is clear that the Board tries to keep close supervision on the operations of the bank. It is worth emphasizing that the United States has a permanent representative on the Board of Executive Directors and thus can be closely involved in the daily decision-making (other countries are divided into groupings in which the position of representative to the Board of Executive Directors is rotated). The President reports to the Board, as in other traditional corporate structures, and it should be noted that the Office of Evaluations is subject to the Board of Directors and not to the office of the President, further indicating the “hands on” approach of the Board. The IDB’s organizational chart illustrates that although the bank is a public, political institution, it adopted elements of a business structure that may resemble a private institution. The staff and management of the 27

28

The Board Directors are elected by the Board of Governors for a two-year term. “The Board of Directors usually meets once a week and, among other duties, is responsible for approving loan and guarantee proposals, policies, country strategies, the administrative budget, setting interest rates, and making decisions on borrowings and other financial matters” (http://www.iadb.org/en/about-us/board-of-executive-directors,6124 .html). Chapter 6 elaborates on the significance of voting shares, their role, and the interest of member states that is associated with RDB membership. “The highest authority of the Bank is vested in the Board of Governors, composed of one Governor and an Alternate Governor appointed by each member country. Governors are usually Ministers of Finance, Presidents of Central Banks or other officials. The Board holds an annual meeting to review the Bank’s operations and to make major policy decisions. The Board of Governors delegates many of its powers to the Board of Executive Directors” (http://www.iadb.org/info/bfingles/bank4.htm).

3.2 Comparing the RDBs

75

institution, including its president (who are supposedly divorced from any country allegiance and are acting as professional employees), are subject to the Board, a political entity charged with voting power that could be driven by partisan political considerations. The question to consider, then, is the nature of the power held by Board members and the extent to which the staff of the institution has control over policy.29 Answering this question is paramount to the analysis of the IDB’s operational procedures, which is essential to the discovery of the nature of the bank (and the other RDBs) and where it stands on the bank/development agency continuum. As for its geographical presence, as mentioned, the IDB’s headquarters are in Washington, DC, and it has regional offices in all its borrowing member countries: “The Country Offices help to identify new projects, supervise and manage loans that have been approved, and monitor project execution.”30 The presence of local offices in member countries facilitates delegating responsibilities through the diffusion of staff tasks and location. It further illustrates the complex nature of information-gathering and the need to disseminate responsibilities to those closer to the sources of the loan-proposals and implementation. The IDB also has offices in Europe (located in Paris until 2012, and currently in Madrid) and Tokyo that assist its relationship with European and Asian donors. These offices underscore the importance the bank attributes to relations with its nonregional members, ensuring physical presence close to central donors. The way in which RDBs are structured, a proximity to nonborrowing members that are imperative for the institutions’ operations, is highly important. The IDB’s offices in Europe and in Tokyo have a distinct purpose of, essentially, fundraising.31 With limited resources, aid and development institutions compete to raise money from donor countries. Initially locating its European representation in Paris, France was not a random choice. Since 1976 the IDB’s office there has shared a building with the World Bank’s European office, in close 29 30 31

See discussion on delegation in Chapter 2. http://www.iadb.org/info/bfingles/bank8.htm In June 2012 the IDB’s Paris office was moved to Madrid, Spain. It still functions in the same way, with comparable staff. The move to Madrid is a natural extension to the 1974 Declaration of Madrid, where European countries were initially admitted to the IDB. More recently, Spain has become the IDB’s single largest contributor to its trust funds, enhancing its relationship with the development bank. Further, for reasons that go beyond shared culture, language, and history, it seems the 2008 financial crisis affecting Europe has made it hard for the IDB to continue operating in Paris and so this move is also a product of cost-cutting measures.

76

Origins, Politics, and Structure of RDBs

proximity to the OECD headquarters, and to the Paris Club.32 Carlos Jarque, the director of the IDB’s office in Europe and Mexico’s former Development Minister, perceives his office’s role as a liaison between donors states trying to meet ODA benchmarks, and the banks’ attempts to offer a fruitful investing environment that serves donor’s interests beyond their ODA contribution.33 The need for – and use of – such a strategy is not unique to the IDB. The AsDB, for example, has a European Representative office in Frankfurt, Germany (since 1996), and a North American Representative office in Washington, DC (since 1995). Although this aspect of the RDBs’ structure is explained in detail in Chapter 6, it is notable that their presence in close proximity to donor countries points to the importance of nonborrowing members, regardless of their voting shares. .

3.2.2

The African Development Bank

In the first published independent academic research on the African Development Bank, Karen Mingst (1990) concludes that it is the least political of the multilateral development institutions (including the IDB and AsDB). She comes to this conclusion after observing that the AfDB is the least prone to hegemonic pressures. However, hegemonic influences and interference are only one aspect of the political nature of RDBs. The political nature of the AfDB lies largely within its institutional apparatus: although difficult to measure, the AfDB is commonly criticized for its nonprofessional practices regarding staff and expert hiring, as well as its nepotism and corruption. Nonetheless, as Mingst points out, being less “political” (she defines politicization as the internal workings of the institution) does not mean being more effective.34 Having a more professional, less corrupt institution, she points out, does not automatically guarantee a positive outcome. She then asserts that despite its shortcomings, the AfDB’s institutional structure is conducive to a positive development strategy. Analyzing the institutional structure of the AfDB (and the

32

33 34

This points to shared interests between the IDB and the US – perhaps also US/donor influence in keeping it close to the WB, but also the bank’s strategy of maintaining geographical presence in the location where discussion and decisions about development aid take place. The Paris Club, a forum for negotiations of debt reduction, does not exist as a formal institution, but the French Treasury facilitates its operations. Interview at IDB Paris Office, April 2012. Mingst, 1990, p. 182.

3.2 Comparing the RDBs

77

other RDBs) offers additional information that can help highlight the method by which the bank works and its possible shortcomings. Over sixty percent of the AfDB’s voting shares are held by African countries, while no nonregional member holds more than 6 percent of the votes individually and no country possesses the ability to veto policy.35 Independence from nonregional hegemons allows borrowing countries to make decisions about the loans they receive, while nonregional donors are seen mostly as financiers.36 However, politicization of a bank’s policies can occur also when hegemons are not closely influencing decisionmaking. Moreover, the goals of alleviating poverty and fostering development are not necessarily best served by an RDB that does not allow hegemonic involvement: since nonregional member countries are limited, the AfDB is more heavily dependent on local contributions. Its budget is therefore significantly smaller than that of the other RDBs. The AfDB’s volatile relationship with donors affects its credit rating; although it is now AAA rated, in 1995 it was downgraded to AA by Standard & Poor’s, causing credit agencies to be wary of the bank’s borrowers’ creditworthiness.37 In fact, a Standard & Poor’s Supranationals Report (2009) suggests that the AfDB’s stable ratings and positive outlook are a product of its preferential treatment in sovereign and sovereignguaranteed loans. As a result, the AfDB benefits from increased global attention to Africa and various debt-relief funds, including those targeted at the highly indebted poor countries (HIPCs). These funds prioritize servicing the AfDB’s loans, making it a worthwhile investment despite the risk involved in making loans to many African countries. These incentives encourage the AfDB to follow a financially aggressive policy that safeguards its credit worthiness rather than creating an agenda that is guided by development concerns. Given the volatile political, economic, and social conditions in Africa, an IFIAC report (2000) suggested that the AfDB should be dissolved and that the WB should become the central lender to Africa. Arguably, lending institutions in Africa that position their policy strategically so that they

35

36

37

As, for example, the US has veto power in the IDB, Japan and the US clearly have a distinctive advantage in the AsDB, and the major (developed) donors of the EBRD are the principal decision-makers. See Appendix 2, Voting Powers of Member Countries (IDB, AsDB, AfDB, EBRD, and WB). In the 2009 Supranationals Report by Standard & Poor’s, the agency explains that the AAA rating reflects the AfDB’s commitment to making loans to credit-worthy borrowing members.

78

Origins, Politics, and Structure of RDBs

can be compensated by debt-relief initiatives to the region may contribute to inefficiency of donor resources – where debt relief and emergency assistance are servicing loans made by IFIs. The AfDB does not have much choice in this case if it seeks to maintain a stable, high rating because it does not have a central nonregional hegemon that can serve as a guarantor. It seems that hegemonic support may make it easier to pursue development-driven, financially risky lending because of the safety net provided by the hegemon as well as the reliable high credit rating an institution receives when a pivotal member state has a strong and stable economy. To that end, although Nigeria has been the AfDB’s largest shareholder, its shares have fallen in the last decade from 10 percent to 9.669 percent; while a decade ago Egypt was the second largest shareholder (still, with under 6 percent of the votes), it now trails behind the United States. Furthermore, South Africa’s original low profile role at the AfDB (a product of the Apartheid era) has been replaced by increased involvement, placing it in with the top five shareholders in recent years. It is noteworthy that increasing the vote shares of nonborrowing members of the AfDB is an indication that external involvement must be reflected in shares. Founded in 1964 and lending by 1966, the AfDB’s fifty-three regional members (borrowing) and twenty-four nonregional members (donor) are expected to follow the institution’s mandate: “. . . to contribute to the economic development and social progress and reduce the poverty level of its 53 regional members” (Standard & Poor’s, 1999, p. 17). The bank’s total capital in 2000 was US $35.5 billion.38 In its 1999 report on Supranationals, Standard & Poor’s concluded that the AfDB’s Board has a recent history of politicization that impaired the bank’s management in the late 1980s and early 1990s. Standard & Poor’s definition of “politicization” stands in contrast to the one suggested by Mingst (1990). While the latter refers to the involvement of nonregional members (hegemons) and the way in which they seek to influence the bank, the former is focused on regional members’ influence on the bank’s management. Specifically, the AfDB’s headquarters and management located in Africa (in the Ivory Coast at that time) with regional members holding majority shares on the Board, coincided with growing regional debt volumes and economic uncertainty. The frequent stalemates at the management level, then, were blamed on the governing

38

AfDB Annual Report, 1999; World Bank Annual Report, 1999; IFIAC Report, 2000.

3.2 Comparing the RDBs

79

apparatus and the central shareholders (also borrowers), who were unable to resolve financial and budgetary issues.39 In addition, lengthy negotiations on the Fifth Capital increase indicated that member governments’ willingness to support the bank is still lower than in many other AAA rated multilateral lending institutions.40 Such observations made by rating agencies highlight the significance of shares held by nonborrowing members in determining the bank’s rating. This demonstrates the conflict between independence from donor control and the need for a hegemon in safeguarding credit worthiness. The AfDB’s institutional design is similar to the IDB’s in its corporate model:41 the President reports to the Board of Directors, that represents the members’ Governors. However, unlike the IDB, the AfDB does not have an evaluation office that reports to the Board of Directors. This can explain some of the problems faced by the bank in assessing projects and communicating problems and shortcomings of loan programs. One plausible explanation for this difference between the two banks (that can be concluded from the founding documents of the institutions) is that while agreeing to a dominant US presence at the IDB, including veto power, Latin American countries prioritized an evaluation mechanism that could, ostensibly, “check” the hegemon. At the same time, in an institution where borrowers had much more “control,” an evaluation office may not have been at the top of the agenda. Notably, over the years and with allowing additional donors since the 1980s, the AfDB has had to find ways to provide more transparency in the face of criticism. Moreover, not reflected in the diagram is the AfDB’s experience with many changes to the makeup of its Board of Governors over the years – much more than in any of the other RDBs. Although the AfDB began its operations with a deep-rooted belief in African control, difficulties in meeting the paid-in obligations forced the bank to increase its external 39

40

41

Overall, if correct, the conclusions reached by Standard & Poor’s (and other rating agencies), furthers the argument that that limited hegemonic involvement at the AfDB brings a financial price for the institution. And while “politicization” that is engendered in hegemonic or donor control is tolerated by rating agencies, the political motives associated with borrowing countries’ involvement raise concern (and may lead to lower ratings, with financial and development implications). Negotiations for the Fifth Capital increase lasted eleven years, while those for previous capital replenishments were typically about five. In contrast, the IDB’s longest capital increase negotiation period lasted seven years (1983–1990), while the negotiations for the six increases preceding it lasted between two and six years. For more detail see www.afdb.org, African Development Report, 1999, 2000; see supplemental website for chart.

80

Origins, Politics, and Structure of RDBs

borrowing in 1973. In 1981, the bank increased its economic base by allowing non-African states to become members. Thus, the AfDB has experienced shortages in resources with greater frequency and magnitude than the other RDBs. The importance of the AfDB as a lender in Africa, similar to that of the other RDBs, is its disbursement of loans with longer maturities, as well as a commitment to lend to projects that would not otherwise be funded. However, interest rates on its loans are not much different than the comparable market rates (although the interest rates vary and are not always disclosed, they can be as high as 10 percent).42 Presumably, capital would not be available otherwise, and then loans that carry with them a market rate interest could be considered a concession of sorts. Thus, if the RDB makes loans available only to borrowers who would not qualify for loans from private institutions, then even at market interest rate, the loans would be unique opportunities for the borrowers.43 The problems of the continent of Africa are exceptional to the AfDB. First, there are structural issues concerning transportation: insufficient roads and inconvenient transportation prevent observers from conducting evaluations and visiting sites of projects.44 Next, the AfDB lacks the web of country offices possessed by its counterparts. Finally, the AfDB has been the subject of criticism centered on bad practices, corruption, and insufficient expertise much more than any other IFI: “. . . good work has been the exception rather than the rule, while even ‘successful’ projects could have been done better. Inadequate project preparation was the most fundamental shortcoming, while the absence of staff involvement at this stage or in subsequent supervision was particularly striking” (English and Mule, 1995, p. 4). The AfDB separates two windows of lending. Eleven borrowers are eligible for its regular lending (OCR), and the rest can only borrow from 42

43

44

“Applying sound banking principles, the bank disburses funds at market rate with maturities ranging between twelve and twenty years. The financing is meant to cover the foreign exchange expenditures incurred on a project, generally less than one-half of the project’s cost” (Mingst, 1990, p. 11). One of the questions examined in Chapters 5 and 6 is whether private lending institutions are, in fact, not likely to be an option. I do not address here the question of whether making such loans available is in fact good for these developing countries, or, as some claim, detrimental as they contribute to the debt problem. In the conclusion, I do include some discussion of this issue since some assumptions can be deduced from the findings of this book. To get from some capitals of Africa to others, one has to go through Europe, traveling for many hours at a significant cost.

3.2 Comparing the RDBs

81

table 3.3 AfDB country classification Classification AfDB (nonconcessional only)

2009

Angola, Cameroon, Cape Verde, Ghana, Kenya, Madagascar, Mali, Mozambique, Nigeria, Senegal, Tanzania, Uganda, Zambia, Zimbabwe Combination Benin, Burkina Faso, Ethiopia, Lesotho, Malawi, Mauritania, Niger, Sierra Leone ADF (concessional) Burundi, Central African ADF (concesRepublic, Chad, sional) only Comoros, Democratic Republic of Congo, Republic of Congo, Ivory Coast, Djibouti, Eritrea, Gambia, Guinea, Guinea Bissau, Liberia, Rwanda, Sao Tome & Principe, Somalia, Sudan, Togo

1995 Algeria, Botswana, Gabon, Libya, Mauritius, Morocco, Namibia, Seychelles, South Africa, Swaziland, Tunisia

Angola, Benin, Burkina Faso, Burundi, Cameroon, Cape Verde, Central African Republic, Chad, Comoros, Congo, Congo Democratic Republic, Djibouti, Equatorial Guinea, Eritrea, Ethiopia, Gambia, Ghana, Guinea, Guinea Bissau, Ivory Coast Kenya, Lesotho, Liberia, Madagascar, Malawi, Mali, Mauritania, Mozambique, Niger, Rwanda, Sao Tome & Principe, Senegal, Sierra Leone, Somalia, Sudan, Tanzania, Togo, Uganda, Zambia

the “soft loan” (concessional) window, the African Development Fund (ADF). However, since concessional funds were cut substantially, this division is not followed with orthodoxy (see Table 3.3). This distinction means that the very poor countries of Africa can receive loans only from a more limited source. The vast majority of the bank’s loans come from the regular lending window.45 This lending 45

For most of the period covered in this study, the ADF comprised a small percentage of AfDB lending (roughly 10–15 percent). More recently, its funds have been increased and it now comprises about 30 percent of the AfDB Group lending.

82

Origins, Politics, and Structure of RDBs

window is better endowed and has a better “reputation.” Thus, the very poor countries that cannot borrow from this source are at a disadvantage. In the 1990s, the concessional fund came under criticism that nearly led to its closure: donor countries were reassessing their endowments to the ADF and analyzing whether it is actually effective. While the ADF still operates, fewer countries are now eligible to borrow from it.

3.2.3

The Asian Development Bank

The AsDB was founded in 1966 to promote the economic and social development of its member states in the Asia-Pacific region through loans, investments, and guarantees. The AsDB has sixteen nonregional (donor) members, and together with Japan, Australia, and New Zealand, they contribute most of the bank’s resources and hold about 55 percent of the voting shares (see Table 3.4).46 Of the three older generation RDBs, this is the only one in which developed members hold majority shares. Financially, the bank’s purpose is “to provide loans and equity investments that promote the economic and social advancements of its member states and to encourage public and private sector investment for development purposes” (Standard & table 3.4 AsDB member countries (2002) Regional (borrowing) members: 41

Non Regional (donor) members: 16

46

Afghanistan, Azerbaijan, Bangladesh, Bhutan, Cambodia, China, Cook Islands, Fiji, Hong Kong, India, Indonesia, Kiribati, Korea, Kyrgyzstan, Kazakhstan, Lao PDR, Malaysia, Maldives, Marshall Islands, Micronesia, Mongolia, Myanmar, Nauru, Nepal, Pakistan, Papua New Guinea, Philippines, Singapore, Solomon Islands, Sri Lanka, Tajikistan, Taipei China, Thailand, Tonga, Turkmenistan, Tuvalu, Uzbekistan, Vanuatu, Vietnam, Western Samoa Austria, Belgium, Canada, Denmark, Finland, France, Germany, Italy, Netherlands, Norway, Spain, Sweden, Switzerland, Turkey, UK, US

It is noteworthy, that only recently have the shares shifted: with additional countries joining both the donor and borrowers’ ranks of the AsDB, borrowing members now hold just over 50 percent of the shares (see Table 3.1).

3.2 Comparing the RDBs

83

Poor’s, 1999, p. 19). The AsDB has been AAA rated since 1971 with total approved financing in 2012 measured at US $21.57 billion.47 With its good financial credentials and solid portfolio, the AsDB was instrumental in the efforts to restructure Asian economies in the aftermath of the 1997 financial crisis. It has been a leading conduit for development, maintaining a noticeably strong presence in the region: the AsDB has local offices in all of its borrowing member countries. Although the headquarters of the bank are in Manila, Philippines, the AsDB is under strong Japanese (hegemonic) influence. In fact, as Krasner (1981) notes, the AsDB is heavily influenced by both the United States and Japan, the central shareholders in the bank. Interviews with AsDB officials reveal that, at times, strong tensions between the United States and Japan dictates policy. While the United States and Japan have many shared interests, they both vie for influence in Asia and are often at odds when it comes to policies concerning economic development. If we keep in mind that major donors may (covertly) expect IFIs to advance their interest (and economies), then it would not be surprising that during much of this time period, when the United States and Japan are global economic competitors, tensions on economic development policy may guide their preferences. Disputes between the key shareholders are usually settled before decisions are made about loans and prior to voting by the Board of Directors, notes a former General Council of the AsDB (interview, 2001).48 The origin of the AsDB provides an explanation for the stronghold of developed countries on the bank: unlike the IDB, where borrowing members pressed for creating a development bank, or the AfDB, where African countries insisted on controlling the bank’s policies, the AsDB was envisioned and promoted by the United States and Japan. As noted, political interests (the Vietnam war) were the catalysts for the AsDB’s foundation. In particular, the United States was seeking both Japanese involvement and a multinationalization of aid to Southeast Asia.49 Thus, from its very inception, the AsDB has been heavily influenced by donor member countries. The AsDB’s borrowers do not have a blocking vote, although they 47

48

49

At the end of 2012, the AsDB’s subscribed capital increased to US $163.1 billion (AsDB Annual Report, 2012). The high credit ratings reflect the confidence that credit agencies have in the donor base of the bank, and these donors’ decisions to increase capital in the last capital replenishment. See also Babb (2009) for a discussion about how any conflict and disagreement are resolved before they are brought to a vote, thus leaving no paper trail of disputes. See Krasner, 1981, p. 317.

Origins, Politics, and Structure of RDBs

84

are able to sometimes advance their interests by taking advantage of the rivalry among developed members (particularly Japan and the United States).50 Evidently, the RDBs that were created with the vision, support, and ambition of nonborrowing countries institutionalized these members’ powers. The AsDB is designed in a similar fashion to the AfDB and IDB.51 The Board of Governors and Board of Directors oversee the operations of the institution, and the President directly supervises many of the oversight and evaluation offices. The AsDB’s president presides over departments ranging from operations (divided into regions and sectors) to finance and strategy, like the presidents of the AfDB and IDB. Similar to the IDB, the AsDB has an evaluation department that is directly connected to the office of the president. While this allows the highest ranking official of the bank direct access to and responsibility for project/program evaluation, it also institutionalized an internal evaluation system rather than an independent one that reports to Board Members.

3.2.4

The European Bank for Reconstruction and Development

The EBRD was founded in 1991.52 Its mandate, “to foster the development of market economies by promoting private and entrepreneurial initiative within recipient Central and East European countries” (Standard & Poor’s, 1999, p. 27), is quite different than that of its counterparts, and includes a political agenda of advancing market economies, while making no reference to the alleviation of poverty. Of the EBRD’s sixty members, twenty-six are regional (borrowing) and thirtyfour nonregional (donor). The EBRD is the only RDB that has more donor than borrowing member countries on its Board (like the WB). It has been consistently AAA rated by Standard & Poor’s since its inception, and its authorized capital in 2013 was US $41.28 billion. This high credit rating reflects the EBRD’s careful financial policy. For example, according to Standard & Poor’s (1999), despite the losses suffered following the

50

51 52

Japan, Australia, and New Zealand are counted as regional members, giving them a share of the regional voting pool even though they do not borrow from the bank. For organizational chart see supplementary website. “I am sure you will all agree that the most important event for Europe, perhaps for the world, since the Second World War is what is happening in Eastern Europe. . . What can Europe do? So much more! Why not set up a bank for Europe.” François Mitterrand, at the European Parliament, October 25, 1989.

3.2 Comparing the RDBs

85

Russian financial collapse (1998), the EBRD’s credit worthiness is protected by conservative financial policies.53 The EBRD is modeled differently than the three other multilateral development banks in this study. On the one hand, its conservative financial policy is rarely dependent on government guarantees and it boasts a market approach that favors the private sector. However, its agenda fosters overt political goals that stand in stark contrast to the philosophy of the older RDBs that posit that poverty alleviation can be separated from political institutions. The EBRD’s political agenda is evident in its own words: The European Bank for Reconstruction and Development was established in 1991 when communism was crumbling in Central and Eastern Europe and ex-Soviet countries needed support to nurture a new private sector in a democratic environment. Today the EBRD uses the tools of investment to help build market economies and democracies in 27 countries from central Europe to central Asia.” (http://www.ebrd.org/about/index.htm)

The EBRD strategically supports projects that are compatible with processes of democratization and open market policies. Although it is more financially conservative and prides itself on its banking-business professionalism, the bank overtly takes a clear political stand in its goals and criteria. EBRD membership (as of 2009) Borrowing (30) Albania, Armenia, Azerbaijan, Belarus, Bosnia and Herzegovina, Bulgaria, Croatia, Czech Republic, Estonia, Georgia, Hungary, Kazakhstan, Kyrgyz Republic, Latvia, Lithuania, Moldova, Mongolia, Montenegro, Poland, Romania, Russian Federation, Serbia, Slovak Republic, Slovenia, Tajikistan, Ukraine, Uzbekistan Donor (34) Australia, Austria, Belgium, Canada, Cyprus, Denmark, Egypt, Finland, France, Germany, Greece, Iceland, Ireland, Israel, Italy, Japan, Republic of Korea, Liechtenstein, Luxembourg, Malta, Mexico, Morocco, Netherlands, New Zealand, Norway, Portugal, Spain, Sweden, Switzerland, Turkey, UK, US, EIB, EU

53

The EBRD stresses its “Prudent Capital Adequacy Policies – economic capital policy which excludes all callable capital, uses a 99.99% confidence interval to underpin the triple A rating” (EBRD, Investment of Choice, October 2013).

86

Origins, Politics, and Structure of RDBs

Created three decades after the establishment of the IDB, AfDB, and AsDB, the EBRD was championed by its nonborrowing member countries, particularly those in Western Europe. Following the collapse of the USSR, there was a concern about the economic and political policies that would be adopted by Eastern and Central European counties. Envisioned by Jaques Attali (special advisor to French President Francois Mitterrand), the EBRD was given the green light only when the United States finally agreed to support its establishment. With its headquarters in London, the EBRD was meant to be mostly a European institution. However, the United States has had substantial influence on it, mostly in behind-thescenes agenda-setting while Western European countries remain invested in the bank and provide a balance to American power. Privatization and democratization in Eastern and Central Europe are important policy goals for both the United States and Western European countries: the United States has a strong interest in seeing political and economic reform in Russia, while Western Europe is concerned with the expansion of the EU.54 Thus, donors dominate decision-making at the EBRD: both on the Board and in the positions of power held within the bank, most of which are occupied by nationals of Western countries.55 The EBRD not only differs from the other RDBs in its open political approach, but also in its lending policies: there is more microlending56 and loans are not required to be guaranteed by governments. In fact, lending to governmentrelated agencies is not encouraged.57 The EBRD appears to be structurally similar to the other RDBs: the Board of Governors, comprised of high-ranking officials from the member 54

55 56

57

And more recently in the wake of revolutions in the Middle East and North Africa, President Obama suggested a refocusing of the EBRD: “And we will work with the allies to refocus the European Bank for Reconstruction and Development so that it provides the same support for democratic transitions and economic modernization in the Middle East and North Africa as it has in Europe” (The White House May 19, 2011). This statement is evidence of strong US involvement – and influence – at the EBRD. A large percent of staff members are nationals of Easter European borrowing members. Microlending is a practice that is becoming more popular especially among smaller and independent development lenders (such as the Grameen Bank). These are usually substantially smaller loans to private borrowers not requiring government guarantees. As Article 11, section 3(i) states: “Not more than forty (40) per cent of the amount of the Bank’s total committed loans, guarantees and equity investments, without prejudice to its other operations referred to in this Article, shall be provided to the state sector. Such percentage limit shall apply initially over a two (2) year period, from the date of commencement of the Bank’s operations, taking one year with another, and thereafter in respect of each subsequent financial year.”

3.3 RDBs’ Capital Structure

87

countries delegates most powers to its representatives on the Board of Directors who preside over the ongoing activities of the bank. The president, elected by the Board of Governors and supervised by the directors, manages the bank. At this point, the design of the institution begins to differ. The various departments are divided into banking, finance, and risk management, to name a few.58 It resembles a “bank” much more than a development agency (and more than the other RDBs); symbolically, this is reflected in the location of headquarters – the heart of London’s financial district. These tangible details are important because they capture the geographical and human influences on the RDBs’ conduct. When the structure and goals of the banks are similar, it may be the cultural, less tangible factors that constitute some of the stark difference between the banks. For example since the shareholders mostly delegate day-to-day decisionmaking, the composition of the staff and the origin of the president may have a large impact on the way the institution functions, since they are responsible for much of the agenda and preference setting.

3.3

rdb s’ capital structure

Based on the institutional design of the World Bank, the RDBs are a unique form of international institution. In contrast to the UN and its agencies, these banks are structured as corporations, with member countries as shareholders. The voting power corresponds to the relative contribution of the members (see Appendix 2). But unlike at the WB, as discussed in this chapter, regional member countries of the RDBs are granted a minimum number of votes as well as secured administrative positions so they can have greater impact on the governing and decision-making of the respective banks. Consequently, though dependent on donor countries’ contributions, borrowing countries are potentially able to partake in shaping RDBs’ policies. Discussing hegemonic influence in each of these banks, Krasner (1981) notes that the involvement of donor countries differs from one bank to another. When the political interests of the hegemon are long-term and general, as in the case of the IDB during the Cold War, the day-to-day activities are less affected by hegemonic influence – the bank staff and 58

The rest of the main department of the EBRD are: Human Resources, Office of the Secretary General, Office of the General Counsel, Office of the Chief Economist, Internal Audit, Office of the Chief Compliance Officer, and Communications. See supplemental website for the EBRD’s Organizational Chart.

88

Origins, Politics, and Structure of RDBs

management are delegated the authority to make many decisions, while the principle donors interfere only to the extent that decisions are not consistent with their long-term interests (for example, Cuba’s exclusion from IDB membership is an American political decision). The AsDB, according to Krasner, is the most closely controlled by developed countries. This is not only because borrowing members have less than fifty percent of the votes (a smaller percentage than in IDB and AfDB), but also because having two hegemons involved – Japan and the United States – does not leave much room for developing countries’ engagement. The AfDB, being an African initiative, was initially relatively free from donor oversight. But, to avoid risking losing its capital commitments, the AfDB increased its nonregional membership in order to secure funding. By doing so it was forced to allow some donor countries more authority. The EBRD clearly emphasizes control by donors who provide the capital. First, the ratio of donors to borrowers is highly in favor of donors (34 donor countries; 26 borrowing, until 2003). Second, at the outset, when there were eight borrowing countries,59 of the one million shares, the borrowing members held only 134,500. And, unlike the other RDBs, EBRD shares are directly proportionate to contribution, with no minimum vote assigned to borrowing members. Figure 3.1 describes the capital structure of the RDBs.

Callable capital Members’ liquid contributions RDB

Market borrowing and investment, loan repayments

Concessional (soft loans) Non-concessional (hard loans) Eligible borrowers (HIPCs)

Most borrowers

figure 3.1 The capital structure of RDBs

59

Bulgaria, Czechoslovakia, German Democratic Republic, Hungary, Poland, Romania, Union of Soviet Socialist Republics, and Yugoslavia.

3.3 RDBs’ Capital Structure

89

This illustration of the sources of RDBs’ income and the way in which they manage their capital, demonstrates the flow of finances in and out of the banks.60 Most of the RDBs liquid capital comes from borrowing, investment, and loan repayments. Donor members pay in a small amount of their commitments (10%–20%),61 and the callable capital provides an “insurance” that the RDBs will remain highly rated and thus attract private investments and higher yields. The output of the banks (apart from the EBRD62) is divided into concessional and nonconcessional lending avenues. The nonconcessional, or “hard-loans” (also known as OCR – Ordinary Capital Resources) have varying degrees of interest rates and maturities.63 Each bank has different policies, but they are all similar in their resemblance to commercial banks. The interest rates on their OCR loans are usually similar or slightly lower than private banks’ interest rates. The maturities are, for the most part, longer than those of commercial banks, at times up to twenty years. In addition, all the RDBs require “sound banking” procedures in their Articles of Agreement, measures that many times resemble those of private banks. What differentiates the RDBs from private banks is their stated commitment to finance projects that would not qualify for loans by a commercial bank.64 In addition, the RDBs are an attractive investment for donors because the majority of their loans are government guaranteed. These sovereign loans are almost “fool proof” as private investors assume that in case of a crisis, financial assistance (bailout) will be made available, assuring that the borrowers will not default on their loans. The concessional, or “soft loan” window, is one that is unique to these public agencies and sets them further apart from private banks. This loan option is very close to grantmaking and helps the banks work in the poorest countries. However, eligibility is not simple: access to these 60

61 62

63

64

Note that the RDBs are nonprofit institutions, thereby having a strong incentive to use all their funds in every budget year. All members of each bank pay the same percentage of their commitments. Although it does not have a “soft-loan” window of lending, the EBRD is the most secretive about the rates of its loans. Officials claim that rates are determined according to various criteria, including need. Rates are thus never made public, and even the bank’s staff is often not privy to this information. It is important to note that in most cases, members earmark the capital they commit to either the concessional or nonconcessional lending window. The AsDB, for example, stipulates that: “The OCR are lent at rates of interest determined by the market. Until 1986, these were set on the basis of the average cost of borrowing by the Bank during the preceding twelve months plus a margin of 0.5 percent” (Kappagoda, 1995, p. 18). How do we know that this is the case? This is problematic since there is no requirement that project proposals first be turned down by private investors before the RDBs approve the loans.

90

Origins, Politics, and Structure of RDBs

funds is mostly limited to HIPCs. These special funds are unique in their very low interest rates and very long-term maturities. But the concessional funds are substantially smaller than the regular funds, only provided to countries with very low GDPs, and are more susceptible to cuts. Moreover, this lending window would not exist without the OCR. Further, the concessional lending window receives its own funding separately from the capital contributions to the OCR.

3.4

geography, states, and people

The location of the RDBs’ headquarters symbolizes the political and financial influences that are a product of geographical proximity to government agencies, commercial banks, and other international institutions (see Table 3.5). Where the banks are located also impacts the composition of staff: headquarters placed in a borrowing member country ensure that a large percentage of the bank’s staff comes from developing countries; while headquarters at a donor state result in a different makeup of staff as well as a tangible difference in atmosphere. Lastly, while the location and culture of the institution affect its conduct in a way that is difficult to quantify, the voting shares held by member countries reflect numerically the overt balance of power within the RDBs’, and potentially has direct impact on policies, strategies, and appearance. 3.4.1

Headquarters and Staff

American voting shares and de facto veto power contribute to the appearance that the IDB is significantly influenced by the United States. Some argue that its Washington, DC, headquarters are evidence of American hegemony. This degree of US influence may be mitigated by the fact that the bank’s staff includes many Latin American nationals and its president is always from a borrowing member country. Despite the strong hegemonic influence of the United States and Japan, the AsDB’s headquarters are located in the Philippines. And while the staff of the bank is comprised largely of residents of borrowing member countries, the president of the AsDB is always Japanese, and its vice president is always American. Its African headquarters reinforce the AfDB’s strong African roots and influence. Its staff is therefore comprised of mostly African nationals, and its president is always from the region as well. The EBRD’s British headquarters reflect the central role played by donor members. Its staff mixes nationals of donor and borrowing countries, but most have a background

table 3.5 Presidents, staff, and shareholders of the RDBs IDB President Staff Shareholders (voting power)

Location

65

Always Latin American (VP American) Mixed Latin American and American 50.005% Latin American; 49.995% non-regional (30% American) N. America – Washington, D.C., US

AfDB

AsDB

Always African

Always Japanese (VP American) Overwhelmingly African Mostly Asian 60% African; 40% nonregional

65.515% regional; 34.485% nonregional65

Africa – Abdijan, Ivory Coast (temporarily relocated to Tunis in 2003)

Asia – Manila, Philippines

EBRD Western European: Three French, one German. Mostly Western European and N. American. 12.35% countries of operation; 87.65% regional and nonregionals (donors) Western Europe – London, UK

The “regional” votes include those of Australia, Japan, and New Zealand (their combined voting power is: 19.696%). Thus, the real voting power of the nonindustrialized members of the Bank is: 45.819%.

92

Origins, Politics, and Structure of RDBs

in banking. The president of the EBRD is always from Western Europe, and every president thus far has been French, German, and most recently British. The IDB’s staff is recruited from its member countries through a competitive process that includes country quotas. While the US headquarters employs a large percentage of American citizens, with offices located in all borrowing member countries, many locals are involved in the IDB’s operations. This extensive apparatus gives the impression that the bank is “connected” to the region by making itself visible in Latin America and enabling locals to at least partake in the country office apparatus. Partly due to the location and partly because of the ideology present during its founding, the AfDB is predominantly run by African nationals. In addition to the nationality of its president, the AfDB, located in Ivory Coast from 1966 until 2003, employs an overwhelming number of African staff and has been struggling with resisting donor members’ influence since its inception. It is well documented that corruption has often taken over the bank’s decision-making, and that many of the staff are not qualified for the positions they hold.66 Lately, with the temporary move to Tunis (after violence in Abdijan made it necessary to relocate) and the election of a new president (Rwandan national) in 2005,67 the AfDB is in the process of administering reforms and attracting more outside experts to supervise changes to its administrative and policy-making apparatus.68 The AsDB, unlike the IDB, has no formal nationality quotas. Since its headquarters are in Manila (Philippines), a large percentage of its staff is comprised of Asian nationals. Moreover, the AsDB has a hiring system based on merit rather than nationality, perhaps because of the large pool of potential employees with requisite technical and professional skills available in the region. However, it is unclear how this system of hiring actually works.69 It should be noted that the major donors of the AsDB, Japan and the United States, have much less representation in all staff ranks than regional members. And professional staff who are nationals of regional members, such as India and the Philippines, outnumber their

66 67 68

69

See Culpeper, 1997; and Standard & Poor’s Supranationals, 2003, 2009. This president, Dr. Donald Kaberuka, was elected for a second five-year term in 2010. Anecdotally, many issues of “The Economist” in recent years have advertised job openings at the AfDB. See Appendix 3 for a table of the AsDB’s professional staff’s nationalities, 1981–1991.

3.4 Geography, States, and People

93

relative entitlements (the minimum required employment of nationals).70 However, since Japan’s population is much smaller than many of the regional members (e.g., China, India, Pakistan, Indonesia), a quota system that is based on population would likely marginalize it. Thus, although there is no clear documentation or explanation of this feature, it appears that not having nationality quotas likely favors Japan and probably assists it in maintaining a stronghold in the bank. The EBRD’s staff consists mostly of former bankers (who previously worked in the private sector). Its headquarters are located in the center of the business district of London – the City of London – surrounded by investment banks. The staff of the bank is selected very carefully based on qualifications and member country quotas. It is ironic that with its high level of professionalism in banking, this bank has come under severe attacks from policy makers and scholars during its short lifetime: it is precisely because of its banking professionalism that many claim that the existence of this institution is a waste of public money. Moreover, some claim that the highly professional “atmosphere” creates a disguise for what is really a very political institution.71 While the staff bankers and economists focus on sound financial decisions (rather than developmentdriven policies), the Board and managers make decisions about the bank’s overall policies. Thus, bankers and bureaucrats carry out the EBRD’s political goals such as democratization and liberalization, not development ideologues. It should be noted, however, that although in its first decade the bank struggled with its identity, during this last decade it seems to have evolved, mostly by making its operations more transparent and trying to differentiate itself from private institutions (for example, by taking a lead on funding projects that focus on sustainable energy).

3.4.2

Presidents

The presidents of the IDB are always from regional (Latin American) countries, as postulated by the bank’s Articles of Agreement. Elected by the Board of Governors to a five-year term, the president conducts the day-to-day business of the institution along with the executive vice president. By tradition, while the president is from a Latin American nation, the executive vice president is American. The president oversees 70 71

See Kappagoda, 1995, p. 38. This critique was reflected in debates over the pages of the Financial Times throughout the 1990s. It was especially acute during the presidency of Jacques Attali.

94

Origins, Politics, and Structure of RDBs

the meetings of the Board of Executive Directors, but has no vote except to break a tie. Further, since voting is infrequent, the president has the discretion to announce that there is consensus. In reality, then, presidents wield significant agenda-setting power that allows them to regulate discussions and decisions. It is important to note, however, that the United States plays a major role in the election of the IDB’s president. The IDB’s close ties with the US government has allowed presidents of the development bank to play a central role in setting the tone for the bank’s activities and reputation. But, on the other hand, America’s central role in appointing the president (the Board needs to approve the president and the United States has the highest percentage of shares) also has significant influence over the IDB’s president’s leadership and policy choices. This included casting a tie-breaking vote, where the president will be cautious not to alienate the United States. The only restriction on AfDB presidency is that she or he be African, while all members (a majority of which are African) receive equal vote in electing the president. Therefore, donor countries exert much less influence on the AfDB’s selection of president than their counterparts in the other RDBs. In addition, a nonformal understanding has developed that linguistic and geographic regions take turns in the presidency (e.g., Arabic, Anglophone, Francophone). But although the borrowing members make an independent decision about the president of the bank, some of the informal mechanisms, including the self-governed nature of the AfDB, can make corruption and nepotism potentially more widespread.72 The AsDB’s presidents, per its Articles of Agreement, must always be Japanese. Although the bank groups Japan with its regional member countries, it is a developed country and a regional hegemon (a donor rather than a borrower). Thus, unlike the IDB and AfDB where the presidents are traditionally from one of the borrowing member countries, the AsDB is an institution in which the donor members, especially Japan, are much more involved not just as board members, but also in key management positions. Although a permanent Japanese appointment for the presidency certainly does not help advance the idea of including borrowing members’ representatives among the ranks of top decisionmaking positions, proponents of the bank claim that the Japanese

72

This is not to say that corruption cannot be found in the other RDBs where the president’s selection is not democratic and more prone to donor/hegemonic pressures – this would mean that potential candidates and elected presidents are more likely to appease the powerful shareholders who were instrumental in their appointment.

3.4 Geography, States, and People

95

influence has been a major contribution to resource mobilization. This has been an asset to the AsDB in successfully maintaining high credit ratings and a positive reputation, as well as securing capital subscriptions. Lastly, in addition to Japan’s central role in the AsDB, the executive director of the Board is always American (set by the Articles of Agreement), and the United States (Treasury Department) makes a point of overseeing the policy of the AsDB since American presence in Asia is a political and financial priority for the United States. The EBRD’s presidents have all been from donor member countries.73 The first president, Jacques Attali (France), who was also the chief founder of the bank, held his position from April 1991 until June 1993. His term was characterized by mounting criticism over the use of funds and public money. His successor, Jacques de Larosiere, another French national and former Director of the Bank of France and General Director of the IMF, stayed in office for five years; from 1993 until 1998. He was followed by Horst Kohler (1998–2000), who cut his term short when he was elected to

73

In its Articles of Agreement, the EBRD defines the role of the president, explains the process by which a president is elected, and the president’s role. This is relatively similar to the other RDBs: Article 30: The President: 1.

2.

3.

4. 5.

6.

The Board of Governors, by a vote of a majority of the total number of Governors, representing not less than a majority of the total voting power of the members, shall elect a President of the Bank. The President, while holding office, shall not be a Governor or a Director of an Alternate for either. The term of office of the President shall be four (4) years. He or she may be reelected. He or she shall, however, cease to hold office when the Board of Governors so decides by an affirmative vote of not less than two-thirds of the Governors, representing not less than two-thirds of the total voting power of the members. If the office of the President for any reason becomes vacant, the Board of Governors, in accordance with the provisions of paragraph 1 of this Article, shall elect a successor for up to four (4) years. The President shall not vote, except that he or she may cast a deciding vote in case of an equal division. He or she may participate in meetings of the Board of Governors and shall chair the meetings of the Board of Directors. The President shall be the legal representative of the Bank. The President shall be chief of the staff of the Bank. He or she shall be responsible for the organization, appointment and dismissal of the officers and staff in accordance with regulations to be adopted by the Board of Directors. In appointing officers and staff, he or she shall, subject to the paramount importance of efficiency and technical competence, pay due regard to recruitment on a wide geographical basis among members of the Bank. The President shall conduct, under the direction of the Board of Directors, the current business of the Bank.

96

Origins, Politics, and Structure of RDBs

lead the IMF. The following president, Jean Lemierre (2000–2008), is a French former head of Internal Revenue Service and Director of France’s Treasury; he was followed by Thomas Mirow (2008–2012), a German national. Finally, the bank’s current president, Sir Suma Chakrabarti (2012-present), hails from British political ranks – most recently as head of the UK’s Department for International Development (DIFD). His surprise election in 2012 as the first British national to head the bank is a possible signal of EBRD’s future presence in developing regions outside of Europe.74 Although the EBRD maintains the most “bank-like” professional atmosphere, its presidents, while deeply involved in the world of finance, all come from a distinctly political background. In fact, presidents of all the RDBs often come from the political world. For example, the longest-serving president of the IDB, Enrique Iglesias, who held his position for seventeen years (1988–2005), was the foreign minister of Uruguay (1985–1988) and a revolutionary (against the nondemocratic regime) in the 1960s and 1970s. Although he has an academic background in economics and experience in banking, it was his dominant persona and diplomatic experience that made him an extremely popular and successful president.75 According to some IDB officials, it is the president who is the most influential figure in the bank; his ability to create ties and influence donors is crucial in advancing policies that are not necessarily “bankable.” Some IDB officials therefore claim that the IDB’s ability to shift to progressive, more development-oriented programs should be credited to Mr. Iglesias who was able to ensure donor support for such reforms (he presided over two crucial capital replenishments – in 1989 and 1994 – which resulted in significant capital increases for the bank).76 And most recently, the AfDB’s president, Donald Kaberuka, an economist by training, was Rwanda’s Finance Minister before taking the helm at the AfDB (2005–2015).

74

75 76

On May 22, 2011, the EBRD adopted a resolution to expand its operations to the Middle East and North Africa to support democratic transitions. Further, Chakrabarti’s former positions at the WB and IMF make him the first EBRD president with global IFI experience, perhaps signaling an expansion of the EBRD’s global role in supporting transition to market economies. Based on interviews with IDB officials 1999–2002 and 2007. Despite the IDB’s lengthy existence, Iglesias was only its third president. His predecessor, Antonio Ortiz Mena from Mexico, presided over the IDB for eighteen years (1970–1988), and the IDB’s first president, Felipe Herrera from Chile, held office for ten years (1960–1970).

3.4 Geography, States, and People 3.4.3

97

Shareholders: the Representatives of the Members

The articles of agreement of the IDB stipulate that borrowing member countries will have a voting power of at least 50.005 percent of the total (up until 1994, when Japan, Germany, Italy and France increased their stakes, it was 53.5 percent vote for the borrowing members). It is, according to most accounts, the regional bank that allows for the most control by its borrowing members – in both formal and informal ways – while maintaining stable donor support. Although the United States has substantial influence over loan decision made by the bank and is the single largest shareholder, regional members hold the overall majority of voting shares on the Board of Directors. Thus, despite the hegemonic influence of the United States, this distribution of votes and ownership of the IDB is markedly different from the WB, where little voting power is in the hands of borrowers. One of the major problems facing the AfDB is the lack of an apparent leader among its nonregional members. During turbulent times in Africa – civil wars and increasing poverty – no donor member is taking leadership. Unique among its peers, the AfDB’s foundations were very African: It was set up to be a development bank for Africa run by Africans and empowered by Africans. More than four decades after its creation, regional members still hold over half of the bank’s voting power. Although African countries conceded more control to nonregional shareholders in the early1980s, they still hold about 60 percent of the voting power (as of 2010).77 Because rich countries are more reluctant to support an institution over which they have little control, the African run AfDB has lacked stability. Although the AfDB seems to boast a formula where borrowing members exercise more control over their destiny, the price it pays in terms of donor support compromises its ability to be a central, effective development institution in the region, compared to the IDB, where donors – especially the United States – play a larger role.

77

Chapter 6 provides an analysis of the power structure within the RDBs and a more detailed account of the meaning of voting shares. It should be noted here that, similar to the global IFIs, many parameters that define the power of members in the institution were set in the Articles of Agreement. For example, both the IDB and the AfDB explicitly state in their founding agreement that borrowing members will hold a majority of the shares (50.005 percent, to be precise, at the IDB). Furthermore, member countries do not necessarily seek as many voting shares as possible. As explained in Chapter 6, being a member of the bank has benefits that are not associated with the amount of shares, but simply with having shares.

98

Origins, Politics, and Structure of RDBs

Japan and the United States are the AsDB’s two largest shareholders. Shareholders’ votes are calculated in such a way that Japan, Australia, and New Zealand are counted as regional members, even though they are not borrowers. Thus, the sum of 65.515 percent shares to regional countries can be misleading since Australia, Japan, and New Zealand hold 19.696 percent of this figure (nearly a third), leaving borrowing countries only 45.819 percent of the votes (shares). Therefore, of the three oldest RDBs, the AsDB is the one in which borrowing developing countries have the least voting shares. Despite its late arrival on the scene, the EBRD follows a similar model of institutional shareholder control as the other RDBs. However, the voting power rests predominantly with industrialized members of the Board. Like the AsDB, the EBRD is a donor-inspired and donorcontrolled institution. It is evident that the major European donors and the United States are the central decision-makers and that the Board usually takes its cue from these members, making sure votes take place only after their positions are clear.

3.5

lending strategies

The concessional lending windows of the RDBs have come under severe attack and scrutiny since the 1990s and were on the verge of closing down due to limited replenishment offers.78 The debate over these funds demonstrates the conflict between the corporate-like structure that seeks to remain in good financial standing and the pressures to help the neediest. Although my focus is on the regular (nonconcessional) and much more substantial lending windows of the RDBs,79 it is important to note that concessional funds distinguish the RDBs from commercial banks.80 The nonconcessional funds (OCR) are at times structured similarly to private sources of financing (e.g., the interest on loans and their 78

79

80

The IDB operates the (concessional) FSO, the AsDB uses the (concessional) AsDF, and the AfDB owns the (concessional) ADF. For a comparison of major donors’ contributions to concessional and nonconcessional funds, see Chapter 4, Table 4.1 (and supplemental website). For example, in 2010 and 2011 approximately three quarters of financial assistance to MDBs was on nonconcessional terms. Although, without the nonconcessional lending window, the banks would not exist at all as donors would have insufficient motivation to support them. Further, these lending windows are funded separately, with distinct replenishment negotiations. Both are, however, subject to the mission of the RDBs.

3.5 Lending Strategies

99

table 3.6 Net capital flows to developing countries, 1986–1995 (at constant 1995 prices) 1986

1991

1993

1995

Net FDI Net portfolio equity Bank lending and bonds

13.0 0.8 15.3 29.1

37.0 8.1 20.1 65.2

74.0 49.5 43.8 167.3

90.3 22.0 54.8 167.1

Concessional loansa Non-concessional loansb Grantsc

17.2 17.7 20.9 55.9

17.2 12.5 39.8 69.5

14.1 11.5 31.9 57.5

14.1 17.3 32.9 64.3

Private

Total Private Official

Total Official

Source: Development Committee, “Recent Trends in the Transfer of Resources to Developing Countries.” Document DC/93–3 Rev. 1 (Washington, DC, April 5, 1996). Reprinted in Culpeper (1997, p. 36). a Predominantly bilateral and MDBs. b Mainly from MDBs but also includes bilateral export credits. c Official development assistance, excluding technical cooperation.

maturities), while the “soft loan” window is a feature that arguably puts RDBs closer on the scale to aid institutions rather than banks.81 Table 3.6 demonstrates the various sources of capital flows that are available to developing countries. There has been a sharp increase in private flows to developing countries since the 1990s.82 At the same time, official flows have been relatively stable. Interestingly, of the total financial flows to developing countries, the share of grants has grown. It is noteworthy that grants are usually made bilaterally, while concessional and nonconcessional lending is primarily made by MDBs. This book’s focus on the nonconcessional lending window is based on the controversy that surrounds it: the OCR is essential for the RDBs to retain their donor capital contributions (they would retain support in the absence of the concessional lending window, but not without the nonconcessional lending window). The concessional lending window disburses loans that are effectively quasi grants. It is therefore more likely placed on the development agency side of the spectrum because 81

82

The AsDB, for example, disbursed 68% of its loans through the nonconcessional window until 1992. This is be further analyzed in Chapters 4 and 5.

100

Origins, Politics, and Structure of RDBs

banking considerations do not apply. It is the nonconcessional lending window that should be scrutinized since it is more prone to various pressures (political and economic) while officially subject to the banks’ fundamental principles that require them to approve loans to projects that would not otherwise attract funding (private or bilateral). Essentially, preference to support projects that would not be financed by other resources determines whether the RDBs’ OCR is development driven. To comply with their own mission RDBs must maintain constant scrutiny over the market opportunities for the projects they support.83 The IDB was originally set up to “bridge the gap” between the developed world and poor countries in Latin America.84 Following the philosophy of the time, loans in the early years were focused on the sectors of housing, sanitation, and education. This is contrary to critical thinking, both then and now, that the bank only lent to “bankable” projects: these sectors were not considered, at the time, profitable. During the IDB’s first decade, the United States was the sole “net” donor and held 42.05 percent of the voting power.85 Many consider this decade the “golden era” of the IDB – taking risks and being true to its mission, it began with great expectations.86 On the other hand, this was a decade in which the prevailing development strategy was import substitution; in retrospect, this strategy may not be the most advantageous for the promotion of sustainable development, but at the time it was viewed as a much needed “pain killer.” Moreover, this was a time when the United States, after the election of Nixon, became more internationally oriented rather than hemispheric, while some of the governments in Latin America took very nationalistic approaches in accordance with their military rule. Financially, this was a period characterized by the belief that external financing could solve everything.87 By the end of the decade, when more conservative theories of development and financing prevailed, the bank’s activities became marginalized. The 1980s were defined by one of the most severe economic crises in Latin American history: 83

84

85 86

87

In Chapter 4, I demonstrate the large difference in amounts of funds between the OCR and the concessional arm of the RDBs (Table 4.1). The funds for OCR are substantially higher, making it the central operative of the RDBs. “The Bank was created in response to a longstanding desire on the part of the Latin American nations for a development institution that would focus on the pressing problems of the region” (www.iadb.org/exr/english/ABOUTIDB/about_idb.htm). Tussie, 1995, p. 3; www.iadb.org/exr/english/ABOUTIDB/about_idb.htm. See Dell, 1972; White, 1970; interview with Leo Harari, special IDB representative to Europe, June 24, 2002. Sunkel, 1993, p. 32.

3.5 Lending Strategies

101

Interest rates escalated and overall international financial conditions deteriorated, wreaking havoc on borrowers. Credit became rationed. By the time the debt crisis erupted, Latin America’s external debt was three times the value of its exports. The volume of exports expanded rapidly during the decade, but per capita incomes declined significantly. Income disparities worsened in almost all countries. The period is rightly known as “the lost decade.” (Tussie, 1995, p. 4)

During these uncertain times, the IDB found itself in a peculiar position; it was suddenly forced to focus on the economic survival of its borrowers. Its failure to adjust to the new situation coupled with the deep economic crisis marked a decline in its prestige and overall support. The AsDB was created with a similar goal in mind. “It was recognized that having a regional character, the AsDB would be better suited to support the developmental needs of the smaller and relatively less developed countries. They had not been served adequately by the World Bank, which was perceived as concentrating its lending mainly in India and Pakistan” (Kappagoda, 1995, p. 14). Even though the United States was initially reluctant to join, American officials quickly realized that contributing to an institution that promotes development and cooperation presents a different face to the otherwise hostile involvement of the United States in Asia. Japan began its role as a regional hegemon in Asia by keeping a low profile, but American support for the AsDB offered the platform for Japan’s increased involvement, as the financial and political benefits that could result from a prominent Japanese role in the bank became evident. Like the IDB, the AsDB primarily lends (from its inception) to the public sector, with government guarantees. In 1983, the AsDB started direct assistance to the private sector, but these loans accounted for only 1.7 percent of the total.88 As for geographic distribution, Southeast Asia was the largest borrower during the first fifteen years of the bank’s operation, whereas South Asia became the principal borrower after 1982 (this is largely due to India’s borrowing which started in 1986). In addition, as the NIEs (Newly Industrialized Economies) began growing and their economies became successful at attracting private investment, they halted their borrowing status at the AsDB (this, of course, changed following the 1997 Asian Financial Crisis).89 Throughout the years, the agriculture sector garnered the largest shares of the loans, partly due to its importance in poverty alleviation. In addition, increased oil prices in the 1970s, made the energy sector 88 89

See Kappagoda, 1995, pp. 25–26; and www.asdb.org. See Appendix 2 for percentage shares of the largest AsDB borrowers.

102

Origins, Politics, and Structure of RDBs

a priority as well. Consequently, the lending strategy of the AsDB changed at the end of the 1970s: from a purely project-lending policy, the AsDB introduced a program lending policy (this was primarily geared to enhance certain sectors and productivity). Like the other two RDBs, the AfDB has become a significant player in its region in the last few decades. The AfDB has consistently increased disbursement to Africa and has managed to maintain positive net official flows (disbursements consistently higher than repayments).90 However, unlike its counterparts, donors seem reluctant and unsure of their contributions in each of the replenishments, creating uncertainty for the bank’s future. Overall, the agriculture sector has been the principal beneficiary of AfDB lending. This is not surprising given the level of poverty and the centrality of agriculture to the livelihoods of Africans. Infrastructure is the second largest sector receiving loans from the AfDB – mainly transportation, public utilities, and industry. These sectors are typically more “bankable” – results are fairly immediate and tangible. In the early years of the RDBs’ operations, loans to these sectors were necessary, but the overall increase of FDI usually indicates that more private resources that are available for these sectors. Moreover, in the late 1980s, as a result of the newly popular structural adjustment lending policy (which the AfDB defines as “multisector lending”), there was a sharp decline in lending to agriculture. This comes at a time when FDI is rising, therefore putting the bank’s development priorities in question. Over the years, the AfDB’s reputation has suffered because of various factors, both internal and external. The central external factors that have affected the AfDB are: (1) The fact that African countries have become more impoverished and it seems that very little, if any, progress has been made in poverty alleviation and development (especially in comparison to Latin America and Asia, where some success stories can be pointed out and where progress is more evident); and (2) The AfDB, unlike the AsDB and the IDB, has substantially less resources than the combination of the WB’s resources devoted to the region and the transfers and loans made bilaterally. It seems, then, that compared to other actors engaged in development in the region, the AfDB is relatively less significant than its counterparts. One of the major differences between the development banks is the source of their control. As mentioned, the World Bank, AsDB, and EBRD were created and controlled by the creditors. The IDB and AfDB were 90

See English and Mule, 1995, p. 65.

3.5 Lending Strategies

103

created and controlled by the borrowing countries (Culpeper, 1997, p. 12). This characteristic lends itself to some interesting observations. First, there is great differentiation among the banks, even when their setup was based on the same structural model. For example, in the most extreme case, the IDB and the AfDB have had very different track records over time. In fact, today, some consider the IDB the most “successful” RDB (as evidenced by the latest increase in funds under the eighth and ninth replenishments) while the AfDB is struggling with a fluctuating credit rating. Given that both banks are based on a similar working model, and that both prioritize borrowing members’ decision-making, it is reasonable to assume that the roots of the difference are either the variation of economic conditions of the regional members, or the extent of hegemonic involvement in the institutions (this is examined in Chapters 5 and 6). In addition, a cultural factor that includes governance, method of operation, and monitoring, may be an important contributing factor to this deviation (tested in Chapter 6). Even if at first the RDBs were viewed as supplemental to the work of the WB, over time the RDBs have become increasingly influential in their regions. In fact, the latest capital replenishments have made their lending capacity (within their respective regions) substantial even in comparison with the global institution.91 Moreover, in many of the smaller and some of the poorest countries, RDBs make more loans in absolute terms than the WB. Finally, just in the last decade, the percentage increase in both the number of loans and their total dollar amount was substantially higher in the RDBs than in the WB. Overall, the RDBs’ total loans to all regions, taken together, have gotten progressively closer to the amount of yearly loans made by the WB (especially the IDB and the AsDB, where the difference between the amount of loans they make to their regions and those made by the WB is shrinking rapidly).92 The creation of the EBRD in the 1990s was of vital importance for the future of RDBs. The EBRD represents an affirmation of the multilateral support for the institution of RDBs. The differences in the EBRD’s structure and mandate from those of its counterparts point to a shift in policy that in turn had an impact on the other RDBs. Its larger focus on a relationship 91

92

“It is evident, however, that recent increases in the capital of the regional banks have given those institutions a lending capacity comparable to (and in the case of the IDB, exceeding) that of the World Bank within their respective regions” (Culpeper, 1997, 12). See Chapter 6 for a comparison of the total amount of loans made by each RDB over the years, to those made by the WB to the corresponding region. See Chapter 6 for more detail.

104

Origins, Politics, and Structure of RDBs

with the private sector and the different shareholders’ role are part of the shifting nature of multilateral financial institutions; this contributes to the debate over their necessity and the prospects for their continued existence while, paradoxically, securing their financial future. This is not to say that the EBRD presents an improved model of an RDB. On the contrary, it serves to highlight some of the most serious shortcomings of the regional banks. Often in the center of controversy, the EBRD’s main mission, the transformation of former Soviet Bloc states into market economies (overtly political) has sometimes overlapped with the work of commercial banks or private sector initiatives, thus putting its necessity as a public institution into question. The EBRD’s existence helped bring to the center of the debate the schizophrenic nature of development banks and raise questions concerning our expectations of these institutions and their own goals and policies. One of the central critiques of the WB was its adjustment lending in the 1980s and 1990s. The idea behind this type of lending was that the only way to combat the debt crisis was to condition loans made to developing countries on the recipient’s commitment to advance policies that promote growth.93 However, as Easterly (2001) argues, adjustment lending was a failure. Even though lending grew substantially, growth actually declined in this period.94 Some countries responded positively to the adjustment loans, albeit after a few years, but, as Easterly concludes, these loans did not address the problem of inflation in countries that borrowed, and therefore results were weak. The data on inflation and adjustment lending is especially telling with regard to the EBRD. Since this bank’s “. . . core business is the financing of projects that advance the transition to market economies,” (The EBRD: Its Role and Activities, 1999), it is alarming that between 1991 and 1998, inflation in former Communist countries rose from 200 percent to 64,000 percent (cumulative inflation of the average former Communist country).95 Moreover, according to Easterly, “. . . a cumulative output decline in the 1990s of 41 percent for the typical ex-Communist economy in Eastern Europe, with the percent of population living on less than $2/day increasing from 1.7% to 20.8%” (ibid., p. 106). If the creation of the EBRD was supposed to help Europe’s developing economies, these figures point to 93

94 95

It should be noted here that the RDBs do not exercise conditionality in the way the IMF and WB do. They have an agenda for policy preferences that guide their decisions to disburse loans, but once they approve a specific project or program, there are no conditions attached (at least not in any official way). See Easterly, 2001, chapter 6, and p. 103 for graph. Easterly, 2001, pp. 106–107.

3.5 Lending Strategies

105

a disturbing transition process.96 Therefore, although there are some success stories, adjustment lending policies by the WB and IMF clearly do not produce magical results, particularly in the face of other competing economic, social, and political factors. Given the financial pressures imposed on them by donors and by their own institutional constraints, RDBs often fall into the trap of lending strategically to satisfy financial parameters rather than for the purpose of development. The RDBs engage in strategic lending when the destination of the loans or the mere disbursement of the loans serve the banks’ political and/or financial interests (and the need to “get the loans out”), rather than development, poverty alleviation, or making sure that no other sources of funding would be available. This often happens due to budgetary concerns – the need to disburse a certain amount of loans by the end of the fiscal year to demonstrate positive activity and a balanced budget. Generally, such loans are made for two reasons: (1) to support powerful shareholders’ preferred policies vis-à-vis relations with certain countries (preferences and interests) and (2) to ensure enough loans are disbursed and “good results” are achieved. As long as these two conditions are regularly met, the banks are sure to sustain their shareholder support, despite all the criticism they receive. For example, Easterly (2001) observes, “A recent World Bank study found that aid does not influence countries’ choice of policies. Nor do donor experts consider the worthiness of countries’ policies in determining which ones are given aid. Aid appears to be determined by the strategic interests of donors, not by policy choice of the recipients” (p. 110). Evidently, aid disbursement is meant to contribute to recipient countries’ development, but even at the WB, as Easterly notes, donors’ interests play a larger role in determining aid policies. This is an important factor influencing bank-like or development agency policies. There is often an informal “power struggle” between the shareholders and staff bankers of the RDBs. At times, while the Governors and Directors receive their guidelines from the political apparatus of their respective countries, the staff bankers are thinking of the balance sheets and financial records. Conversely, it is sometimes the Board that is concerned with the finances of the bank, while the managers and staff are more aware and

96

And since the EBRD has been accused of essentially competing with the private sector and making loans to “bankable” projects (suits the goal of advancing market economies) it would not be surprising that its contribution to development is not great.

106

Origins, Politics, and Structure of RDBs

concerned with development issues facing their borrowers.97 In this delicate relationship between donors, borrowers, and professional staff, needy and higher-risk recipients can fall by the wayside; time spent on strategizing detracts from improving evaluation of lending, adjusting policies, facilitating research, and informing decision-makers. Hence, when RDBs report loans as “successful” we cannot deduce whether policy reforms have been made, which preferences were utilized by the banks when approving a loan, and whether the recipient could have received the loan from other sources. There are many difficulties associated with enforcing supervision of adjustment loans, since it is close to impossible and certainly not easily feasible to monitor not only the implementation of the projects for which loans were given, but also the actions of the borrowing government, the accuracy of their reports, their accounting techniques, etc. Perhaps adjustment lending is not a good way to promote development. This is especially true if it encourages coverups and, ultimately, bad policies, since (1) adjustments often do not constitute the initial building blocks necessary for sustainable reform, and (2) the close scrutiny of the adjustment policies creates fear among developing countries, essentially encouraging them to “show” results rather than actually undertake policy changes. This problem is augmented when the poor countries that receive this conditional aid use it for patchwork rather than for real adjustments necessary to alleviate poverty. The trap multilateral financial institutions fall into is entirely counterproductive for achieving their goals: from the point of view of the banks, loans were approved and disbursed after thorough investigation into their worthiness, expected value, and returns. The various country departments also compete for budgets and therefore must disburse their given budget during the year so as not to lose funds for the following year.98 Therefore, making the loan can sometimes be more important than the loan’s substance. For the poor countries, these loans can be constructive in a positive scenario, an impediment to development

97

98

This reversal of role on the part of the Board and mangers/staff stems from various systemic and internal constraints that guide the policy preferences of each of these players (sometimes altering them). “Most donor institutions are set up with a separate country department for each country or group of countries. The budget of this department is determined by the amount of resources it disburses to recipients. A department that does not disburse its loan budget will likely receive a smaller budget the following year. Larger budgets are associated with more prestige and more career advancement, so the people in the country departments feel the incentive to disburse even when loan conditions are not met” (Easterly, 2001, p. 116).

3.6 Conclusion: What Are They Like, What Do They Do

107

in a negative scenario, or a reenforcement of bad policies already undertaken by corrupt or dishonest governments.

3.6

conclusion: what are they like, what do they do

Some of the most common criticisms of RDBs are that they essentially perform tasks that could be undertaken by the private sector and commercial banks, or that their activities largely overlap with each other, the WB and the private sector, causing unnecessary competition, duplication, and waste. In its suggestions for reform, the IFIAC Report concluded that there is merit to the claim that overlap in the activities of the development banks contributes to inefficiency. Others argue that over the years the development banks evolved and now supply more public goods – a niche not sufficiently addressed by private sector and other institutions.99 This chapter compared how each of the RDBs was created. It showed that the factors that led to their foundation – under the pressure of developing countries (bottom-up) or as a result of a decision by developed countries (top-down) – had an impact on the institutional structure of each of the RDBs. An examination of the RDBs shareholder makeup, presidents, staff, and even location demonstrated the differences between the development banks, and the effect that their unique foundation circumstances have on the way RDBs operate and how they are perceived. Specifically, this chapter included an examination of the RDBs membership (shareholders) and the staff and management that are responsible for the institutions’ operations. In addition, it explained some of the important constraints on the RDBs’ policies – their need to maintain high credit ratings and balance budgets, and to protect the sources of their capital. In this context the chapter assessed the differences between concessional lending and nonconcessional lending (OCR). All this explains not only the importance of these factors to the members and staff, but also how they contribute to the RDBs’ balancing of their duties as a development agency with the need to maintain a strong financial portfolio. This discussion, then, puts the following quantitative analysis of the loans made by the RDBs in a historical and institutional context. Tying the foundations of each of the RDBs to their current status as development aid institutions is crucial in a political environment where international organizations are being scrutinized and public expenditures overhauled. The variation in evolution of each of these banks contributes 99

See Culpeper, 1997, p. 8.

108

Origins, Politics, and Structure of RDBs

to the explanation of how their policies (and thus effectiveness) differ. Overall, this chapter lays the foundation for questioning the logic of having numerous institutions perform a similar task (whether overlapping in region or not, and in light of the World Bank’s lending). Those who support the banks see this competition as a natural and healthy way to improve development aid and benefit developing countries. The RDBs may lower interest rates and provide better services as a result of the competition. And since the RDBs, like other lending institutions, justify their existence by pursuing their yearly lending goals, they are searching for projects to fund. As for donor countries, in addition to being “part of the club” by joining the institution, they are also providing opportunities for their own private sectors since in most cases only companies based in shareholder countries are eligible to bid for projects sponsored by the RDBs (except for the EBRD – its bids are open to all). These issues will be further explored in the next three chapters.

4 RDB Loans and Developing Countries

If you lend money to my people, to the poor among you, you shall not deal with them as a creditor; you shall not exact interest from them. – Exodus 22:25

Can a bank whose primary function is making loans (with interest) perform the task of development in poor countries? This chapter unpacks the loaded term “development bank” by juxtaposing each regional development bank’s overall lending in the last few decades to the socioeconomic changes that took place in the developing regions that are on the receiving ends of the loans. On the micro level, I take a closer look at the most developed country and the least developed country from each region serviced by the RDBs, and, returning to the macro level I examine the RDBs’ performance in comparison to the World Bank’s. Below, I examine the social and economic parameters that form the basis of the statistical analysis in Chapter 5. I first evaluate the levels of development of the regions that receive loans, and analyze selected countries relative to each other within each region. I then evaluate the loan amounts made by the RDBs over the years of their existence, including how RDB loan amounts compare to those of the World Bank. These figures support the conclusion that the loan amounts to all regions have significantly increased since the 1960s, with a particular sharp rise in the 1980s. The development mission of the RDBs is constantly put to the test of changing global socioeconomic conditions. Specifically, the levels of GDP per capita, foreign direct investment (FDI) flows, health, and education indicators, have all changed over time in developing countries that receive 109

RDB Loans and Developing Countries

110

RDB loans. For the banks to remain relevant as development institutions, their lending should be adjusted to account for these changes. When, for example, some developing countries begin to attract more private capital (FDI), we would expect that the RDBs that lend to these countries are no longer the sole source of capital: they could very well be competing with private sector lenders – a violation of their stated mission. To be sure, all four regions have improved their GDP per capita and lowered infant mortality rates. They also have all been able to attract more FDI as of the late 1980s and early 1990s. However, the differences between the regions and among countries within each region are sometimes stark, complicating RDB lending policy. While Chapters 5 and 6 lay out an in-depth analysis of these results, the present chapter raises other important questions. For example, why are RDBs disbursing much larger amounts of loans at the same time that FDI is becoming more prevalent in developing countries? Or, looking at the gross-versus-net loans to Africa, for example, we can ask, has the debt servicing that began in earnest in the 1980s had a negative impact on the borrowing countries (i.e., when the balance sheet is negative in favor of loan servicing, are recipient countries really benefitting from development loans)?

4.1

data: description and sources

To test whether RDBs undertake strategic lending by providing more loans to (1) borrowing member countries that have more advanced economies and fare better on social indicators, and/or (2) borrowing countries that serve the political interests of the powerful shareholders of the banks (hegemons), I assembled an original panel time-series data set that includes economic, social, and political parameters of the borrowing members. In their borrowing membership, these four RDBs make loans to most of the developing world. In fact, the only developing countries that are not currently borrowing members of any of these RDBs are Cuba, North Korea, and the countries of the Middle East1 (Afghanistan during 1

There is no development bank in the Middle East. However plans to create one have been discussed and written about in detail, especially since the signing of the Oslo Pease Accords in 1993. It is unlikely that a regional financial institution can be created in the Middle East until the political conflicts of the region are resolved. Additionally, the Islamic Development Bank should not be confused with the RDBs; it operates in Islamic countries and is not regional-specific, and it is governed by a different set of rules and guidelines, closer to those of a charity organization. As mentioned in the previous chapter, the EBRD

4.1 Data: Description and Sources

111

the Taliban rule was suspended from membership in the AsDB). The RDBs’ regions of lending are geographically based: the IDB lends only to Latin American and Caribbean countries; the AfDB lends only to African countries; the AsDB lends to Asian countries; and the EBRD to Central and Eastern European countries. The only area of overlap in borrowing countries’ membership is between the AsDB and the EBRD: Some of the former Soviet Republics, east of Russia, can borrow from both banks.2 Moreover these four regional banks work in the same countries that are eligible for World Bank loans. Their political importance is therefore compounded by the fact that they perform similar functions to the WB in the same countries. 4.1.1

A Note on the World Bank

Since the WB and the RDBs perform similar functions in the countries that are eligible for RDB loans, the RDBs interaction or “strategy” vis-à-vis the WB can provide an insight into their institutional constraints.3 Four main forms of possible interaction between the RDBs and the WB emerge: the first is cofinancing loans (collaboration), which is the highest level of cooperative endeavor. There are a growing number of projects that are financed this way – allowing for the risk to spread and for the institutions to coordinate their efforts. The second, lower on the scale of cooperation, is that of coordination. This happens when the banks inform each other of their intentions and actions, making sure they work in tandem. The third possible interaction is that of “noninteraction,” when the RDBs and the WB work on different matters and focus on different projects. It is lower than coordination on the cooperation spectrum, although not entirely noncooperative since it suggests that the organizations came to an active agreement in defining different spheres. The fourth form of interaction is that of lending to the same projects, competitively. Many of the officials I interviewed at the various RDBs state that the level of all types of cooperative interactions are minimal at best when the banks make their decisions. They share a sense that the competitive nature of the lending process between the institutions stems from the need

2 3

has recently been refocused to make loans available to Middle Eastern and North African countries undergoing processes of democratization, but the full scale (and future) of this expansion is yet to be determined. These are the countries that border Asia, such as Kirgizstan and Tajikistan. For more on the World Bank lending and decision-making see Weaver, 2008; Nielson and Tierney, 2003, 2005; Easterly, 2001; Wade, 1996, 2002; and Stone, 2011.

112

RDB Loans and Developing Countries

to “get the loans out” and to finance projects with good prospects for success. In addition, some officials suggested that developing countries, knowing that this competition exists, try to capitalize on it by seeking more favorable terms on loans. The following diagram represents these categories of levels of interaction between the RDBs and the WB: Cooperative interaction

Co-finance coordination Non-interaction competition

Non-cooperative interaction

Although the WB is not the focus of this book, the above diagram provides a useful illustration of the contradiction in the RDBs’ design: that is, the conflict between their roles as banks versus as development agencies. Especially when they are moved to act in ways that are closer to the “competition” end of the spectrum, the RDBs may overlook need-based criteria for making loans in their rush to win the contest. But even at the cooperative end of the spectrum, the existence of a number of institutions performing similar tasks suggest that they each have to make sure that they remain indispensable, and perform in accordance with demands made by shareholders to ensure an ongoing flow of contributions.4 The relationship between the RDBs and the WB can inform us about the relationship these banks are developing with the private sector: as more FDI is available to developing countries, RDBs are forced to reconcile their commitment to maintain a “lender-of-last-resort”5 status. 4.1.2

Concessional and Nonconcessional Loans

The data of loans disbursed by each RDB was collected in the following way: the total yearly amount (of loans disbursed) to each borrowing member 4

5

Gould’s (2003, 2006) discussion of supplementary financiers at the IMF is helpful in explaining the IO’s institutional constraints in collaborating with other lenders/financiers. I use this term to refer to the commitment the RDBs make in their Articles of Agreement not to compete with other sources of financing that are available to their borrowing members.

4.1 Data: Description and Sources

113

table 4.1 Top ten recipients of concessional and nonconcessional loans, 2010 IDB (million $US) Mexico Brazil Argentina Venezuela Colombia Jamaica Ecuador El Salvador Peru Panama TOTAL

IDB-FSO (million $US) 3,010 2,260 1,165 890 685 630 509 435 341 340 10,265

AsDB (million $US) India China Bangladesh Indonesia Pakistan Kazakhstan Philippines Vietnam Thailand Regional TOTAL

103 85 60 18 16 14

TOTAL

296 AsDF (million $US)

2,370 1,588 800 785 629 606 600 510 504 475 9,163

AfDB (million $US) Egypt Morocco South Africa Tunisia Multinational Senegal Nigeria Rwanda Mali Cameroon TOTAL

Honduras Nicaragua Bolivia Guyana Paraguay Guatemala

Vietnam Bangladesh Afghanistan Pakistan Uzbekistan Nepal Kyrgyzstan Cambodia Laos Tajikistan TOTAL

580 449 352 270 265 263 168 161 152 122 2,782

AfDF (million $US) 1.021 815 633 465 209 111 106 50 40 36 3,486

Source: CRS 7–5700 (R41170).

Ethiopia Multinational Congo, Dem. Rep. Tanzania Kenya Ghana Niger Cameroon Benin Mali TOTAL

352 291 248 203 183 172 84 76 67 64 1,740

114

RDB Loans and Developing Countries

country since each of the RDBs began their lending operation (i.e., total loans for each country per year). The data in the statistical analysis include only the nonconcessional loans made by the RDBs. Nonconcessional loans from the ordinary capital resources (OCR) have constituted the lion’s share of the loans for all RDBs. OCR is the main lending window of the RDBs (and the only one at the EBRD). The vast majority of loans – over 90 percent – have been made from this lending window over the years. Although there is a shift, in some of the cases, toward increasing the concessional lending window, the RDBs are sustained via the OCR, which was the first type of loan created at all RDBs, and has been at the center of their operations. For example, in 2010 and FY2011, “about three quarters of the financial assistance provided by the MDBs to developing countries was on non-concessional terms . . .” (CRS Report for Congress, April 2012). Thus, although the share of concessional lending has grown, especially in recent years, it is still a small part of their operations, and not their central mission. Moreover, the data for concessional loans is sparse and covers fewer years.6 Table 4.1 compares the top ten recipients of concessional and nonconcessional lending of the RDBs in 2010. The significant variance between nonconcessional and concessional lending remains at least the same when loan amounts to all 2010 recipients are aggregated. These stark differences paint the picture of the development banks as institutions that conduct most of their work via nonconcessional lending. The AfDB, whose concessional lending window is relatively larger in relation to nonconcessional lending for 2010 (approximately 1:2), also has the smallest overall lending amounts. The reason then for focusing on the nonconcessional lending window in the analysis of the banks’ lending operations is threefold.7 First, the nonconcessional lending window is endowed with substantially larger funds and reaches more countries. For example, by the end of 1992, 68 percent of the AsDB’s loans were disbursed via the nonconcessional window (Kappagoda, 1995, 25).8 Second, this window is the most controversial.

6

7

8

In the supplemental website, I provide additional data on concessional loans, comparing them to the OCR, so that the RDBs’ overall activities are put in perspective. See Table 3.4 for the net capital flows to developing countries, with a breakdown of concessional and nonconcessional loans. To be eligible for concessional funds, a country has to prove a certain level of poverty, and since the funds are smaller, the scales of the projects approved through the concessional window bear direct relation to these smaller loans. However, given the level of poverty of the countries eligible for these loans since these loans, their relative impact can be substantial.

4.1 Data: Description and Sources

115

Since the interest rates on the loans are at times not significantly different from commercial lending,9 the primacy of nonconscessional lending raises the question of whether the RDBs are more concerned with “getting loans out the door” while giving preference to “bankable” projects, or to identifying the most urgent development needs. Moreover, concessional loans are funded through donations to the development funds, while the nonconcessional lending window, which is based on OCR, is funded through market borrowings and the paid-in capital of the members. Furthermore, these “hard loans” are much less flexible than the “soft,” concessional loans. The “soft loans” (concessional) are not a source of liability in the same way as nonconcessional loans: they do not present a conventional balance sheet of assets and liabilities and, thus, are not the subject of bondholders’ or credit-rating agencies’ scrutiny. And finally, the concessional window of lending is not comparable to private lending. There is an understanding that concessional loans are, essentially, credit for which there is a net loss: overall this lending window is closer in character to grants and requires a different type of analysis. The concessional window of lending is subject to a debate about its place within the RDBs and, in the case of the AfDB, the concessional lending window has been substantially marginalized. The African Development Fund (ADF) – the AfDB’s concessional lending entity – has been operating on a minimal scale since donor countries, unhappy with the performance of the AfDB, have limited contributions to the Fund.10 While it may be the case that the nonconcessional loans are a “necessary evil” for the institutions to be able to also have a “soft” lending arm, it seems unreasonable for an institution to devote only a fraction of its resources to actually fulfill its mission.11 Furthermore, members’ financial contributions – both paid in and callable – are only made to the OCR. The concessional lending funds are separate funds with a separate mechanism for fundraising.12 The institutional structure of the RDBs, then, places OCR at the center of operations. 9

10

11

12

Most RDBs do not disclose information about the interest rates they charge for loans. This is especially true for the EBRD. This information was obtained mostly through interviews. See supplemental website for a comparison of concessional and nonconcessional lending by the IDB, AsDB, AfDB, and WB. Admittedly, this is a hard test for the RDBs, but if their central operations do not carry out the mission they were meant to fulfill when they were created, then are these operations really justified? Further, the EBRD’s more recent creation without a concessional lending window is evidence for donors’ interests. Contributions to the concessional window do not succumb to the same institutional structure of shareholder commitments/votes as the OCR. It is therefore impossible to

116

RDB Loans and Developing Countries 4.1.3

Data

Most of the data for this analysis was obtained from the banks themselves: for the IDB, I collected the loan amounts from the bank’s internal data portal (intranet) and annual reports. Loan amounts are calculated in yearly, aggregate figures, for each recipient country, from 1961 until 2009. For the AsDB, the data was computed based on internal records documenting all loans to all countries in all the years of the bank’s operations;13 the original documents covered each loan, and the loans had to be added for each county, each year from 1967 to 2000, while loan amounts for 2001 to 2009 were retrieved from the AsDB’s annual reports. The EBRD’s accounting department provided loan data. For the years 1991–1994, the EBRD’s record was an aggregate figure (all these years grouped). I divided the total equally between the years as a form of estimation for the analysis. For the remaining years (1995–2000), data was provided as yearly aggregate for each country, and for the 2001 to 2009 time period yearly loan disbursements were extracted from the EBRD’s annual reports. Yearly data for the AfDB was much more difficult to obtain directly from the bank. I retrieved the 1966–2000 figures from the OECD/DAC database (2002), and 2001–2009 data from various annual reports of the AfDB. All the loans analyzed are in gross figures (to capture the transfers from the banks to the borrowers and not the interest and loans returned). The indicators that represent borrowing members’ economic and social development were obtained from various sources: FDI from the United Nations Conference on Trade and Development (UNCTAD); GDP, population, trade openness, primary school enrollment, and infant mortality rates from the Word Bank and World Development Indicators (WDI). As for the political indicators, the level of democracy is from the Polity data set, and the level of “shared interest” (relations between hegemons and borrowing states) is represented by the “Affinity Measure,” a coding of UN General Assembly votes between pairs of countries.14 The pairs used

13

14

include these funds in the analysis as they are subject to different rules in a different institutional setting. “Loan, Technical Assistance and Private Sector Operations Approvals,” Asian Development Bank, December 2000. This data set was compiled by Erik Gartzke, UC-San Diego: http://dss.ucsd.edu/~egartzke/ htmlpages/data.html and updated more recently by Eric Voeten, Georgetown University: Anton Strezhnev; Erik Voeten, 2012–06, “United Nations General Assembly Voting Data,” http://hdl.handle.net/1902.1/12379 UNF:5:iiB+pKXYsW9xMMP2wfY1oQ== V3 [Version]

4.2 Variables and Explanation of Models

117

for this study are between nonborrowing hegemon(s) and all the borrowing members in their respective regions (the United States for the IDB, Japan and US for the AsDB, France and US for the AfDB, and US, UK, Germany, and France for EBRD).15

4.2

variables and explanation of models

The social and economic conditions in the countries that borrow from the RDBs provide some evidence for respective poverty and economic and social development. By examining indicators that point to levels of health and education (infant mortality rate and primary school enrollment), and the borrowers’ relative economic development (GDP per capita), we can put the loans made by the RDBs in the context of relative levels of development of the countries receiving them. Furthermore, FDI is included to illustrate the borrower’s comparative ability to attract private capital: higher levels of FDI are indicative of financial markets’ confidence reflected in higher incoming capital flows. Access to FDI may allow a government to direct larger portions of its budget to sectors such as health and education, since more capital flows may release some government funds.16 GDP per capita is the most commonly used indicator of economic development and is used as proxy for the relative well-being of the recipient countries.17 For this study, the association between FDI and loans is particularly important in observing how FDI changes over time, and whether there is a correlation with changes in RDB loan amounts: if higher levels of FDI are a measure of more development – or at the very least, increased access to private financing – we would expect loans by RDBs to decrease where FDI increases (to that particular country). Examining whether loans and FDI

15

16

17

This Affinity Measure is a value between −1 and 1 that includes the degree to which pairs of countries vote in a similar fashion in the UN General Assembly in any given year (since 1945). The original data had to be transformed from the original dyadic form to the monadic format of my data set. This entailed the selection of the relevant countries and rearranging of the affinity measures in relation to hegemons. This obviously depends on the level of corruption and the strategy of local governments, in addition to the pressures of private investors. This study does not go into these issues, but they should be kept in mind (see Rodrik, 1999 for more on FDI and the role of governments). See Neumayer, 2003, p. 105. As he points out, despite criticism that GDP per capita “does not adequately and comprehensively reflect recipient need” the existing literature has often used only income per capita because it is more comprehensive and readily available in terms of its scope and reliability of the data.

118

RDB Loans and Developing Countries

have similar destinations generates findings about whether RDBs are following their “lender-of-last-resort” mission.18 The relationship between FDI and RDB loans can go both ways: either the RDBs’ lending precedes and encourages FDI (perpetuating development), or, loans are disbursed to countries that have already been attracting FDI (strategic lending). The former paints a picture of RDBs as development agencies, providing, through lending, incentives for private sector investment. The latter implies that their risk-averse nature increases the likelihood that RDBs approve loans to countries that are already attracting international private financial resources, making the loans a safer investment (and possibly compromising RDBs’ stated commitment to poverty alleviation). It should be stressed, however, that if the former is true, that RDBs pave the way for FDI, we should expect to see a decrease in the amount of loans to a country that attracted larger FDI amounts the previous year (as FDI increases, loans decrease). Finally, FDI is the best measure for private investment because of its long-term focus, which facilitates more reliable evaluation over time. Other private flows such as portfolio and equity investment do not share this long-term characteristic, as they are much more easily withdrawn.19 FDI represents all capital transactions that are made in order to acquire a lasting interest in an enterprise operating in an economy other than that of the investor, where the investor’s purpose is to have an effective voice in the management of the enterprise.20 Essentially, FDI is a reflection of the creation of economic units in developing countries by private actors. Unlike portfolio investment that is short-term, with profit making reliant on exchange rates and interest rates, FDI engages in investment activities that are closer in nature to those made by the development banks themselves. As shall be further discussed, FDI can engender problems in

18

19

20

While the problem of endogeneity between lending and FDI is real and difficult to solve in its entirety, my main objective is to find association rather than causation. Lagging FDI in the statistical analysis that follows in Chapter 5 provides some further assurances, and tests whether loans are more highly associated with increased FDI in the preceding year. See Bird and Rowlands (2003), and Edwards (2006) for a discussion of catalytic finance at the IMF. Adding portfolio investment to the analysis in future research could enhance insight into private financial flows to developing countries. It is not, however, imperative for this analysis, particularly since the tools of RDBs – loans – are long-term financial investments and therefore FDI would be a better private sector comparison. Since FDI represents net values, it could not be logged. The units of the data are therefore in $1000 US. (Thus, as we shall see in Chapter 5, the coefficients are often very small, reflecting the larger values of FDI.)

4.2 Variables and Explanation of Models

119

measuring these attributes, but since the RDBs themselves, as well as many analysts, consider FDI an indicator of developing countries’ growth and development scope, its inclusion in the analysis is imperative. By many accounts, FDI signifies economic stability, the strength of private sectors, and the confidence of foreign investors. For example, since the 1990s the amount of official (bilateral and multilateral) flows into most developing countries declined substantially relative to inflows of private capital. The end of the Cold War altered investor confidence resulting in rising FDI to emerging markets, now considered investment opportunities. As a result, changes in the global financial landscape raise some important questions about the impact of RDBs. Most obviously – since the mission of these institutions is to fill a certain gap at a certain point in time – does this gap still exist? And if not, have the RDBs adjusted their policies to reflect these changes? The degree of trade openness is another important economic indicator of development. For this analysis it is especially important when comparing lending policies during and after the Cold War. Coupled with FDI, trade openness is a proxy for countries with market-friendly policies, and if more loans are directed to these countries, the results can inform conclusions about RDB policy-preferences: for example, whether developing countries that adopt free market-oriented policies are able to attract more loans. Preferring countries that espouse free-trade policies may add to a set of considerations adopted by the RDBs that do not necessarily follow a need-based, development-driven strategy.21 Primary school enrollment represents, in this study, the level of education for each country in the analysis: Enrollment percentages in primary school are indicative of a basic degree of education that is essential for measuring development and socioeconomic well-being. Since education is, usually, a long-term investment and does not yield immediate results, a high level of enrollment in primary schools can reflect government spending policies and available resources, as well as being a sign of longterm planning.22 The relationship between education and growth is, notably, controversial. Easterly (2001) explains that studies find that there is no relation between growth and years of schooling, but that initial schooling is positively correlated with productivity growth. He brings 21 22

It does directly relate, though, to donors’ motivation to open additional markets. The Human Development Index (HDI) of the UNDP includes primary school education in its measurement of “knowledge.”

120

RDB Loans and Developing Countries

examples from a study that shows that during times of higher expenditures on education, growth rates have fallen sharply: “The median growth rate of poor countries has fallen over time. The growth of output per worker was 3 percent in the 1960s, 2.5 percent in the 1970s, −0.5 percent in the 1980s, and 0 percent in the 1990s. This study noted that the decline in growth happened at the same time as the massive educational expansion in the poor countries” (Easterly, 2001, p. 74). These findings reinforce the claim that education is a long-term investment. Moreover, it questions whether growth and development are positively associated. Thus, while investment in education can be considered an integral part of development policy, economic growth is not necessarily an obvious indicator for development.23 In this book then, the measure of education serves to highlight the level of long-term development commitment exercised by the RDBs (which would be in line with their mission). Low levels of school enrollment and literacy are often highly correlated with poverty, and although the improvement in education levels do not reflect immediate changes in the level of poverty (or growth levels), they do relate to the goals and strategies of the development banks. The rate of infant mortality is a useful indicator for a country’s state of health services and poverty. The UN’s Human Development Index (HDI) takes into account infant mortality rates in its “longevity” variable.24 In very poor countries, the infant mortality rate is typically very high. For example, the UNDP’s 2007 Human Development Report ranks the life expectancy at birth in Japan as the highest, at 82.7, while Sierra Leone’s is one of the lowest, at 47.3. This nearly twofold difference reflects the disparity between developed and developing countries: while most developing countries rank at 78 percent or below in the life-expectancy-at-birth indicator, no developed country ranks under 78. Like education, investment in health is a long-term commitment. Countries with higher infant mortality rates are more likely to be in need of aid, and as lenders that are committed to eradicate poverty, to provide capital where it is most needed, and where it is not otherwise available; the RDBs would be expected to target those countries. The affinity measure is used as a proxy to assess the possible influence of hegemons on RDB policy. It indicates a level of shared interests between 23 24

See Easterly, 2001. See Appendix 1 for the Human Development Report’s life expectancy chart.

4.3 A Historical Look at RDB Loans

121

a borrowing member country of an RDB and the bank’s hegemon(s).25 For the IDB, for example, the pairings of the United States and every Latin American country are examined via their shared voting at the UN. If borrowing countries that vote alongside the United States are also the recipients of a greater share of the loans, then there is an association between the borrowing country’s degree of shared interests with the hegemon(s) of the institution (to the extent that UN votes demonstrate shared interests) and their likelihood to receive loans. Finally, including a measure of democratization (Polity II data set) is also particularly important for differentiating between policies that were subject to Cold War politics, and those that are driven by a post–Cold War world order. In other words, if global spheres of influence were the priority of (Western) hegemons during the Cold War, then the degree of democracy of the borrowing country should not be of concern. I expect that after the Cold War we would see a policy shift, commensurate with changes in the interests of the RDBs’ major donor countries (especially the United States). In sum, the variations in social, economic, and political development among the countries that borrow from the RDBs furnish the setting in which the institutions operate. In examining the pressures that result with the possible conflicting policy preferences of a “development agency” and a “bank,” the relationship between the development banks and their borrowing countries calls for analysis of the borrowers in the context of their regions (i.e., their fellow borrowing members). First, an overview of the RDBs’ lending provides a context. Chapters 5 and 6 then put together these two components – the development needs of the borrowing countries on the one hand and the policies of the RDBs on the other.

4.3

a historical look at rdb loans

The loans I analyze in this book are actual disbursements rather than commitments. Disbursements represent the real transfers each year, and they portray a more accurate measure of how much money the banks actually dispense to their borrowing members. Commitments are the approved loans by the Executive Committees (boards) of the RDBs. Once commitments are made, negotiations take place for obtaining contracts to conduct projects or run programs. The disbursements, which 25

See Copelovitch (2010, pp. 45–49) for discussion of Collective Principal as it applies to the IMF.

RDB Loans and Developing Countries

122

table 4.2 IDB authorized and disbursed funds, 1961–1970 Year

Authorized (million $US)

% Disbursed

1961 1962 1963 1964 1965 1966 1967 1968 1969 1970

272.3 322.5 247.1 281.8 367.7 381.0 487.3 432.7 631.8 644.4

6.6 58.7 141.0 198.1 181.9 212.1 242.4 290.7 393.4 427.7

Source: IDB Annual Reports (Dell, 1972, p. 125).

follow what is at times a prolonged negotiation process, symbolize a more tangible, measurable lending activity: disbursements are usually a fraction of the commitments, which could be disbursed over a number of years – and sometimes not disbursed at all, if, along the way, things do not go well. I therefore analyze only those loans that countries receive, rather than those that are promised in principle. To illustrate the differences in yearly commitment versus yearly disbursements, Table 4.2 compares the two indicators in the first decade of the IDB’s operation (in millions of $US, 1961–1970). There is, at least for the IDB in this period, a large discrepancy between authorized and disbursed loans. However, it is notable that authorized loans versus commitments still inform the analysis. In particular with regard to investor confidence that derives from ratings and signals, the authorization of loans may have an impact on, for example, FDI. Future research that looks into the effects that ratings and loan authorization have on private financial sources could benefit by incorporating loan commitments into the analysis. Yet, this must be done with caution since authorized loans get disbursed over a number of years, and if the unit of analysis is country-year (as it is in this study), the results can be misleading. In this study, I analyze gross loan amounts rather than net amounts. Although the differences between the net and gross amounts of loans do

4.3 A Historical Look at RDB Loans

123

not significantly affect the results of the analysis, the central reason for examining gross loan amounts is that they represent the decision-making by the institutions. They account for the total amounts approved (authorized and disbursed) by the board of each bank to be earmarked for loan transfer and capture the (yearly) outputs of the RDBs. Net amounts include loan repayments, and therefore do not give a clear picture of the RDBs decision-making about loans. Also since net transfers include negative numbers, analysis must include the two-way stream of flows, which is not the main interest of this study. Overall, for all RDBs, a rise in lending reflects donor countries’ confidence in the institutions – and donors’ desire to ensure the RDBs’ continued relevance. Both the RDBs’ management and the donor countries would agree that for a development bank to remain relevant it has to be well funded. Thus, at a time when private global financial transactions are growing in such a way that threatens to make the RDBs marginal, increasing lending seems necessary for an institution that seeks to at least guard its influence.

4.3.1

The IDB

As pointed out in Chapter 2, the creation of the IDB was a result of a long period of lobbying by Latin American countries for such an institution. With the United States finally on board, the development bank was created. In its early years, the lending of the IDB focused not only on infrastructure but also on more long-term projects such as education and health: IDB loans in those years targeted the housing, sanitation, and education sectors, thus embracing financing challenges in sectors that were not viewed as reliably “bankable” in international banking circles (Tussie, 1995, pp. 2–3). This differentiated the IDB from any other lending financial institution to the region. In addition, since there was very little investment and support even for infrastructure projects, each additional unit of aid potentially made a big impact. As Dell (1972) states in his early analysis of the IDB, the average amount of loans approved by the IDB rose from $200 to 300 million in the first four years, to $600 million a year in 1969–1970 (p. 121). This is evident in the following graph, albeit with a lag, since the loan data in this analysis is disbursements rather than approvals (thus, the loans approved in the first four years would be disbursed, on average, over the following three to four years, and therefore the changes displayed in the Figure 4.1 are not quite as dramatic in

RDB Loans and Developing Countries

124 10000

Amount (million USD)

8000

6000

4000

2000

0 1961

1967

1973

1979

1985

1991

1997

2003

2009

figure 4.1 IDB loans, 1961–2009

the first two decades of the bank’s existence and shows a steady and gradual rise).26 Starting out with less than $300 million (US), the IDB now lends close to $60 billion annually (US) to Latin American countries. As Figure 4.1 demonstrates, the gradual rise in lending was interrupted in the second half of the 1980s. This occurred during lengthy discussions over the Seventh Replenishment of the bank when IDB officials and the US administration disagreed (during the Reagan-Bush presidency, with James Baker as Secretary of the Treasury). During these negotiations, the US government was as close as it has ever been to withdrawing support from the IDB, and that episode marked the most contentious period in US/IDB relations.27 The second noticeable change comes in the wake of the Asian Financial Crisis of 1997, when an increase in the bank’s role was approved. At its peak output of this period, IDB’s lending reached

26

27

“An examination of the figures . . . shows that, up to the end of 1970, the IDB had authorized loans to a total value of $4,068.6 million, against which it had disbursed $2,152.6 million. This appears to imply an average time lag of three to four years between authorizations and disbursements . . .” (Dell, 1972, pp. 122–124). For more discussion see Babb, 2009.

4.3 A Historical Look at RDB Loans

125

close to $100 billion (US). Evidently, the sharp rise in the amount of loans disbursed by the bank after 1991 coincides with the end of the Cold War; it appears that the economic and political changes of the 1990s had a positive impact on the amount of loans disbursed by the IDB. However, along with this distinct surge in overall lending, the IDB, like all multilateral financial institutions, was the target of growing global criticism in the 1990s. The magnitude of global financial changes and the end of the Cold War prompted (and allowed for) an increase of public scrutiny of IFIs, including the RDBs. The center of the controversy surrounding the IFIs was the growing concern about the mounting debt accumulated by developing countries and the inability of some of these countries to meet the loan payments that were now maturing. Therefore, the accelerated increase of loan disbursements was met with skepticism about the likelihood that the “development bank” model would succeed in meeting its poverty alleviation and development objectives.28 Given their institutional model, which requires loan disbursements (“getting the loans out”), critics have argued that the rise of FDI at the end of the Cold War may have induced greater loan amounts, rather than the other way around. 4.3.2

The Asdb

In a similar trend to the IDB, the AsDB’s loan disbursements to its regional members have increased significantly over the years (see Figure 4.2). Starting its operations in 1968, with lending only slightly over $40 million (US), the AsDB now lends about $55 billion (US). Like the IDB, until the mid-1980s the increase in the AsDB’s total lending was steady, but gradual. At the end of the 1980s and in the 1990s loan disbursement experienced a steep ascent. Figure 4.2 demonstrates that the AsDB increased its activities in the aftermath of the Asian Financial Crisis in 1997, when it undertook a significant role in helping the struggling Asian economies – in cooperation with the WB and IMF (multiple accounts claim that the AsDB’s policies to accelerate economic recovery after the crisis were more central to the region’s economy than those undertaken by the global institutions).29 This increased role translates, in Figure 4.2, to the spike shown between 1997 and 1998. With significant (hegemonic) 28 29

See Easterly, 2001 and 2006. See Standard & Poor’s Multilateral Financial Institutions Report, 2000 and 2009.

RDB Loans and Developing Countries

126 150000

Amount (million USD)

125000 100000 75000 50000 25000 0 1961

1967

1973

1979

1985

1991

1997

2003

2009

figure 4.2 AsDB loans, 1966–2009

involvement of two nonborrowing members – the United States and Japan – the AsDB would be a likely candidate for adopting policies that resemble donor-dominated IFIs such as the WB and IMF.

4.3.3

The Afdb

Of the four RDBs examined in this book, the AfDB espouses the most active involvement and control by the regional (borrowing) member countries. There is also more controversy surrounding the AfDB than any of the other banks. First, its relative prominence as a source of finance in Africa has been volatile, as is reflected by fluctuating credit ratings resulting from (or driven by) weak donor confidence (both private and state). Second, the AfDB has generated a reputation for corruption and for a staff lacking essential professional skills. Third, compared to the WB operations in Africa, the AfDB has been marginal, and according to some critics, it has been ineffective.30 These observations should be interpreted within the “African context”: overall, Africa is much poorer than the other developing regions. Indeed, developing 30

The authors of the IFIAC Report make the point that the AfDB may be beyond reform, and that the World Bank should assume its responsibilities.

4.3 A Historical Look at RDB Loans

127

Africa is an increasingly difficult task, even for an institution that has abundant funds, support, and expertise. Mired in civil and ethnic wars, the HIV/AIDS epidemic, and political and economic instability, donors are often wary of the challenge given the grave risks and the unlikely – at least in the short run – payoffs (in the form of observable success in development, in addition to financial returns). Since investment in Africa appears more risky, IFIs are faced with a dilemma surrounding the objective and success of their loans. For that reason, in recent years, Africa has become the center of global efforts that encourage aid in the form of grants and charity, rather than loans. The following graph depicts the loan disbursements (gross) made by the AfDB throughout the years. This graph illustrates the stark difference between the AfDB and its peers: a small but steady increase in loan disbursements in the 1970s and early 1980s, followed by a sharp increase in the late 1980s (as some Western countries and Japan join the bank’s Board in 1982), peaking between 1991 and 1995 at close to $1.5 billion, but then reversing direction, with loan disbursements quickly falling (with a total of a little over $500 million in 2000). To examine whether the process of loan repayment by borrowers – the servicing of the debt – affects this decline in lending in the 1990s, Figure 4.4 illustrates the AfDB’s net loan disbursements. Net loan amounts are a reflection of the reciprocal relationship between the bank and its borrowers. Overall, the gross and net loan amounts are similar in their trajectory: the net amounts peak in 1991–1992 at $1.2 billion, after a steep climb beginning in 1985. However, they decline rather quickly, more so than when we look at the gross amounts (this suggests an expansion of debt servicing). After 1997 the balance of loans becomes negative when borrowers’ debt repayments are higher than loans being disbursed. It should be noted that Figures 4.2–4.4 represent an aggregate of the borrowing members, and does not take into consideration fluctuations between countries, so we cannot draw conclusions about individual countries.31 Since the debt servicing is a product of past loans, the

31

For example, the AfDB’s net loans to Tunisia, one of Africa’s “most developed” (according to HDI rankings) were positive in 1998 and 2000, but overall the positive and negative are close to even. Sierra Leone, one of the poorest countries in the region (lowest ranking in the HDI) had a negative net loan balance from 1984 to 1994, after which it did not receive more loans.

RDB Loans and Developing Countries

128 2500

Amount (million USD)

2000

1500

1000

500

0 1967

1973

1979

1985

1991

1997

2003

2009

figure 4.3 AfDB (gross) loans, 1967–2009

1400 1200 1000 800 600 400 200 0 1969 1972 –400

1975 1978 1981 1984 1987

1990 1993 1996 1999

year

figure 4.4 AfDB (net) loans, 1966–2000

repayments would not substantially burden countries if they had been able to advance their economies using this financing. Taking measures to control (or reduce) the debt of countries that would be crippled by it should be a central concern for development agencies – and

4.3 A Historical Look at RDB Loans

129

the RDBs.32 However, the question remains whether under these parameters of lending, the RDBs are capable of operating as banks while alleviating poverty.

4.3.4

The EBRD

The EBRD began lending to Eastern and Central Europe in 1991. Its mission – and guidelines for lending – clearly diverges from that of the three older RDBs: it espouses political objectives to support the states it lends to in transitioning to market economies and democracies. Thus, unlike its counterparts, who over the years have disbursed countless loans to governments regardless of their regime types and economic philosophies,33 the EBRD has set itself apart by committing to clear political goals in its Articles of Agreement. Aggregate EBRD loan disbursements to its borrowing members since 1991 are seen in Figure 4.5. Evidently, the EBRDs budget – and available funds for loans – has been rising. Since specific data for the years 1991–1994 is not available, the total amount of loans for these years was divided equally between those four years, for estimation purposes. We can assume, however, that real disbursements increased steadily between 1991 and 1994. This assumption is based on the EBRD’s budget report for 1991– 1994, which describes a yearly increase in loan amounts during those years.

32

33

A negative net would then suggest that the bank is not mounting more debt on countries, and that repayments surpass disbursements; this could be indicative of responsible lending policies or limited funds. During the Cold War, loans were presumably disbursed in support of Western spheres of influence – RDBs did not officially state any preference for regime types in their loan approval process. Since many of the countries borrowing from RDBs during those years were not democratic, and yet they received loans, it is plausible that Cold War politics trumped preferences for democracy (anecdotal accounts support this, as well as the significant developments that led to the RDBs creation in the 1960s). The statistical analysis in Chapter 5 includes a Cold War dummy variable, as well as interaction effects between Cold War and Polity Index, that tests these assumptions.

RDB Loans and Developing Countries

130 20000

Amount (million USD)

15000

10000

5000

0 1991

1997

2003

2009

figure 4.5 EBRD loans, 1991–2009

For example, the 1995 annual report provides the following revalued figures (in ECU million)34 for loan commitments in 1991–1994: 1991 1992 1993 1994

21 million ECU 728 million ECU 1,530 million ECU 1,649 million ECU

Between 1997 and 1999 the EBRD experienced a decrease in the total amount of loans, although total loan disbursement still remained above 1.5 billion euro. The reduced lending during this period was a product of the mounting criticism and scrutiny of the EBRD’s policies. After the first few years of operation, the EBRD was targeted by journalists and nongovernmental organizations, which heavily criticized its activities and conduct. The central complaints included the absence of transparency and claims that the bank’s loans were competing with private institutions, questioning the need for a high cost public institution.35 Moreover, the 34

35

Revalued figures represent subsequent changes due to exchange rates, cancellations, syndications, or restructuring. The difference in calculating the figures in those years and varying methods of accounting explain the absence of accurate total values available for those years. See, for example, Head, 2004.

4.4 The State of Development

131

EBRD’s creation at the end of the Cold War was the pretext, according to critics, for a so-called public institution that mainly operates in Russia.36 Critics further sought to expose the EBRD’s secretive decision-making and reveal its interest rate policies. At the very end of the 1990s the bank’s reputation started improving. It began exercising more transparency (although interest rate policies are still not public) and engaged in a more aggressive public relations campaign. In its public disclosure of the impact of the transition, the EBRD publicized that from 1996 to 2000 it accounted for the following medium- and high transition impact on evaluated projects:37 1996 83%

1997 70%

1998 74%

1999 74%

2000 81%

The evaluation of transition impact includes promotion of privatization, the development of skills, the encouragement of competition, and support for market expansion. Since Russia is a central borrower of the EBRD, the bank explains stagnation in transition impact from 1997 to 1999 as a result of the Russian financial crisis. In making an effort to foster a positive image, the bank has defended itself by arguing that it provides an additional value to the development process of the borrowing members (what the EBRD officially refers to as “additionality”), thus distinguishing itself from other lending institutions in the region and making its operations unique.38

4.4

the state of development

Determining whether RDBs’ lending is driven by the need of borrowing members to secure credit or by financial considerations that ensure safe 36

37

38

This critique still holds. Various bank officials referred to the EBRD as “the Russia bank.” In the year 2000, the EBRD’s total operating assets were more than €7.5 billion and undrawn commitments of more than €4.6 billion, out of which operating assets in Russia were almost €1.5 billion and undrawn commitments more than €0.73 billion. See EBRD 2000 Annual Report (p. 48) for detailed transition impact ratings. “Project evaluation involves the assessment of operations once investment has been completed. This normally occurs one to two years after full disbursement has taken place. Performance ratings are assigned according to how the project has fulfilled the EBRD’s mandate. The ratings focus on impact on the transition process and the application of sound banking principles. They also assess ‘additionality’ and environmental performance. In addition, technical cooperation operations are evaluated, applying the same means of assessment that are used for EBRD investments.” Ibid. More on this issue will be discussed in Chapter 6. Interviews with EBRD officials and economists, January 2002.

132

RDB Loans and Developing Countries

investment informs the central puzzle of this book – the dichotomy of a development agency and bank, both embedded in a single institution. The independent variables that are included in the analysis are selected to provide a snapshot of the development level of borrowing member countries, as well as their relationship with global and/or regional hegemon(s). Measures of private flows (FDI), per capita GDP, trade volume, health (Infant Mortality Rate), and education (primary school enrollment) all inform us about countries’ socioeconomic circumstances. These benchmarks are important because they help place the loans disbursed by the RDBs in the context of the level of development of member states. The following section overviews a selection of the variables, demonstrating overall global and regional trends.

4.4.1

FDI

A common measure of private sector awareness, Foreign Direct Investment to developing countries has become an increasingly central component of many developing countries’ economies.39 Since the late 1980s private foreign investment in developing countries has grown exponentially, significantly surpassing any form of public – bilateral and multilateral – investment in developing countries. Figure 4.6 illustrates the steady increase of FDI to developing countries since 1970. It is evident that since the end of the Cold War (1991) there has been a surge in FDI flows to poor countries. The rise in FDI to developing countries that are also borrowers of the RDBs alters the economic and financial environment in which the RDBs operate: an increase in FDI to the same countries that receive RDB loans changes the relative significance of those loans.40 Thus, as the amounts of FDI to developing countries increase, RDB loans as a percentage of the total transfers decrease. In explaining this phenomenon, some officials at the RDBs claim that their lending contributes to investor confidence that encourages private investment and generates higher amounts of FDI. On the other hand, critics claim that an increase in FDI is not a result of RDB presence, and even if it were, then by increasing lending to countries that can now attract FDI, 39

40

And countries that were not able to easily attract FDI made efforts to adopt policies that could potentially bring them more FDI. In short, whether increased or not, attracting private investment in the form of FDI has become a central concern for governments in developing nations. See also Table 3.5 for a comparison.

4.4 The State of Development

133

500000000

Amount (million USD)

400000000

300000000

200000000

100000000

0 1970 1974 1978 1982 1986 1990 1994 1998 2002 2006 2010

figure 4.6 FDI to developing countries, 1970–2000

RDBs may very well be straying from their mission. The statistical analysis in the next chapter includes the lag of FDI to examine whether FDI precedes loans made by RDBs.41 There are differences in the amounts of FDI to each developing region. The reasons for this variation include: (1) the level of economic development; (2) institutional and political stability; and (3) financial opportunities (such as resources) – all of which are considered by private investors when making decisions. Presumably, from the RDBs’ perspective, successful economic development is characterized by, among other things, an ability to attract investment, so an increase in FDI flows is welcome and should lead to changes in RDB loan policy. The breakdown in Figure 4.7 of FDI to each of the four regions that correspond to the RDBs in this study illuminates the differences between the regions. In Africa, FDI had a rocky climb in the past four decades. Despite irregularities, however, Africa began attracting a steady increase in private investment in the late 1980s. By the late 1990s FDI is over fifteen times the amounts invested privately in Africa in the early 1970s. Yet, even at its height in the 1990s, FDI to Africa is still a relatively

41

This analysis in Chapter 5 demonstrates an association – and not a causal relationship.

RDB Loans and Developing Countries

134 500000000

Amount (USD)

400000000

300000000

200000000

100000000

0 1970

1974

1978

1982

1986

1990

1994

1998

figure 4.7 FDI to Africa, 1970–2002

small share of the total FDI to the developing world: in 1998 the total FDI to developing countries was almost $190 billion while total FDI to Africa was only $7.7 billion.42 The flow of FDI to Latin America corresponds to the general increase in FDI to developing countries. A steady and steep rise in the 1990s reflects post–Cold War openness that can be identified in other regions. However, in terms of the percentage of the total FDI, the share of Latin America and the Caribbean has been decreasing from more than half in 1970 to about one third in 2000; Latin America’s portion of FDI declined at a time that Asia was becoming more attractive for private investors. FDI to Latin America grew steadily and peaked in 1999 at over $110 billion, but then began declining (in the year 2000, FDI stood at a little over $86 billion). When analyzed in conjunction with IDB loans, FDI to Latin America is particularly interesting for this study. The IDB, the oldest of the four RDBs, has been praised especially for its early work in Latin America. In the 1960s and 1970s, the IDB was a central development agent in Latin America, making loans and financing projects that would

42

See supplemental website (www.ruth-ben-artzi.com) for a breakdown of percent of FDI to each region for selected years.

4.4 The State of Development

135

Amount (million USD)

150000

100000

50000

0 1970 1974 1978 1982 1986 1990 1994 1998 2002 2006 2010

figure 4.8 FDI to Latin America and the Caribbean, 1970–2009

otherwise not get funded. The data on FDI show that even if Latin America experienced growth and development see figure 4.8, its shares of total global FDI declined. Asia, like Latin America, experienced a steady growth in FDI starting in the 1990s. But unlike Latin America, Asia’s portion of total FDI to developing countries has been growing steadily since 1970 (see Appendix 1 for more details). Figure 4.9 shows that Asia received little FDI in the 1970s, with a significant increase in the 1980s, and then the typical steep increase in the 1990s (with a dip right after 1997 due to the Asian financial crisis). At its height in 2000, the Asian region attracted more than $140 billion, significantly more than Latin America. Thus, from a share of 29 percent of the developing world’s FDI in 1970 to a 54 percent share of the total FDI to developing countries in 2000, Asia has consistently increased its appeal to private investors. Understandably, Eastern and Central Europe were mostly closed to private investment until the late 1980s. But immediately following the collapse of the Soviet Union, more FDI entered the region, first gradually, and since 1995 at an accelerated pace. Because of its relative size and its more recent addition to the group of emerging markets, developing Europe

RDB Loans and Developing Countries

136

250000

Amount (million USD)

200000

150000

100000

50000

0 1970 1974 1978 1982 1986 1990 1994 1998 2002 2006 2010

figure 4.9 FDI to Asia, 1970–2009

shares a very small portion of the FDI to developing regions (even so, in relative terms, the percentage of total FDI to developing countries that developing Europe attracted grew from 1 percent in 1970 to 10 percent in 2000; see Figure 4.10). Furthermore, the financial crisis in Russia in the late 1990s had a negative impact on private investors leaving Eastern and Central European markets volatile, as evidenced by FDI flows. In sum, a closer look at private flows to developing countries helps determine whether the RDBs disburse loans to borrowing countries with more access to private capital, or whether their lending policies target the poorest and most needy of their borrowing members. The analysis of an association between FDI and loans is a first important step: although developing countries that receive higher levels of FDI are not unaffected by poverty (and they very well may need development assistance), it is reasonable to assume that those developing countries that are more attractive to private investors are also more likely to have better prospects for repaying loans and that RDB loans made to them are arguably more likely to compete with available private financing.43 43

In this book, I do not analyze the sectors to which funds are directed. Further research examining the sectors would complement this study to get a more detailed picture of the destination of the loans. Moreover, as project-level data becomes available (see Tierney et al., 2011), assessing the likelihood that projects will receive private investment could become easier.

4.4 The State of Development

137

Amount (million USD)

200000

150000

100000

50000

0 1990

1993

1996

1999

2002

2005

2008

figure 4.10 FDI to developing Europe, 1970–2009

4.4.2

Per Capita Income and Health

In a similar fashion to FDI, the GDP per capita of the countries that borrow from the RDBs has changed significantly since the RDBs began lending. A measure of a country’s relative wealth, a lower GDP per capita is more closely associated with higher rates of poverty and underdevelopment, while an increase in GDP per capita is an indication of growth. Given the rising costs of living and currency fluctuations, the analysis interprets the values of GDP per capita in relation to other variables. For this reason, this analysis mostly focuses on the GDP per capita of the each borrowing member country relative to one another, and it does so by testing panel time-series data (in Chapter 5) that include GDP per capita over a period of time, and between countries (to highlight the direction of change and the comparison between borrowers, in association with loan disbursements). It is important to keep in mind that a rise in GDP per capita does not necessarily indicate alleviation of poverty. GDP per capita increase points to an improvement in the overall economic conditions, but it does not reflect changes in economic disparity within countries or show whether certain sectors reap greater shares of the benefits of economic growth.44 44

I do not look at these factors because my analysis treats countries as units, but further research that digs deeper into changes within countries would certainly improve our

138

RDB Loans and Developing Countries

Between 1967 and 1997 the overall GDP per capita of African countries grew by more than 30 percent. For many of these countries this can be significant (being an average figure, there is obviously variation between the borrowing countries).45 In Asia, the changes in GDP per capita have been more dramatic (starting at a point that is much higher than that of Africa). With an average annual per capita GDP of a little over $900 in 1966, developing Asia’s average GDP per capita rose to over $2,000 by 1997 – more than a twofold increase.46 Even when averages are skewed by the fast-paced growth of the “Asian Tigers” up until the late 1990s, Asia’s overall GDP per capita values have been growing.47 In Latin America, GDP per capita has been steadily rising – not as dramatically as in Asia, but the baseline in 1961 was much higher: $2,000 per capita GDP (compared with $900 in Asia). Overall, in Latin America, like in Africa and Asia, the GDP per capita has risen progressively throughout the last five decades. Finally, GDP per capita in Eastern and Central Europe has, for the most part, been on the rise. It is not surprising that after the end of the Cold War, former Soviet Bloc countries that opened their economies to the West have enjoyed substantial growth in GDP.48 But at the same time, the transition into market economies has been laden with obstacles and poverty rates have soared as well in some countries. Like GDP per capita, health and education measures in developing countries are better today than they were a few decades ago. Infant mortality rates are a benchmark to the quality of healthcare in the most basic of services – the birth and care of infants (up to one year). Typically, countries with higher infant mortality rates are also poorer. Western countries all have very low rates of infant mortality. This is,

45

46

47

48

understanding and bolster these conclusions. This would include project-level analysis, sector analysis, and case studies. In the next section, featuring samples of countries from each region, this idea will be further developed. Qualifying this indicator with the given availability of the data, it is important to note that, for example, data for Afghanistan is sparse and unreliable. Poor quality data from Asia’s poorer countries skews the averages upwards. For more on the “Asian Tigers” see Bruce Cummings “The Origins and development of the Northeast Asian political economy”; Krugman, 1994; Radelet and Sachs, 1997. GDP per capita data show that despite some volatility following the crisis, recovery was fairly steady. See Easterly (2001) for the argument that growth does not necessarily lead to development.

4.4 The State of Development

139

table 4.3 GDP per capita and infant mortality rate, by region

Africa

GDP per capita Infant mortality rate Asia GDP per capita Infant mortality rate Latin America GDP per capita Infant mortality rate Developing GDP per capita Europe Infant mortality rate

1971

1981

1991

1997

733 137 1214 96 2503 79 – –

918 119 1555 83 2900 57 – –

970 103 2066 66 2888 42 2179 36

991 93 2116 56 3234 35 2591 26

therefore, an important indicator for assessing loan-recipients’ level of development and placing RDBs’ policies in the context of their borrowers’ level of poverty. Just as GDP per capita in Africa has been rising, infant mortality rates in Africa have decreased steadily in the past five decades. Although still significantly higher than the rates in developed countries and developing countries in other regions, infant mortality has shrunk from over 140 deaths per 1000 births in 1967 to 90 deaths per 1000 births in 2000 – a decrease of more than one third. This is certainly substantial, although still quite high for the early twentyfirst century. The infant mortality rate in Asia has also been declining. Although still much higher, on average, than that of developed countries, it dropped by over 30 percent by the late 1990s, and is substantially lower than the infant mortality rate in Africa (although Africa’s starting point is also higher, the decline in Asia is more significant than in Africa). The decline in infant mortality rates in Latin America is substantially more significant than in Asia. Finally, infant mortality rates in Eastern and Central Europe have also steadily declined over the past two decades. From nearly forty deaths per thousand births in 1991, the rate of infant mortality has declined to a little over 20 per thousand in 2000. It is also noteworthy that like GDP per capita, infant mortality rates in developing Europe reflect higher levels of development than those we observe in the other regions. Most countries in developing Europe are industrialized, where the quality of health-care has been better than in the rest of the developing world.

140

RDB Loans and Developing Countries

Table 4.3 summarizes aggregates of GDP per capita and infant mortality rates, for each RDB, in the last three decades of the twentieth century. In sum, the changes in these indicators of development over the years offer a context within which we can analyze the loans made by the RDBs, and help place them on the continuum of development agency/bank.

4.5

selected countries

A closer look at RDBs lending to specific countries offers another perspective on the banks’ decision-making. I selected two countries from each region based on the Human Development Report’s (HDI) ranking (2001). The highest and lowest ranking HDI country from each borrowing region of the RDBs are included here to test the differences in the amount of loans they each receive from the corresponding RDB. By assessing the net loans to each of the countries the following analysis takes into account the burden of loan repayment and other transfers.49 Significant differences in loan disbursements for the “richest” and “poorest” countries of each region, provides further evidence that the RDBs engage in strategic lending that favors financial parameters over those of need (and poverty). The Figures 4.11 and 4.12 of IDB loans to Argentina (HDI rank 34) and Haiti (HDI rank 134) show stark differences. There is an unambiguous difference in the amount of loans these two countries receive: while loan disbursements to Argentina are much higher in absolute terms, its net is also positive and on the rise. In contrast, Haiti received loans only in the initial years of the bank’s operations. The loan amounts to Haiti were much smaller,50 but more importantly; Haiti has mostly had a negative balance of payments with the IDB. Thus, while Argentina enjoys a positive net for most of the years that it received loans from the IDB, it seems that for Haiti, IDB lending has been a financial burden rather than a relief. The AsDB’s net loans to Singapore (HDI rank 26) and Bangladesh (HDI rank 132) are seen in Figures 4.13 and 4.14.

49

50

See supplemental website (www.ruth-ben-artzi.com) for the graphs of the gross amounts of loans to these countries. Taking into account its much smaller size, this is understandable. In the statistical analysis in the next chapter the proportion of loans to the borrowing member’s GDP per capita is taken into account.

4.5 Selected Countries

141

10000

Amount (million USD)

8000

6000

4000

2000

0 1961

1967

1973

1979

1985

1991

1997

2003

2009

figure 4.11 IDB net loans to Argentina, 1961–2009

10000

Amount (million USD)

8000

6000

4000

2000

0 1961

1967

1973

1979

1985

1991

1997

figure 4.12 IDB net loans to Haiti, 1961–2009

2003

2009

RDB Loans and Developing Countries

142 150000

Amount (million USD)

125000 100000 75000 50000 25000 0 1967

1973

1979

1985

1991

1997

2003

2009

figure 4.13 AsDB net loans to Singapore, 1967–2009

150000

Amount (million USD)

125000

100000

75000

50000

25000

0 1967

1973

1979

1985

1991

1997

2003

figure 4.14 AsDB net loans to Bangladesh, 1967–2009

2009

4.5 Selected Countries

143

2500 2250

Amount (million USD)

2000 1750 1500 1250 1000 750 500 250 0 1967

1973

1979

1985

1991

1997

2003

2009

figure 4.15 AfDB net loans to Tunisia, 1967–2009

There are significant differences here as well: while up until the mid1980s Singapore received substantial positive net loan amounts, in the late 1980s and early 1990s the net loans were negative and in the second half of the 1990s they stopped completely. Given Singapore’s economic development and success, this observation reflects initial investment on the part of the AsDB, followed by some years of overall negative net (for Singapore), culminating with Singapore’s independence from the AsDB. At the same time, while Bangladesh’s net loans were negative throughout the 1980s and most of the 1990s (albeit small), there is a sudden surge in 1999 and 2000. This surge seems to be a reaction to the Asian financial crisis and it does not reflect a sudden significant increase in loan amounts – they remain relatively stable throughout the 1990s; rather, it is a result of a relief in debt servicing that made net amounts positive for Bangladesh. Although Bangladesh continued to receive loans in the 2000s, until 2007 the overall amounts were smaller than in the 1990s. The net loans to Tunisia (HDI rank 89) and to Sierra Leone (HDI rank 162) seen in Figures 4.15 and 4.16, demonstrate striking differences among African borrowing countries as well: Sierra Leone, the lowest HDI ranking country, stopped receiving gross loans in 1985 (see supplemental website), and has had a negative balance

RDB Loans and Developing Countries

144 2500 2250

Amount (million USD)

2000 1750 1500 1250 1000 750 500 250 0 1967

1973

1979

1985

1991

1997

2003

2009

figure 4.16 AfDB net loans to Sierra Leone, 1967–2009

with the AfDB since that same year (negative net loans). At the same time, Tunisia has enjoyed a positive balance of loans, also receiving substantially larger amounts. The decrease in the loan amounts in the late 1990s, coupled with a negative balance (which suggests loan repayments by Tunisia), points to the AfDB’s financial woes and the challenges it faces in securing capital. Lastly, net loans for Slovenia (HDI rank 29) and Uzbekistan (HDI rank 99) suggest a different approach undertaken by the EBRD thus far (see Figures 4.17 and 4.18). While Slovenia, the highest HDI ranking country in the region, began receiving loans only in 1994, Uzbekistan has been receiving a steady stream of loans (positive net) since 1992. Moreover, the overall averages of the size of loans are not vastly different (with Uzbekistan receiving an average of about $40 million, and Slovenia receiving an average of about $30 million). Of the four RDBs, the EBRD is the only one that has been giving the poorest country in its region more loans than the wealthiest country in its region. This observation can be interpreted in one of three ways: (1) that the policy of “additionality” the EBRD employs is successful in providing loans to projects that cannot be financed otherwise, and thus gives more loans to the poorer countries of the region, (2) that former

4.5 Selected Countries

145

20000 17500

Amount (million USD)

15000 12500 10000 7500 5000 2500 0 1991

1994

1997

2000

2003

2006

2009

figure 4.17 EBRD net loans to Slovenia, 1991–2009

20000 17500

Amount (million USD)

15000 12500 10000 7500 5000 2500 0 1991

1994

1997

2000

2003

2006

figure 4.18 EBRD net loans to Uzbekistan, 1991–2009

2009

RDB Loans and Developing Countries

146

Soviet Bloc countries, such as Uzbekistan, are considered sound financial investments, unlike the poorest countries of other developing regions, or, (3) that Uzbekistan’s development is politically important for EBRD members and shareholders, and so the higher investment represents political strategy by a public institution. This closer look into sample countries partially demonstrated what is obvious – that in making loans, RDBs must “think” as banks and ensure that loans are being repaid. And while even the more developed of the borrowing members may have a negative balance with the bank (e.g., Singapore), a positive net flow may mean that the bank is disbursing additional loans to conceal previous debt, thereby ensuring positive net income of the borrowing country (however, higher net loans could be a precursor for future indebtedness, when the larger loans mature).

4.6

a comparison to the world bank

When comparing the disbursements made by the RDBs to those made by the WB, the RDBs’ central role among development institutions is evident: over the years they have become more prominent in their respective regions than the World Bank. Since the RDBs ostensibly ensure greater involvement in decision making by borrowing members than in their global counterpart, a comparison of the global and regional development banks sheds some light on donor preferences.51 Figure 4.19, comparing WB and IDB loans, illustrates the IDBs’ growing presence in Latin America: since the 1990s the IDB has surpassed the WB in loan disbursements to Latin America. As far as US influence in the region is concerned, the IDB’s reputation in Latin America seems more positive than that of the WB, and the fact that the United States is a central supporter of the IDB serves US interests well.52 Thus, despite their similar lending amounts in the initial years of the IDB’s existence, in recent years the IDB has been a more pivotal development bank in the region. For example, in 2000, the IDB was the first to provide Argentina with a substantial loan at the outset of its financial crisis, and the IDB has secured an agreement with the Brazilian National Development Bank (BNDES) 51

52

Central donor countries on the RDBs’ boards are by and large the same as those on the board of the World Bank, and decisions concerning financial allocations take into account the myriad of institutions (see IFIAC Report, 2000). Based on interviews with IDB officials, OECD officials, and authors of the IFIAC Report (2000).

4.6 A Comparison to the World Bank

147

9000 8000

US $ (million)

7000 6000 5000

WB

4000

IDB

3000 2000 1000

19

61 19 63 19 65 19 67 19 69 19 71 19 73 19 75 19 77 19 79 19 81 19 83 19 85 19 87 19 89 19 91 19 93 19 95 19 97 19 99

0

Year

figure 4.19 IDB and WB gross loan disbursements to the Americas, 1961– 2000

providing it with annual support, thus securing the IDB’s involvement in the region’s largest economy. The AsDB has followed a similar trajectory and is now a major credit resource in Asia – even more than the WB (Figure 4.21 of the AsDB and WB loans demonstrates the central role of the AsDB in the wake of the Asian Financial Crisis, as the total amounts of its loans more than doubled during that period while those made by the WB hardly changed). Although at the end of the 1990s the total disbursements of the IDB and AsDB were reduced, they increased again in the last decade and have sustained levels higher than those of the WB. The AfDB, although surpassing the WB in its loans to Africa in the 1980s and early 1990s, has steadily decreased its disbursements, and by the late 1990s became a smaller lender to Africa than the WB (see Figure 4.20). This reflects the AfDB’s shaky relationship with donors and suggests that donor shareholders are more comfortable using the WB as a central development bank in Africa and have thus not granted the AfDB a greater role as a development institution in the region, as reflected by their contributions.53 The EBRD’s overall loan amounts to its borrowers are smaller than those of the WB, but given the development bank’s narrow political 53

The 2000 IFIAC Report makes this point and reiterates the importance of the WB in Africa given its conclusion that the AfDB does not function well.

RDB Loans and Developing Countries 1600 1400 1200 1000 800 600 400 200 0

WB AfDB

72 19 75 19 78 19 81 19 84 19 87 19 90 19 93 19 96 19 99

19

19

69

US $ (mill)

148

Year

figure 4.20 AfDB and WB gross loan disbursements to Africa, 1969–2000

6000

US $ (mill)

5000 4000 WB

3000

AsDB

2000 1000

83 19 86 19 89 19 92 19 95 19 98

77

80

19

19

19

74 19

71 19

19

68

0

Year

figure 4.21 AsDB and WB gross loan disbursements to Asia, 1968–2000

focus and specific agenda, this remains consistent with expectations: since the EBRD does not aim to alleviate poverty, and it concentrates on market reform and the private sector, the EBRD’s mission is different than that of the WB. For example, in response to the financial crisis in Russia, the WB significantly increased its loans to Eastern and Central Europe in the late 1990s (see Figure 4.22) (while the EBRD did not alter its lending volumes). This is quite different than the central roles played by the IDB and AsDB in the wake of the crises in their regions around the same time.

US $ (mill)

4.7 Conclusion

149

1600 1400 1200 1000 800 600 400 200 0

WB EBRD

1994

1995

1996

1997 Year

1998

1999

2000

figure 4.22 EBRD and WB gross loan disbursements to developing Europe, 1994–2000

Overall, if major donor countries are concerned with their influence over the institutions, then the increasing assistance from RDBs, at times at the expense of the WB, suggests that donors are finding different ways to exert power, perhaps with more cooperation and delegation. When it comes to the soft power exercised via these institutions, RDBs’ reputation may yield better policy results for donors than an institution in which donor power is more evident. The WB’s strong association with criticism directed at the “Washington Consensus” generates a different response to US involvement than the one associated with the RDBs, where even if the United States takes on a hegemonic role, it could be more easily diffused or masked.

4.7

conclusion

This chapter began painting a picture of the overall levels of economic and social conditions in each developing region serviced by one of the RDBs, as well some of the differences between them. The overview of the total loans disbursed to each region by the RDBs since their inception examines the general trajectories, outlining differences between the development banks. Furthermore, the snapshots of the loans made by each of the development banks to two selected countries from their region demonstrated the dilemma inherent in the institutions’ role as development agencies constrained by banking standards. An examination of the highest and lowest HDI-ranked borrowing member of each RDB demonstrates that of the four RDBs, the EBRD is the only one that lends more total dollar amounts to the lowest HDI ranking country in

150

RDB Loans and Developing Countries

its region than to the one with the highest ranking. But the reason for this could be rooted in the differences of baseline socioeconomic needs rather than that the EBRD’s policy is more development-oriented. As for the other three RDBs, the IDB and the AfDB disburse higher loan amounts to the higher HDI ranked country, while the AsDB seems to have reversed this trend in the last few years. An overview of selected independent variables reveals, as expected, that Africa stands out as a much poorer region than all the rest: its overall GDP per capita is the lowest, its infant mortality rate the highest (even at the end of the twentieth century and beginning of the twenty-first), and flows of FDI to Africa are the lowest. However, like all the other RDBs, AfDB borrowing members have experienced overall improvements in the conditions measured by all these variables. Lastly, in the final section of this chapter, a comparison to the WB establishes the RDBs’ central role among multilateral development banks. The RDBs (except for the EBRD) were designed to address poverty and development issues and all four institutions are committed to extending financing to projects and programs that would not otherwise receive credit. Thus, mapping the association between RDB loans and recipient countries’ socioeconomic conditions sheds light on the RDBs’ actual disbursements, putting them in the context of the countries that borrow. The next chapter analyzes the changes within countries over time, and compares the RDBs to each other, as well as the countries within their respective regions: it statistically analyzes the socioeconomic variables described here, and tests whether (and to what degree) they are associated with loans.

5 Banks or Development Agencies?

Every individual necessarily labours to render the annual revenue of the society as great as he can. He generally, indeed, neither intends to promote the public interest, nor knows how much he is promoting it . . . He intends only his own gain, and he is in this, as in many other cases, led by an invisible hand to promote an end which was no part of his intention. – Adam Smith, An Inquiry into the Nature and Causes of the Wealth of Nations, book IV, ch. II, 1776

With growing volumes of financial flows to developing countries, how do we know whether the RDBs are fulfilling their stated mission and not competing with private capital? When all borrowing members in their respective regions are developing countries, what drives RDBs’ decisionmaking? Are they driven by the borrower’s need for credit (even if risky for the RDB)? Or, do they prefer safer investments where they are more likely to get paid back, but also where there is a higher likelihood of other competing credit sources? Further, do the political interests of powerful donors, or the recipient’s level of democracy drive loan disbursements? This chapter provides the first step in answering these questions by evaluating RDB loans according to the economic, social, and political contexts of their borrowing members. The previous chapter established a basis for the aggregate, regional level lending by the RDBs and examined the indicators for a select group of countries in each region. In this chapter, I statistically locate the RDBs’ policy (represented by loan disbursements) on the development institution to bank continuum, as represented by select socioeconomic indicators. Furthermore, in this chapter, I test whether hegemon(s) of the RDBs use their power to influence loan making. Finally, I examine the extent to 151

152

Banks or Development Agencies?

which the Cold War is associated with the level of loans that borrowing member countries receive. International Financial Institutions (IFIs), including the RDBs, have been at the subject of policy debates concerning their role. The central themes of the controversy are (1) whether they meet the objectives they were created to fulfill as public institutions, which include their contributions to poverty alleviation and development, and (2) whether IFIs serve the national interests of their largest donors (hegemons), making their lending practices commensurate with donor interests rather than with institutional objectives (to the extent that they differ). The comparison of the four RDBs in this analysis highlights the ways in which different hegemonic configurations and/or regional-borrowers’ involvement affect lending. We can then draw inferences about how RDBs execute their main task, and whether institutional mechanism, hegemonic involvement, or both, guide each development bank’s policy and its ability to achieve the expressed intentions of its founders. To evaluate whether the RDBs operate more like banks or development agencies, the analysis includes measures of borrowing countries’ economic and social development. If the banks are closer to the development-agency end of the spectrum, countries with lower social and economic levels of development will receive more loans from the RDBs. If the RDBs’ performance is more bank-like, I expect that countries whose economies are stronger and social indicators more promising (and therefore arguably safer investment venues) would receive more loans.1 And next, I test the extent to which the RDB’s carry out a political agenda based on Polity scores and the extent to which central shareholders hold sway over the banks’ agendas. Whether an RDB’s policy falls on the bank or development-agency spectrum could be a product of the development banks’ own preferences, or those dictated by hegemon(s). Political motivations for aid are not necessarily associated with meeting development goals or prioritizing banking concerns. Political interests could manifest in setting an agenda that favors allies or supports allied countries that display certain political ideologies. For powerful donors to exert their preference over the policies of the RDBs, an institutional mechanism that enables donors’ hegemonic influence must be in place. The political component in this study examines whether an association exists between hegemon(s) and the lending output of the RDBs through their UN General Assembly voting. Furthermore, to measure the impact of the Cold War and its aftermath, I include Polity 1

For more on (private) bank lending to developing countries see Jeanneau and Micu, 2002; Goldberg, Dages, and Kinney, 2000; and Arena, Reinhart, and Vazquez, 2006.

5.1 Analysis

153

scores in the analysis to control for RDB loan preference based on the level of democracy. The findings of this chapter demonstrate that the poorest countries do not get the bulk of loans. Except for the AfDB, loans by the RDBs are positively associated with FDI, and at the IDB and EBRD they are also positively associated with GDP. These RDBs are then more likely to lend to borrowers whose economies fare better. For the most part, they are also less likely to lend to countries where indicators for health and education point to poorer conditions. Of the four RDBs, only the EBRD prefers lending to democracies. Further, the AfDB is the only RDB whose lending to borrowing members that are more or less democratic is a factor of the Cold War: borrowers with higher Polity Index scores were more likely to receive loans during the Cold War than in its aftermath. Finally, the influence of hegemon(s) differs from one bank to another: the United States is strongly involved at the IDB, Japan influential at the AsDB, and the AfDB appears to be the most independent of hegemonic influence. The EBRD, with multiple hegemons, seems to make loans regardless of borrowers voting records at the UN.

5.1

analysis

I test the relationship between these predictors and the loans made by banks using ordinary least squares regression with (i) country-fixed effects to control for the average differences across countries in any observable or unobservable predictors (reducing the threat of omitted variable bias); and (ii) a lagged dependent variable to correct for time trends or autocorrelation in the data.2 Each of the RDBs is analyzed separately because I expect the variables to have different effects for each of the banks. For example, the United States, being a single hegemon at the IDB may influence the IDB’s policy differently than it would at the EBRD, given the EBRD’s multiple hegemonic structure. The comparison of the four institutions presupposes that there are inherent differences between them that are manifest in the effects the variables have on the loans made by each RDB. Finally, since the banks have operated for different time periods, by analyzing them separately I maximize the utility of the data. For robustness, I pooled the data and tested all the models on the pooled 2

See Beck and Katz (1995) for more on issues in estimation in time-series cross-section models; and Keele and Kelly (2005) on appropriate use of lagged dependent variable (LDV) in OLS regressions.

154

Banks or Development Agencies?

data sets. The results of the pooled data are not substantive or significantly different and are included in the supplemental website.3 The analysis covers the years each of the RDBs has been lending through 2009: for the IDB, 1961–2009; for the AfDB, 1966–2009; for the AsDB, 1967–2009; and for the EBRD, 1991–2009. For each bank, all regional borrowing countries are included in the data set: 28 borrowing members of the IDB, 53 borrowing members of the AfDB, 41 borrowing members of the AsDB, and 26 borrowing member countries of the EBRD. As discussed in Chapter 4, there are compelling reasons for analyzing disbursed loans. Although selection bias could be a problem with analyzing only these loans, since it does not take into account all the loans that recipients requested, the way in which the RDBs conduct their business makes it impossible to assess requested loans. Much of the deliberations and decisionmaking take place outside the institutions’ “formal space” and thus actual formal denials of loan requests are rare. Since the process of loan applications and selection is, in a way, predetermined (countries apply mostly when assured that loan application will be approved), analyzing actual loans disbursed is the most informative metric given available information.4 The potential for endogeneity poses difficulties when analyzing loans disbursed by RDBs to countries that are members of these institutions and seek those loans. One of the ways I address this is by lagging all the independent variables by one time period, as has been done in previous studies.5 In addition, I control for country fixed-effects, which Simmons and Hopkins (2005) argue attenuates selection bias. Addressing the endogeneity problem using instrument measurements offers another solution, but is not possible in this case. It is very hard to find appropriate 3

4

5

See also Kam and Franzese (2007, p. 103) for an explanation of the differences between separate analyses and pooled samples. Overall, there are advantages to both and the authors suggest that choosing one over the other should be driven by theory; and for this study, separating the analysis for each of the banks is preferable from a theoretical standpoint. In the previous chapter, in Chapter 6, and in the supplemental website, I discuss concessional loans (“soft loans”) and demonstrate their relative share in the banks’ lending. But it is important to reiterate that there are numerous reasons for including only nonconcessional loans in the statistical analysis. In addition to the fact that soft loans have been disbursed for fewer years and do not exist at all for the EBRD, concessional funding is separate and their share of the total lending by the RDBs is relatively small. Thus, I test the RDBs’ central lending mechanism – and the one that is their main operation: the nonconcessional lending window falls under the purview of the banks’ Articles of Agreement and should satisfy their objectives. See, for example, Copelovitch, 2010; Dreher and Vaubel, 2004; Stone, 2002; Thacker, 1999; and Arena, Reinhart, and Vazquez, 2006.

5.1 Analysis

155

instruments for the independent variables in this study (see Bartels, 1991 for discussion of the problems associated with the misspecification of instrumental variables).6 Like the IMF, RDBs operate according to a fiscal-year timetable and thus a one-year lag reflects RDB decisionmaking timetable (Copelovitch, 2010, p. 86). Although endogeneity concerns obviously still exist despite my steps to minimize it, this analysis examines association and not causation, somewhat neutralizing endogeneity as a problem. Establishing association in the statistical analysis, along with the qualitative evidence in Chapter 6, forms a valuable foundation that addresses the questions posed in this study. Demonstrating association between the loans and the independent variables does not inform us about the direction of the association (e.g., do loans follow FDI or does FDI follow loans?), but examining this relationship over time allows us to draw some conclusions about this question (see Table 5.1). For example, when FDI amounts are higher or rise in a particular country, then we could expect loans to that country to decrease in accordance with the RDBs’ mission, given that private financial resources are available.7 If loans do not decrease when FDI is on the rise, then it is more likely that RDB lending could be competing with private financing. The tables in Section 5.2 include all the variables described in Chapter 4. The statistical analysis below examines economic, social, political, and combined models for each of the RDBs. All the models include the lagged dependent variable. The combined models include the economic, social, and political variables to see if the significant effects found in the separate models still hold once we take into account all the effects, combined. In these combined models, I try to balance three issues: (1) include variables from all three areas (economic, social, and political); (2) keep models parsimonious (excluding unnecessary/insignificant variables, as demonstrated in the separate models); and (3) minimize and ideally eliminate multicollinearity by excluding the highly collinear variables (tested by postregression VIF), and/or including them in alternate combined models.8

6 7

8

See also Sovey and Green, 2011. While there could very well be other factors involved, this analysis assumes (based on the RDBs’ own Articles of Agreement and Mission Statements) that loans would not compete with other financing opportunities. Subsequent studies should look specifically into countrycases where, for example, FDI rises, and examine whether there are any changes in the loans if the amounts are not reduced (thus, if both variables are associated positively, we need to make sure they are not competing with each other). I inspect for collinearity by checking the correlation and variation inflation factors (VIF) for these variables. In cases where VIFs are high, I separate the variables into two models.

156

Banks or Development Agencies? table 5.1 Variables1

Variables* Loans per capita (DV) GDP per capita Trade FDI Primary Education Infant Mortality Polity Index Post–Cold War USA, Japan, France, Germany, UK Polity*Cold War USA*Cold War

Description log** (data from RDBs, OECD) Divided by 1000 Trade volume Divided by 1000 Percent of children who complete primary school Number of infant deaths, out of 1000 births, in first year of life Country’s polity score on the regime scale, −10 to 10 (Marshall and Jaggers data set) Dummy variable, the effect of the Cold War Affinity (the extent of borrowing countries’ support of these hegemons in voting at UN General Assembly)*** Interaction term: polity scores during and after the Cold War Interaction term: hegemonic influence during and after the Cold War (US, Japan for AsDB, etc.)

The data include “zero” values – indicating years when a given country received zero loan disbursements. * All independent variables are lagged one time period; units are all country-year (for loans, the aggregate of all loans made per country-year). ** It is logged because of its wide distribution and the expectation that very large loan values (outliers) have a smaller effect. *** Anton Strezhnev; Erik Voeten, 2012–08, “United Nations General Assembly Voting Data,” http://hdl.handle.net/1902.1/12379 UNF:5:fWzDBiI+iY41v5zJF3JnoQ== Erik Voeten [Distributor] V4 [Version] 1

Lastly, the indicators for which coefficients are significant are graphed with marginal effects of these independent variables to provide further context and evidence of their impact. A “margin” is a statistic computed from predictions from a model while manipulating the values of the covariates. There are different ways for controlling or manipulating the values of covariates, hence many ways of computing marginal effects. Older methods generally default to using the “means” for variables whose values have not been otherwise specified. That is, they estimate marginal effects at the means (MEMs). Since fixed values, such as the mean, are often not relevant for interpretation, especially for dummy variables (no year could be classified as “49% post–Cold War”),

5.2 Findings

157

I decided to compute and graph marginal effects at representative values (MERs). These are also a variant of “predictive margins,” that is, predictions where some of the covariates are not fixed, but rather set at several “representative” values. With MERs, we choose ranges of values for one or more independent variables and then see how the marginal effects or the partial effect of one covariate differs across that range. MERs can be intuitively meaningful because they show how the effects of a covariate vary by the effects of other variables. For instance Figure 5.6 shows that the effect of Japan’s affiliation on loan amounts is different during the Cold War than after the Cold War. Since all my models are OLS models, all marginal effects graphed in this chapter are linear predictions. They all show the effect on the conditional mean of y of a change in one or more of the repressors while keeping some other repressors at representative values. For a continuous covariate, the marginal effect is the first derivative of the response with respect to the covariate. For a discrete covariate, the marginal effect is the effect of a discrete change of the covariate. Marginal effects are also particularly relevant when interactions are present. In this case, I compute and graph the effect of each variable, taking into account both its main effect and its effect from the interaction term, thus offering one of the best ways to visualize significant effects in models with interactions between variables. The significance of the interaction effects is based on the joint significance of the interaction terms and the main effects. The individual coefficients are not reliable because interaction terms introduce collinearity into models, inflate standard errors, and reduce the significance of the individual coefficients.9

5.2 5.2.1

findings The IDB

In brief, the analysis of the IDB indicates that the bank does not provide more loans to less developed countries in its region and there is some evidence that it gives more loans to more developed countries. It further indicates that the IDB is influenced by political factors. The IDB, then, makes loans that are not entirely commensurate with a development agency. The results of this analysis are presented in Table 5.2: 9

See Brambor, Clark, and Golder, 2005 (especially pp. 11–15).

table 5.2 IDB loans per capita Economic Model 1 Lagged Dependent Variable GDP per capita (lag)

Trade (lag) FDI (lag) Primary Education (lag) Infant Mortality (lag) Polity (lag) Post–Cold War Polity*Cold War USA Affinity (lag)

0.27*** (0.04) 2.92e-04** (1.15e-04)

Economic Model 2 0.13** (0.05)

Social Model 0.04 (0.04)

Political Model 0.10* (0.04)

4.63e-03 (2.72e-03) 4.55e-05*** (9.62e-06)

Combined Model 1 0.05 (0.05)

Combined Model 2 0.08* (0.04)

2.95e-5*** (9.16e-06) −4.77e-03 (0.01) −0.02*** (3.06e-03)

−0.01*** (3.34e-03) −0.01 (0.01) 1.20*** (0.29) −3.43e-03 (0.03) −0.90*** (0.11)

0.96*** (0.19)

0.81*** (0.20)

−0.51*** (0.14)

−0.58*** (0.14)

USA Affinity*Cold War Constant Observations R-squared

−8.82*** (2.41) 783 0.26

−3.81*** (0.75) 614 0.20

−3.60*** (1.34) 580 0.26

1.67*** (0.35) −5.17*** (0.33) 754 0.37

*** p < 0.01, ** p < 0.05, * p < 0.1 Robust standard errors in parentheses. a FDI and GDP per capita have high VIF and thus are in separate models (economic models 1 and 2). b FDI and Infant Mortality have high VIF and thus are in separate Combined Models.

1.36*** (0.36) −5.41*** (0.40) 633 0.26

1.52*** (0.35) −4.80*** (0.33) 784 0.37

Banks or Development Agencies?

160

The specifications in Economic Models 1 and 2 generate clear conclusions about the IDB’s economic motivations (GDP per capita and FDI were not included in the same model due to high VIF). According to Economic Model 1, GDP per capita is significantly associated with the amount of loans IDB member states received. In other words, higher GDP per capita predicts more loan amounts (a $1000 rise in GDP per capita is associated with about $3 million in additional loan amounts). FDI is significant according to Economic Model 2 and this significance holds in Combined Model 1.10 Countries that receive higher amounts of FDI are more likely to receive loans. For example, a $1 million rise in FDI leads to approximately $45 million more in loans.11 To put this in the context of the bank – development agency spectrum, the regression results for the economic variables point to a preference for making less risky loans (financial rather than need-based considerations), to countries with higher GDP per capita and the ability to attract FDI (see Figure 5.1). The social specifications yield mixed results with regard to the IDB’s lending policies. According to the Social Model, primary school

–5

Linear prediction –4.5 –4

–3.5

Inter-American Development Bank Predictive Margins with 95% CIs

–171

4829

9829 14829 Foreign direct investment

19829

figure 5.1 The effect of FDI on loan amount (US $) for the IDB

10 11

I did not include GDP per capita in the combined models due to high VIF. See Cameron and Trivedi (specifically p. 333) for more on marginal effects.

24829

5.2 Findings

161

–6.5

–6

Linear prediction –5.5 –5

–4.5

–4

Inter-American Development Bank Predictive Margins with 95% CIs

9.1

29.1

49.1

69.1 89.1 Infant mortality rate

109.1

129.1

149.1

figure 5.2 The effect of infant mortality rate on loan amount (US $) for the IDB

enrollment is not a significant predictor of loans, but infant mortality rates are a significant predictor of loans. Results of this model demonstrate that higher rates of infant mortality are associated with receiving a lower amount of loans from the IDB: for each additional infant death (rates are per thousand births) a borrowing country receives about $3 million less in loan amounts. This negative and significant association also holds in Combined Model 2. These findings are in opposition to the stated goals of a development agency – of an institution that purports to shape its policy on the premise of the promotion of social and economic development for its borrowing members – with its purported focus on poverty alleviation (see Figure 5.2). In the context of its poverty alleviation mission, I expected some association between loans and education and health. Specifically, if low school enrollment and high infant mortality rates are predictors of loans, then it would indicate that the bank is targeting countries that are in greater need of assistance (less developed), with loans that may improve long-term health and education services – sectors that do not usually attract private investors or FDI (this is not the case for the IDB). At the same time, the economic variables point to lending that is driven by the financial standing of the recipient country. Aid to countries with lower GDP per capita and

162

Banks or Development Agencies?

less access to FDI would constitute a policy that is more likely driven by a development agency. FDI is a predictor for loans by the IDB, but not in a way that signals primacy of development concerns: more loans are likely where there is more FDI. At the same time, the bank also gives preference to borrowing countries with higher GDP per capita, targeting loans at middle – and upper – income countries in the region. Not only do the economic and social indicators point to policies that are not aligned with development agency priorities, the analysis also indicates that the IDB is influenced by political concerns in its lending practices. While the interaction effect between the Polity Index and the post–Cold War period is not significant according to the Political Model, the post–Cold War variable is significant in this model, as well as in the Combined Models. According to the Political Model, a country’s score on the Polity Index is not significantly related to the amount of loans that it receives. This finding also holds for the interaction effect of Polity and the Cold War (including the impact of the end of the Cold War on the relationship between loans and the Polity Index). In the Political Model, I include an indicator variable for the post–Cold War era, coded 1 for all years after 1989, and 0 otherwise. According to this model, democracies do not necessarily receive more loans from the IDB when the post–Cold War control is included. This finding challenges the notion that American political motivation to secure a stronghold in Latin America during the Cold War facilitated the founding of the IDB, but that once the Cold War was over American foreign policy interests became more heavily intertwined with regime-type, namely, democracy. In fact, no association between Polity and loans suggests that the Polity score does not impact loans post–Cold War. Thus, political in its overall goal of preventing the spread of Communism, the IDB’s lending during the Cold War is not correlated with levels of democracy, according to these findings.12 When US interest in the region changed after the Cold War, the IDB did not pursue aggressive preference for like-minded political ideologies (now that the sphere of influence was no longer a central policy goal). Loans from the IDB have not, in fact, been associated with the level of democracy.13 It could be, then, that with the end of the Cold War, the IDB focused on socioeconomic indicators of its borrowing members and the “bankability” of its loans (taking into account that the central political

12

13

It should be noted that “Cold War politics” arguments of US foreign policy often stress the preference for allies over regard for polity. This could also be explained by the reduction in variation of Polity scores after the end of the Cold War.

5.2 Findings

163

pariahs – Cuba and more recently Venezuela – are already precluded from borrowing). When translating these results onto the bank–development agency spectrum, we find that the Cold War made little difference in rewarding borrowing countries for political reform. However, there is some evidence that with the end of the Cold War IDB lending appears to be more bank-centered.14 To test US hegemonic influence, I include in the Political Model a measure of countries’ political affinity with the United States’s voting in the UNGA (UN General Assembly). In this model, being aligned with how the United States votes in the UNGA is significantly associated with the amount of loans that countries receive from the IDB. Furthermore, with the effect of the Cold War taken into account, a significant association is found. These results hold in both Combined Models, where I find joint significance of the Cold War indicator and affinity with the United States at above the 99 percentile.15 During the Cold War, affinity with the United States as represented by UNGA votes was a negative predictor for loans from the IDB. But after 1989, there is a positive association between Latin American countries aligned with the United States (as reflected in UNGA votes) and the amount of loans received.16 Therefore, during the Cold War, voting opposite the United States in the UNGA is associated with more loans – suggesting that credit is not a reward for political alliances as represented by the UNGA. And once the Cold War ended, voting alongside the United States in the UNGA is a predictor for loans (see Figure 5.3).17 To sum up, these findings demonstrate that the IDB favors financial considerations over a borrowing country’s need for aid based on its socioeconomic status. While these preferences are perfectly rational from a banking standpoint, they do not meet the expectations of a development agency, particularly when, according to the RDBs’ Articles of Agreement, 14

15

16

17

With recent political shifts in Latin America and the rise in anti-Americanism, future research should reexamine these indicators to see whether the IDB has started incorporating polity preferences similar to those during the Cold War – not identified by ideological kinship, but, rather, by strategic alliance (as measured by the Affinity indicator). The significance of the interaction effects is based on the joint significance of the interaction terms and the main effects. The coefficients of the individual coefficients are not reliable because interaction terms introduce collinearity into models, inflate standard errors, and reduce the significance of the individual coefficients. For more on comparison of aid during the Cold War and post–Cold War, see Dunning, 2004. While it is likely that the United States shifted preference after the Cold War and that IDB borrowers had to compete for financial support in a different global political environment, it is also plausible that by then many Latin American countries had undergone political reform that made their preferences more aligned with those of the United States.

Banks or Development Agencies?

164

–6

–5

Linear prediction –4 –3

–2

Inter-American Development Bank Predictive Margins with 95% CIs

–.86

–.66

.14 .34 .54 –.46 –.26 –.06 USA (affiliation measure based on UN voting) During Cold War

.74

.94

Post–Cold War

figure 5.3 The effect of US affiliation during and after the Cold War on loan amount (US $) for the IDB

they are committed to making decisions based on need. Politically, US influence is evident, and strongly associated with the Cold War. While higher Polity Index scores are not associated with IDB lending both during and after the Cold War, borrowers’ UN voting at the General Assembly is significantly associated with IDB lending: after the Cold War there is a shift to “reward” countries that have voted alongside the United States. These findings raise an interesting point (further explored in Chapter 6) about the changing nature of hegemony over time: the United States, as the central hegemon responsible for the creation and funding of the IDB, has an interest in ensuring alliances during the Cold War, without regard to political ideologies (neorealism). However, when the Cold War ends, a quasi-latent hegemon emerges as more actively concerned with rewarding preferred policies (neoliberalism), but at the same time oversees an institution that follows financially prudent policies.

5.2.2

The AsDB

The results of the statistical analysis for the AsDB (presented in Table 5.3) are somewhat different from those of the IDB. However, like the IDB,

5.2 Findings

165

table 5.3 AsDB loans per capita Economic Model (1) Lagged Dependent Variable GDP per capita (lag) Trade (lag)

0.54*** 0.47*** (0.05) (0.07) −4.33e-06 (1.(1.01e-04)

Combined Model

0.57*** (0.06)

0.47*** (0.07)

3.64e05** (1.38e05) −3.93e-04 (0.01) 1.80e-03 (4.25e-03)

Prime Education (lag) Infant Mortality (lag) Polity (lag) Post–Cold War Polity*Cold War

0.01* (0.01) −0.01 (0.01) −2.04*** (0.63) −0.04 (0.21) −0.27 (0.45)

US Affinity (lag) US Affinity* Cold War Japan Affinity (lag)

Observations R-squared

0.52*** (0.06)

Political Model

−4.80e-03 (3.00e-03) 3.47e05*** (1.37e05)

FDI (lag)

Japan Affinity* Cold War Constant

Economic Social Model (2) Model

−6.82*** (0.34) 542 0.48

2.85*** (0.91) −3.62*** −3.69*** −1.91*** (0.70) (1.16) (0.92) 492 459 462 0.47 0.43 0.49

−1.93*** (0.54)

−0.68* (0.40)

2.88*** (0.81) −2.68*** (0.81) 440 0.50

*** p < 0.01, ** p < 0.05, * p < 0.1 Robust standard errors in parentheses. a Interactions (Polity*Cold War) and (US Affinity*Cold War) were not significant and therefore not included in the models (to prevent unnecessary collinearity).

166

Banks or Development Agencies?

overall, the AsDB does not favor the countries that are statistically most in need. Further, the results for political factors highlight the unique governance structure of the AsDB: Japan is influential, while the United States, despite trying to maintain a strong political presence in the bank (at least equal to that of Japan’s impact), does not seem to have as much influence as it does at the IDB.18 Economic Models 1 and 2 demonstrate that AsDB lending is not associated with poorer economic measures. Thus, poorer Asian countries with lower GDP per capita are no more likely to receive loans from the AsDB than borrowing member countries with higher per capita GDP. The volume of trade is not associated with loans, but the ability to attract investment does predict the likelihood of loans from the AsDB: countries with higher FDI amounts are more likely to receive larger loan amounts (for every additional $1000 in FDI a borrowing country is expected to receive an increase of $3.5 million in loans).19 While these results are mixed with respect to market openness and its association with loans from the AsDB, the finding that lending is associated with higher FDI suggests a policy that is “safer” and points to a preference for banking concerns (see Figure 5.4). Certain characteristics of the AsDB and of development in Asia offer some explanations for these findings. First, while many Asian countries have grown rapidly over the past two decades, attracting larger amounts of foreign investment, some have been slow to open their borders to trade. The AsDB then may not put much weight on trade openness when considering loan requests. However, the findings for FDI suggest prioritizing financial considerations over development motives: the AsDB makes significantly more loans to Asian countries that are also the recipients of more FDI (and thus more likely to have the ability to secure private credit). These findings are in line with the AsDB’s policy as stated in its Articles of Agreement that does not require foreign economic policy to be a condition or even a central factor in considering loans. Further, Japan’s shared hegemonic status with the United States may drive the bank toward less controversial – financially prudent – policies that are naturally less need-driven; the dual hegemonic status plays a role in preventing either

18

19

Kilby (2006) also finds some evidence that the AsDB favors countries that vote alongside Japan at the UN. The significant results hold in the Combined Model. GDP per capita was included in a separate economic model because of multicollinearity (high VIF).

5.2 Findings

167

–6

–5.5

Linear prediction –5 –4.5

–4

–3.5

Asian Development Bank Predictive Margins with 95% CIs

–7

4993

9993

14993 19993 24993 Foreign direct investment

29993

34993

39993

figure 5.4 The effect of FDI on loan amount (US $) for the AsDB

hegemon from using the development bank to fulfill its particular political interests. The analysis of the social variables yields similar mixed results for the AsDB. The Social Model’s findings demonstrate that neither primary education nor infant mortality rates are associated with loans, while infant mortality rates are positively associated with loans in the Combined Model. Thus, while the Social Model finds that there is no association between loans and health or education according to these measures, the Combined Model’s result suggests that for each additional infant death, a $100 increase in loan amounts is expected. These findings point to a weak relationship between loans and the bank’s stated development goals – possibly targeting countries with greater healthcare challenges.20 Combined with the findings of the economic indicators, the AsDB’s lending

20

There could be numerous explanations for this finding. One possibility is that the countries that are targeted for financial reasons – those that are more likely to receive higher FDI volumes, are also countries with health concerns that are worthwhile for the Bank to support. Another, related, possibility is that Asian overall growth and rise in FDI has affected countries that while able to offer economic incentives for investors and loans, are still struggling with health issues. The AsDB in this case would have some incentive to support health-related initiatives if the economic indicators look promising.

Banks or Development Agencies?

168

–7

Linear prediction –6 –5

–4

Asian Development Bank Predictive Margins with 95% CIs

4

19

34

49

64 79 94 Infant mortality rate

109

124

139

154

figure 5.5 The effect of infant mortality rate on loan amount (US $) for the AsDB

yields mixed results when it comes to lending that is development-driven versus lending that is motivated by financial considerations. While it may not meet the development agency standards set by the RDBs themselves, the AsDB does appear to fare better than the IDB (see Figure 5.5). The AsDB, like the IDB, does not seem to base its lending policy on the borrowers’ level of democracy. AsDB member countries’ score on the Polity Index, according to the Political Model, is not significantly related to the amount of loans they receive. Likewise, when the effect of the Cold War is taken into account, the Polity Index, the post–Cold War period, and the interaction effect between the two are not jointly significant. Thus, both during and after the Cold War, lending by the AsDB is not associated with recipient countries’ level of democracy. The role of the hegemons in the AsDB is similar to that of the IDB, but with the AsDB dual-hegemonic structure, it is only Japan – not the United States – that is directly affecting borrowing members’ UNGA voting. Affinity with the bank’s hegemons in UN voting is not a predictor for loans according to the Political Model. However, both the Political Model and the Combined Model demonstrate that changes associated with the end of the Cold War do impact lending as it relates to Japan’s hegemonic

5.2 Findings

169

–8

Linear prediction –7 –6

–5

Asian Development Bank Predictive Margins with 95% CIs

.1

.2

.3

.4

.5

.6

.7

.8

.9

JAPAN (affiliation measure based on UN voting) During Cold War

Post–Cold War

figure 5.6 The effect of Japan affiliation during and after the Cold War on loan amount (US $) for the AsDB

role in the bank: the Japan Affinity measure, the post–Cold War, and the interaction effect between the two are jointly significant at better than the 99 percent confidence interval, rendering a change of policy with the end of the Cold War as shown in Figure 5.6. During the Cold War, according to the findings presented in Figure 5.6, countries that voted in opposition to Japan at the UNGA were likely to receive more loans. However, this changed when the Cold War was over, with AsDB borrowing members more likely to receive more loans if they voted alongside Japan in the UNGA. Japanese hegemonic influence may have shifted to rewarding supportive countries (represented by UNGA voting patterns) in the post–Cold War era once the Cold War need to secure spheres of influence (politically) was removed. Notably, although the United States and Japan both hold equal voting shares at the AsDB and share hegemonic influence, Japan seems to hold more political sway, according to these results.21 21

In Chapter 6, I discuss in greater detail the qualitative findings that point to a strong American interest in the bank. Although not manifested in its association with UNGA voting, it appears that US hegemonic preferences remain on the macro level – in ensuring that rival regimes do not receive loans – and that despite its reported “competition” with

170

Banks or Development Agencies? 5.2.3

The AfDB

The findings in the analysis of the AfDB paint a different picture than findings for the IDB and the AsDB. The economic factors are not predictors for loans by the AfDB, but the bank favors countries with better healthcare. On the political front, findings are mixed and point to the bank’s relative independence from influential hegemons. Like the IDB and the AsDB, the AfDB also appears to pursue policies that are not centrally driven by the development level of the borrower, although it seems less driven by economic indicators compared to the other RDBs.22 Further, the AfDB is less constrained by political considerations. The results for the analysis of the AfDB are presented in Table 5.4. According to Economic Model 1, GDP per capita is not a predictor for receiving loans. Further, Economic Model 2 reveals that FDI and trade are also not predictors for loans.23 These findings demonstrate that the overall measures of recipient countries’ economies, as represented by these variables, are not determinants of lending by the AfDB. A rise in GDP per capita or FDI for these countries does not lead to increased loans. Arguably, it is not surprising that FDI and Trade are not associated with loans. After all, Africa lags in attracting FDI and in trade openness compared to other developing regions (see Chapter 4 for more detail on FDI by region).24 Therefore, while this finding (that there is no association between loans and these indicators of borrowers’ economies) may be indicative of a riskier lending policy – where loans are not directed at countries whose economies fare better, it also may reflect the regions’ overall economic state.25

22

23 24

25

Japan over the governance of the bank, the United States does not influence lending decisions when they are within the parameters of the Articles of Agreement. It is important to note, however, that the post–Cold War Affinity finding is similar to that of the IDB (US). Thus, this result points to an active Japanese role in decision making – not surprising since Japan’s role in Asia is central to its regional policies throughout these years. While all African countries are in need of aid, this analysis treats countries on a comparative basis, showing how the institution distributes its limited resources. It therefore highlights the motivations, constraints, and results of the bank’s policy, demonstrating that even in a donor-controlled institution, a development bank struggles balancing banking and development needs. The economic variables were separated into two models because of multicollinearity. Compared to other developing regions, Africa still attracts only modest FDI and trade openness has been hampered by unstable regimes, conflict, and extreme poverty, especially in sub-Saharan Africa. Whether an increase in loans precedes higher FDI amounts or vice versa does not matter here: we know that there has been a rise in FDI (Chapter 4) and yet no association is found here.

table 5.4 AfDB loans per capita Economic Model 1 Lagged Dependent Variable GDP per capita (lag) Trade (lag) FDI (lag) Primary Education (lag) Infant Mortality (lag) Polity (lag) Post–Cold War Polity*Cold War USA Affinity (lag) USA Affinity*Cold War

0.54*** (0.04) 1.95e-04 (1.37e-04)

Economic Model 2 0.56*** (0.04)

Social Model 0.44*** (0.04) 3.51e-04* (1.88e-04)

Political Model

Combined Model

0.49*** (0.04) 4.05e-04** (1.84e-04)

0.45*** (0.04) 1.89e-04 (2.47e-04)

0.02 (0.02) −0.10 (0.17) −0.04* (0.02) −0.90*** (0.19)

−0.01*** (0.01) −0.01 (0.02) −0.24 (0.17) −0.04* (0.02) −0.61*** (0.18)

−1.20e-03 (4.21e-03) −1.17e-09 (9.78e-10) −0.01 (4.31e-03) −0.02*** (4.50e-03)

France Affinity (lag) France Affinity*Cold War Constant Observations R-squared

0.22 (.34) −2.48*** (0.73) 807 0.66

−1.64*** (0.36) 768 0.66

1.04*** (1.12) 764 0.66

−3.78*** (0.43) 807 0.66

−3.01*** (0.58) 745 0.66

*** p < 0.01, ** p < 0.05, * p < 0.1 Robust standard errors in parentheses. a Combined Model controls for GDP instead of FDI. b Interaction between (US Affinity*Cold War) and (France Affinity*Cold War) were not significant and therefore not included in the models.

5.2 Findings

173

–9

Linear prediction –8 –7 –6

–5

African Development Bank Predictive Margins with 95% CIs

13

28

43

58

73 88 103 118 Infant mortality rate

133

148

163

178

figure 5.7 The effect of infant mortality rate on loan amount (US $) for the AfDB

The Social Model reveals a different trend with regard to the social variables: while primary education is not associated with loans, infant mortality rates are negatively associated with loans, suggesting that more infant deaths predict less loans amounts. For example, each additional infant death (out of 1000), leads to approximately $20,000 less in loans. The negative association between infant mortality rates and loans does not point to development-oriented preferences. The level of education in borrowing countries, then, does not seem to be a factor in AfDB lending (based on primary enrollment). At the same time the bank does not make higher loan amounts available to the countries that score lower on health (represented by infant mortality rates). Although these findings do not decidedly fall on the “development agency” or “bank” ends of the spectrum, we can deduce that the bank’s social investment interest is not motivated by education and is risk-averse when it comes to health-care (see Figure 5.7). The analysis of the political variables for the AfDB yields different results than the IDB and the AsDB. The Polity Index is not significantly related to receiving loans – and therefore not a predictor for loans. However, the interaction effect between the Polity Index and the

Banks or Development Agencies? African Development Bank Predictive Margins of Cold War (During & After) at Different Levels of Polity with 95% Cls

–8

Linear prediction –7.5 –7

–6.5

174

–8

–7

–6

–5

–4

–3

–2 –1 0 Polity score

During Cold War

1

2

3

4

5

Post–Cold War

figure 5.8 The effect of Cold War on loan amount (US $) at different levels of polity for the AfDB

post–Cold War period is significant according to the Political Model (and remains significant in the Combined Model). The main effects for the Polity Index, the post–Cold War period, and the interaction effect between the two are jointly significant at better than the 99 percent confidence interval. During the Cold War, a higher Polity Index score predicts larger loan amounts, while after the Cold War borrowing countries with lower Polity Index scores are the recipients of more loans(see Figure 5.8). This finding points to development-agency preferences, where need drives decisions and is not influenced by the level of democracy of the borrowing state. When taking into account one of the central differences between the IDB and the AfDB – hegemonic involvement – this finding reinforces the conventional thinking: American interest in the IDB is commensurate with US foreign policy preferences. During the Cold War, alliances were valued over governance factors, and after the Cold War, the importance of democracy is elevated. Thus, the findings in the analysis of the IDB imply a certain degree of US influence. American role at the AfDB is smaller, with less influence and involvement. While a positive association between loans and Polity Index during the Cold War can be explained by US preferences, it is not clear that that is the case. During the Cold War,

5.2 Findings

175

–8.4

–8.2

Linear prediction –8 –7.8 –7.6

–7.4

African Development Bank Predictive Margins with 95% CIs

–.1

0

.1 .2 .3 .4 .5 USA (affinity measure based on UN voting)

.6

.7

figure 5.9 The effect of US affinity on loan amount (US $) for the AfDB

when many African countries were newly independent, lending to countries with higher Polity Index scores could point to a preference in lending to countries that provide a more stable political environment. But the end of the Cold War altered the predictability of democracy in attracting bank loans: democratization does not seem to be a factor for the AfDB’s decisionmaking when it comes to loans after the Cold War.26 This finding for US affinity in the Political Model reinforces this conclusion: voting with the United States at the UNGA (affinity measure) is a negative predictor for loans. Thus, countries that vote alongside the United States are likely to receive smaller amounts of AfDB loans than those that vote opposite the United States. The interaction effect between US (and France’s) Affinity and loans is not significant (with the Cold War interaction term). Thus, the end of the Cold War did not alter the direction of the association generated by affinity: the AfDB’s lending to countries that are not as closely allied with the United States remained consistent after the Cold War (see Figure 5.9). 26

An alternative explanation may be that the Polity Index for Africa is much different than that for Latin America – with fewer countries that are democracies or in the transition phase. However, if political reform were a factor for the AfDB then even small changes in Polity would have yielded some results.

176

Banks or Development Agencies?

Given the United States’s smaller role in the AfDB – both in material terms through its voting shares on the board, and more generally in terms of its political interests – the similarity between the IDB and AfDB with regard to US Affinity is unexpected. For both RDBs, Affinity with the United States is negatively associated with loans during the Cold War, suggesting that regardless of American involvement in the RDB, the United States did not use its influence to reward allies (based on UNGA voting) with loans. However, the banks differ in the Cold War’s aftermath, where IDB loans are positively associated with UNGA voting while the negative association remains constant at the AfDB. Particularly in the post–Cold War era when American political weight (and Washington Consensus) is considered more central to influence over policies of IFIs, these findings demonstrate the independence of the AfDB from US hegemony – at least in comparison to the IDB.27 5.2.4

The EBRD

The newest addition to this group of RDBs, the EBRD, represents a different breed of development bank. Specifically, post–Cold War politics are factored into its foundation as it began lending in 1991 (the Cold War interaction term is therefore irrelevant). In its Articles of Agreement, it commits to facilitate a transition to market economies among its borrowers rather than emphasize the alleviation of poverty. And while there is no Cold War and post–Cold War comparison for the EBRD to indicate policy changes, comparing the findings of the EBRD’s lending to those of the other RDBs, post–Cold War, provides insight into both the political and financial motivations for lending in a post–Cold War environment. The analysis (results in Table 5.5) reveals an institution that prefers to lend to countries that are more developed. Furthermore, in accordance with the bank’s mission, the findings demonstrate a strong political preference for countries that score higher on the Polity Index. The hegemonic influence, however, is mixed based on the results. Overall, the EBRD’s lending policy appears both financially and politically driven. The GDP per capita of the EBRD’s borrowing members is a solid predictor for receiving loans (Economic Model 1). And while trade does not predict the likelihood of receiving loans (Economic Model 2), this model further establishes that FDI is associated with loans – EBRD borrowing countries that attract more FDI are more likely to receive larger 27

Chapter 6 examines more closely the advantages and the price paid for this independence.

table 5.5 EBRD loans per capita Economic Model 1 Lagged Dependent Variable GDP per capita (lag) Trade (lag) FDI (lag) Primary Education (lag) Infant Mortality (lag) Polity (lag) USA Affinity (lag)

0.45*** (0.08) 3.51e-04*** (6.98e-05)

Economic Model 2 0.45*** (0.08)

Social Model 0.42*** (0.08)

Political Model 1 0.44** (0.09)

0.50*** (0.08)

3.13e-03 (3.01e-03) 4.50e-05*** (1.40e-05) −0.03** (0.01) −0.12*** (0.02) 0.05** (0.02) −1.20*** (0.30)

0.08*** (0.03)

Combined Model 1

0.14 (0.89)

Combined Model 2

0.32*** (0.08)

0.36*** (0.08)

1.92e-05** (9.37e-06) −0.03*** (0.01) −0.09*** (0.02) 0.05** (0.03) −0.74* (0.40)

1.64e-05* (9.35e-06) −0.02** (0.01) −0.09*** (0.02) 0.08*** (0.03)

−3.13*** (0.86)

France Affinity (lag) Germany Affinity (lag)

Political Model 2

−1.81* (0.95) 1.44* (0.87)

UK Affinity (lag) Constant Observations R-squared

−6.15*** (1.03) 308 0.49

−2.96*** (0.84) 294 0.47

*** p < 0.01, ** p < 0.05, * p < 0.1 Robust standard errors in parentheses. a Variables were separated in models due to high VIFs.

8.26*** (1.56) 268 0.61

0.45 (0.52) −4.39*** (0.99) 284 0.52

−1.86*** (0.57) 284 0.51

0.02 (0.53) 4.82*** (1.83) 254 0.60

6.66*** (1.63) 254 0.60

5.2 Findings

179

–5

–4.8

Linear prediction –4.6 –4.4

–4.2

European Bank for Reconstruction and Development Predictive Margins with 95% CIs

0

2000

4000

6000 8000 10000 12000 14000 16000 18000 Foreign direct investment

figure 5.10 The effect of FDI on loan amount (US $) for the EBRD

loan amounts: a $10 million increase in FDI leads to an increase of $4.5 million in loans (see Figure 5.10). According to the Social Model, primary school enrollment is a predictor for loans (see Figure 5.11), consistent with a developmentdriven policy – the higher the enrollment, the lower the loans (for every 1% rise in primary school enrollment, a borrowing member receives approximately $3 million less in loans). However, the results for infant mortality rates show contrasting effects with respect to developmentcentered policy: higher rates of infant mortality predict lower loan amounts (see Figure 5.12). Thus, a borrowing country that fares worse on health as represented by infant mortality is less likely to receive loans, with a decrease of about $12,000 in loans for every additional infant death out of one thousand births (all these findings are reinforced in both Combined Models). Overall these findings paint a mixed picture vis-à-vis the banking versus development-agency goals; with loans more likely to countries with lower primary school enrollment rates (less educated), but also more likely to countries with lower infant mortality rates (healthier). These conflating findings could be rooted in the particulars of Eastern and Central Europe. Infant mortality, overall, are lower for this

Banks or Development Agencies?

180

–6

–5.5

Linear prediction –5 –4.5

–4

European Bank for Reconstruction and Development Predictive Margins with 95% CIs

83

88

93 98 103 108 Percent enrolled in primary education

113

118

figure 5.11 The effect of primary education on loan amount (US $) for the EBRD

–15

Linear prediction –10 –5

0

European Bank for Reconstruction and Development Predictive Margins with 95% CIs

3

13

23

33 43 53 Infant mortality rate

63

73

83

figure 5.12 The effect of infant mortality rate on loan amount (US $) for the EBRD

5.2 Findings

181

–6.5

–6

Linear prediction –5.5 –5

–4.5

–4

European Bank for Reconstruction and Development Predictive Margins with 95% CIs

–9

–7

–5

–3

–1 1 Polity score

3

5

7

9

figure 5.13 The effect of polity score on loan amount (US $) for the EBRD

regions than other developing regions. An association between higher loan amounts and lower infant mortality rates, in this case, may reflect a lower priority compared with lending to education, and an overall trajectory that meets both development needs and the mission of the bank.28 Politically, in accordance with its deliberate agenda, the EBRD values democratic states: according to Political Models 1 and 2 (and the Combined Models) countries that score higher on the Polity Index (see Figure 5.13) are more likely to receive larger loan amounts (a one unit shift toward democracy on the Polity score leads to an increase of nearly $8 million in loans, according to Political Model 2). As for hegemonic influence at the EBRD, Political Models 1 and 2 and the Combined Models demonstrate that hegemons, while ensuring that policies are within the framework of its Articles of Agreement, delegate decision-making about loans to staff bankers, who follow financial considerations rather than political alliances. Of the four hegemons tested – the United States, UK, France, and Germany – only voting alongside Germany in the UN’s General Assembly predicts more loan amounts 28

See Section 5.3 and Figure 5.15 for a more in-depth examination.

182

Banks or Development Agencies?

(Combined Model 1). However, voting with the United States and France at the UNGA is a negative predictor for loan amounts, while affinity with the UK is not significant. Germany’s central role in European finance and banking (home to the ECB) is a plausible explanation for borrowing countries’ alignment with Germany at the UNGA (see Figure 5.14).29 The EBRD’s London headquarters, or the central role played by France and the United States in its foundation, do not seem to translate into borrowing members’ political allegiance with these hegemons, according to the Affinity Measure. Overall, these findings potentially reinforce the delegation hypotheses, whereby central shareholders distance themselves from decision-making as long as policies remain within the agreed-upon framework. And, by keeping the institution at arm’s length, hegemons do not “reward” or “punish” borrowing members based on their political preferences at the UN (and borrowing members do not expect retaliation or support when casting these UN votes).

5.3

comparison summary

Although their purpose is similar, the results reveal some differences in lending policy between the development banks. The examination of their unique governance structure and regional circumstances inform the analysis of the RDBs’ divergent policy outputs on the one hand. And on the other hand, it helps explain why, under certain circumstances, the development banks are institutionally constrained to meet their objectives in ways that are similar. In Table 5.6, I compare the results of the analysis for each variable across the RDBs. Figure 5.15 shows the effect of infant mortality rates on loan amounts at high (90 percentile) and low (10 percentile) levels of FDI, for all RDBs. These graphs combine social and economic variables to paint a nuanced picture of loan destination. The differentiation between high and low FDI percentiles emphasizes the “types” of countries on the infant mortality continuum. Countries that receive higher FDI amounts are considered attractive destinations by private investors and we can assume that they are able to attract credit more 29

The negative association of loans and voting alongside the United States may be a testament to the EBRD’s European nature – where borrowing members are not concerned about their political allegiance with the United States. Overall, these mixed findings with regard to all four hegemons may indicate that politics does not play a role in decision-making and that the EBRD follows its banking guidelines with hegemons remaining hands-off.

Linear prediction

.25

.35

.45

.55

.65

.75

.85

–6 –5.5 –5 –4.5 –4 –3.5

–4.5 –5 –5.5

Linear prediction

–4

European Bank for Reconstruction and Development Predictive Margins with 95% CIs

–.9 –.8 –.7 –.6 –.5 –.4 –.3 –.2 –.1 0 .1 .2 .3 .4 .5 .6

.95

UK (affiliation measure based on UN voting)

–4 –4.5 –5

Linear prediction

–5.5

–5 –5.5

Linear prediction

–4.5

–3.5

USA (affiliation measure based on UN voting)

.7 .72 .74 .76 .78 .8 .82 .84 .86 .88 .9 .92 .94 .96 .98 1 Germany (affiliation measure based on UN voting)

.5

.55

.6

.65

.7

.75

.8

.85

.9

.95

France (affiliation measure based on UN voting)

figure 5.14 The effect of hegemon affiliations on predicted loan amount (US $) for the EBRD

1

Banks or Development Agencies?

184

table 5.6 Results (summary) IDB

AsDB

AfDB

EBRD

GDP FDI Trade Primary Education Infant Mortality Polity Index Polity x postCW

Positive (B) Positive (B) None None

None Positive (B) None None

None None None None

Positive (B) Positive (B) None Negative (DA)

Negative (B)

None

Negative (B)

Negative (B)

None None

None None

Hegemony (Affinity)

–-

–-

Hegemony x post-CW

Negative (US) during CW Positive (US) after the CW

Negative (Japan) during CW; Positive (Japan) after the CW; US – none

None Positive Positive during N/A CW, negative after US negative; US, France France none negative; Germany positive; UK none None N/A

B = Bank preferences; DA = Development Agency preferences

easily; while countries on the lower FDI scale are more likely to struggle getting credit for development projects. When this dichotomy is juxtaposed with the health measure (represented by infant mortality rates) we can assess the relationship between loans and health based on a country’s ability to attract FDI on its own regional scale (at the 90th percentile and at the 10th percentile). Higher rates of Infant Mortality are associated with lower loan amounts and higher FDI amounts are associated with increased loans at the IDB. Thus, countries with the most FDI and lowest infant mortality rates receive the highest loan amounts. This finding demonstrates a deviation from the purported development commitment of the bank. This finding for the AsDB is somewhat more consistent with a development mission, where countries with higher rates of infant

Predictive Margins with 95% Cls Asian Development Bank

Linear prediction –6 –5

–4

–7 –6 –5 –4 –3 –2

–3

Inter-American Development Bank

9.1

29.1

49.1

69.1

89.1

109.1 129.1 149.1

4

19

34

49

Infant mortality rate

64

79

94 109 124 139 154

Infant mortality rate European Bank for Reconstruction and Development

13 28 43 58 73 88 103 118 133 148 163 178 Infant mortality rate low FDI

high FDI

–10

–8

–6

–10–9 –8 –7 –6 –5

–4

African Development Bank

20

30

40

50

60

Infant mortality rate FDI = Foreign Direct Investment:

low FDI = 10th Percentile

high FDI = 90th Percentile

figure 5.15 The effect of infant mortality rate on loan amount (US $) at high (90th percentile) and low (10th percentile) levels of foreign direct investment for the IDB, AsDB, AfDB, and EBRD

186

Banks or Development Agencies?

mortality receive higher loan amounts. But AsDB borrowers that receive more FDI are also the recipients of higher loan amounts. This seems puzzling: when controlling for infant mortality rates, development appears to be a motivation, yet more loans are also being disbursed to countries that attract more private investment. A possible explanation for this finding is the overall higher FDI amounts to Asia.30 For example, Asia’s global FDI shares more than doubled from 1970 to 1990 (from 29 percent to 63 percent); at the same time, Latin America’s share of FDI went from 55 percent in 1970 to 28 percent in 1990. Given its rapidly growing shares of FDI, it could be concluded that Asian countries faring poorly in health are still an attractive FDI destination.31 If this is the case, then loan association with countries that have higher infant mortality rates puts the AsDB slightly closer to the development end of the continuum. The analysis of the AfDB in the previous section demonstrated that none of the economic indicators are predictors for loans. Since the significance of FDI in Africa may be weak due to the substantially lower FDI amounts to the region, FDI may not be a factor in lending as of yet. Infant mortality, on the other hand, is a predictor for loans: countries with higher infant mortality rates are likely to receive less loan amounts, a finding that questions the AfDB’s commitment to development-driven policy. Interpreting findings for the EBRD, yields conclusions similar to those reached for the IDB: higher Infant Mortality rates are associated with lower loan amounts and higher FDI amounts are a predictor for more loans.32 Like the IDB, the EBRD’s performance in this assessment is banklike, favoring loans to countries with higher private investment levels, while making fewer loans to countries with more health challenges. Examining the EBRD’s social policy more closely – where both FDI and primary education as they related to loans were significant variables in the regression analysis – Figure 5.16 delves into the contradiction described in the previous section:33

30 31

32

33

See Chapter 4 and supplemental website (www.ruth-ben-artzi.com). This is also explained by Asia’s impressive growth during that period, including the prominent rise of the “Asian Tigers.” Overall, private investors appear to have determined that Asia is a worthy investment, despite lingering development (health) problems. The difference between countries that receive higher and lower amounts of FDI is smaller than for the IDB. This may be because there is less variation in FDI for EBRD borrowers. The contradiction between the two social variables – infant mortality rates producing development-agency leaning results, while primary education producing bank-like results.

5.3 Comparison Summary

–5.5

–5

–4.5

–4

–3.5

European Bank for Reconstruction and Development Predictive Margins with 90% Cls

–6

Linear prediction

187

83

88

93 98 103 108 Percent enrolled in primary education Low FDI High FDI

113

118

Median FDI

FDI = Foreign Direct Investment Low = 10th Percentile High = 90th Percentile

figure 5.16 The effect of primary education rate on predicted loan amount (US $) at high (90th percentile), median and low (10th percentile) levels of foreign direct investment for the EBRD

Figure 5.16 examines the effects of education at different levels of FDI at the EBRD (at the 10th percentile, 50th percentile, and 90th percentile). The results demonstrate that although loans decrease for countries with higher percentages of primary school enrollment for all levels of FDI, the countries that receive the highest FDI amounts are still likely to receive substantially greater loan amounts than those at the bottom end. The effect of primary education rates on predicted loan amounts at different FDI percentiles paints a picture of some preference for financial considerations over those of development. Loan amounts are higher at the high FDI percentile (90th) than at the lower FDI percentile (10th). Therefore, closer examination of the association between loans and primary education at varying FDI levels reveals a preference for higher FDI: countries that rank lower on primary school enrollment AND receive higher FDI amounts are more likely to receive more loans (than EBRD borrowers that receive less FDI). This finding substantiates the EBRD’s concern for financial solvency and “bank” driven policies in its loan decisions.

188

Banks or Development Agencies?

For the other RDBs, the fact that loans are not associated with education, although not a definitive finding, casts doubt on their commitment to development-driven policies since one would expect that countries with lower education scores would receive more aid. Both education and health are generally more risky (and long-term) investments, and less likely to be supported by private financial resources. Adding to this the statistical results for health (infant mortality rates), the IDB, AfDB and the EBRD generate findings that lean toward banking preferences: while at the EBRD primary school enrollment predicts loans in a way that is consistent with development concerns (less education, more loans), infant mortality rates predict loans in a reverse direction (less loans to countries with higher rates of infant mortality). The AsDB may be a bit closer to the “development” end of the spectrum in this regard (compared to the other RDBs), although results are inconclusive in the absence of association between the social indicators and loans for the AsDB. Proceeding to an examination of political variables, the Figure 5.17 demonstrates the effect of US Affinity on loan amounts at different levels of infant mortality rates at the IDB (high – 90th percentile; and low – 10th percentile). While loans are negatively associated with US affinity during the Cold War, and positively associated with it in the post–Cold War era (see Figure 5.18), inserting infant mortality rates at different levels also points to banking-motivated preferences: both during and after the Cold War, IDB borrowing members with higher infant mortality rates (at the 90th percentile) are likely to receive less loan amounts than countries at the lower end of the infant mortality spectrum. This finding is consistent with the overall statistical findings that point to the IDB’s preference for financial solvency in its borrowers. Along with the results generated by the economic indicators, it appears that overall the RDBs prefer lending to countries with stronger socioeconomic foundations (with the AsDB slightly more “development” oriented, and the IDB slightly more “bank” oriented). For commercial banks this strategy is undeniably reasonable, but for development banks that promise not to compete with the private sector while alleviating poverty and contributing to development (except for the EBRD), these findings question the degree to which the RDBs fulfill their fundamental objectives. Next, a discussion is warranted on the comparison of the effects of hegemon affiliation on loan amounts during and after the Cold War for the IDB, AfDB, and AsDB. The political indicators – polity and affinity – are central to the analysis because I test whether the level of democracy

5.3 Comparison Summary

189

–7

–6

Linear prediction –5 –4 –3

–2

Inter-American Development Bank Predictive Margins with 95% Cls

–86

–66

–46 –26 –.06 .14 .34 .54 USA (affinity measure based on UN voting) Low IMRT, during CW High IMRT, during CW

.74

.94

High IMRT, post CW Low IMRT, post CW

figure 5.17 The effect of US affinity on loan amount (US $) at high (90th percentile) and low (10th percentile) of infant mortality rate and during and after the cold war for the IDB.

and the involvement of hegemons are associated with the likelihood to receive loans. Moreover, examining these variables during the Cold War and after it ended furthers the study of the extent to which the RDBs’ lending policy is subject to political pressures – those that correspond to ideological preferences (Polity Index) and those that correspond to powerful donor-member interests (UNGA affinity). During the Cold War, African countries with higher Polity scores receive more loans, while after the Cold War ended, higher Polity scores predict less loan amounts for AfDB borrowing members (see Figure 5.8). The level of democracy of the IDB and the AsDB’s appear not to factor into lending decisions, even when controlling for the Cold War. At the same time, Eastern and Central European countries with higher polity scores receive larger loan amounts from the EBRD, consistent with the bank’s mission. It is surprising that of the three older RDBs that were to a large extent a product of the Cold War politics, democracy scores matter in conjunction with the Cold War only at the AfDB – the RDB with the least Western hegemonic official involvement. These varied results for all four banks seem to suggest that hegemonic or political considerations weigh in on polity preferences mostly

Predictive Margins with 95% Confidence Intervals Asian Development Bank

–7

–5

–8

–6

Linear prediction

–4

–6

–3

–5

–2

Inter-American Development Bank

–86 –66 –46 –26 –0.6 .14

.34

.54

.74

.94

USA (affinity measure on UN voting)

.1

.2

.3

.4

.5

.6

.7

.8

.9

Japan (affinity measure on UN voting)

–9–8.5–8–7.5–7–6.5

African Development Bank

–1

0

.1

.2

.3

.4

.5

.6

.7

Post CW = 0

Post CW = 1

USA (affinity measure on UN voting)

figure 5.18 The effect of hegemon affiliations on loan amount (US $) during and after the Cold War for the IDB, AsDB, and AfDB

5.3 Comparison Summary

191

in a global way – as part of the Articles of Agreement for the EBRD, as a policy of exclusion of states for the IDB (e.g., Cuba), and with no particular overarching policy preference for the AsDB. In fact, the Affinity results demonstrate that after the Cold War, the United States and Japan prefer that the IDB and AsDB lend to countries that support their policies (in the UNGA) – countries with similar ideological leanings (polity), while during the Cold War loans are higher for countries that do not support the United States (IDB) or Japan (AsDB), possibly in an effort to sway them. The findings for these two banks are similar. At the IDB during the Cold War, voting alongside the United States is negatively associated with loans; once the Cold War ends, the association becomes positive. Thus, UNGA votes, even if opposed to those of the United States, did not increase the likelihood of loans – they decreased it. One possible explanation for this could be an American concerted effort during the Cold War not to reward its closest allies that would remain loyal, but to offer incentives for countries that could potentially pose a problem to US foreign interests.34 These results are similar for Japan and the AsDB, suggesting that its central role in the bank was similar to that of the United States at the IDB.35 Affinity UNGA voting is not associated with loans for the AfDB both during and after the Cold War, highlighting the bank’s relative independence from hegemons and nonregional donors.36 At the same time, for the EBRD, which has multiple hegemons, voting with the United States in the UNGA is also a negative predictor for loans. Either due to the diffusion of power in a multihegemonic system, or because European hegemons with vested interests in the EBRD are simply more dominant than the nonregional US, the EBRD does not seem to follow policies dictated by the United States. However, since political goals are already embedded into the EBRD’s Articles of Agreement, it may not be necessary for hegemons to closely monitor or use the bank as a political tool, especially given that it already makes more loans to countries with higher Polity scores.

34

35

36

For more on the issue of swing/core voters see Cox and McCubbins, 1986; Dixit and Londregan, 1996. American influence over Japan during the Cold War (the decades following WWII) can explain this, perhaps even in an effort not to exert overt influence, but allow Japan to play a more public role in the region (while perception of the United States in Asia was often negative). This also explains in part US reluctance to support the AfDB and its significantly smaller endowment and more erratic credit rating.

192

Banks or Development Agencies?

Interestingly, comparing the three European hegemons of the EBRD, we find that voting alongside the UK at the UNGA is not a predictor for loans; voting with France is a negative predictor for loans; and voting together with Germany is a positive predictor for loans (see Figure 5.14). Perhaps this highlights the UK’s relative distance from the Eurozone and Germany’s central influence with its strong economy and its host status to the ECB. This finding certainly reinforces recent European experience of Germany’s influential economic role (in the aftermath of the 2008 recession).

5.4

conclusion

Are the RDBs development agencies or banks? Which factors determine a borrowing member country’s likelihood to receive loans? This chapter has revealed that despite institutional variances based on their origin, the RDBs face similar constraints in their lending policies that bear on the “types” of countries they lend more to. Further, the different compositions of hegemonic roles on the RDBs’ boards – a central feature of their decision-making policy – do not seem to have a major impact on the RDBs’ place on the bank/development agency spectrum: none of these RDBs demonstrates a clear and coherent “development agency” agenda. Evidently, their “bank” feature – with the exception of the Africandominated AfDB – is substantial and poses constraints on the institution as a whole, leading to policies that favor less risky financing, as well as to a higher likelihood of competing with private institutions in the process.37 Specifically, the paradox inherent in the banking features and development goals of the RDBs is supported by the mixed results of the analysis of their loans.38 This quantitative assessment highlights the actual activities of the banks that are the reason for their existence and the source of controversy 37

38

Although the AfDB’s lending is not associated with economic factors, it is still risk averse when it comes to the social factors examined in this study: it fares better than the other RDBs on the development agency/bank continuum, but still seems to fall short of the ideal. Some factors cannot be quantified and measured and thus were not included in this part of the analysis. Furthermore, other social, economic, and political variables that could possibly contribute to the analysis are not included here due to technical difficulties: either data was sparse, incomplete, or nonreliable for the year-span and countries in this analysis, or the process of collecting the data was complicated. As data collection techniques become more sophisticated with more data on developing countries widely available (see Tierney et al., 2011), future research could add to the scope of this study.

5.4 Conclusion

193

surrounding them. The analysis in Chapter 6 that incorporates information gathered through interviews, archival research and secondary sources, supplements these statistical findings and produces a more detailed account of the RDBs’ working agendas – what guides the RDBs’ policies and influences their programs – to produce the output analyzed in this chapter. In sum, the findings generated in this chapter establish that, despite differences in the extent to which political factors are associated with loans, the resulting lending policy in the context of poverty levels and development needs of borrowing states is not substantially different. On one end, the oldest of these development banks, the IDB, with dominant American involvement but with considerable regional participation, avoids risky lending and can be placed closer to the “bank” end of the spectrum. The AsDB, with US and Japanese dual-hegemonic institutional structure does not lag much behind the IDB, demonstrating an agenda that is closer to “development agency” in the social realm, but overall – when economic variables are included – seems to prefer “bankable” lending. And on the other end, the AfDB, which resolutely rejected donor domination in favor of regional control, also falls into the trappings of its banking concerns: it inhibits an agenda that is truly guided by the objectives of poverty alleviation and development, though it is less driven by economic concerns than all of its counterparts. And finally, the newest member to the family, the EBRD, opting for a different institutional structure that embraces political goals, also falls short on preference for policies driven by development concerns.

6 Political and Economic Constraints, Principals and Agents, and Prospects for Development

Normally speaking, it may be said that forces of a capitalist society, if left unchecked, tend to make the rich richer and the poor poorer and thus, increase the gap between them. (Jawaharlal Nehru, “Basic Approach,” reprinted in Vincent Shean Nehru, The Years of Power, 1960, p. 295) If a free society cannot help the many who are poor, it cannot save the few who are rich. (John F. Kennedy, Inaugural Address, January 20, 1961) For the first time in our history, it is possible to conquer poverty. (Lyndon Baines Johnson, Speech to Congress, March 16, 1964)

How do the RDBs make policy decisions? With hegemon(s), states, boards, managers, and staff all involved in the governance and activities of the RDBs, how is their influence integrated into the policy output of each bank, and does the balance of power in the governance structure change over time and/or under diverse circumstances? This chapter enriches the findings in the Chapters 4 and 5 by adding a central dimension to the results of the quantitative analysis: it describes the context in which the policy outputs are undertaken by unpacking the governance structure of the RDBs and taking a closer look at the decision making processes that influences outcomes. By examining the loans RDBs disburse in the context of borrowing countries’ economic, social, and political indicators, the quantitative analysis in the last two chapters demonstrated the differences (and similarities) between the development banks. In addition to “big picture” 194

Political and Economic Constraints

195

assessments – which “types” of countries get more loans from the RDBs, whether the Cold War marked a turning point in lending policy, and the extent to which hegemons reward allies – Chapter 4 examined a sample of borrowing countries to provide further tangible evidence and context for Chapter 5’s statistical analysis. These findings, however, do not complete the picture of the decisionmaking processes and political strategies utilized by the RDBs and their member countries. Through an analysis of the typology of RDBs’ governance structures, this chapter complements the quantitative analysis through a more nuanced consideration of the way RDBs develop lending policies. An examination of the culture and institutional mechanisms of the development banks reveals their unique attributes as well as their similar features. As we will see, the varying levels of RDB shareholders’ involvement has implications not only for how the RDBs are perceived, but also for how these institutions pursue their objectives. This chapter argues that the governance of RDBs plays a crucial role in the RDBs’ ability to fulfill their mission of development. The major findings for each bank are: (1) The IDB has the potential to be development-driven. With a single most powerful hegemon, the bank’s majority shareholders (borrowing countries) and its mostly Latin American executives, have substantial control over policy as long as the hegemon’s central preferences are satisfied. (2) The AsDB is the most bank-driven, perhaps because the political motives of the two major hegemons offset each other: lending policy within the parameters of sound banking policies are most likely to meet the approval of both hegemons. (3) The AfDB is potentially closest to a development institution, but as a result it lacks funding and necessary political support of donors. (4) The EBRD represents a new model of RDBs. It is donor-controlled and explicitly driven by new political and economic goals beyond just development. First, I examine the varying hegemonic configurations of the RDBs and how each particular institutional design affects both donor and borrowing members. Next, I consider shareholder motivations to join the RDBs; this includes the implication of voting percentages or shares, the benefits associated with shareholder status, and the scope of shareholder involvement in setting lending policies. Then, I delve into the role – and power – of RDBs’ borrowing members. Finally, this chapter concludes by mapping

Political and Economic Constraints

196

the RDBs’ place in the overall geography of development banking, including RDB prospects to reduce poverty.

6.1

political interests and rdb s

An organization in which states are members is inevitably subject to political motivations. Such is the case with the RDBs, where states are not simply shareholders with voting power, but also hold noneconomic stakes in the outcomes of loans. Whereas in other international organizations, such as the UN, the political nature and interests of member states are evident in day-to-day public activities, RDBs project images of professional organizations that function independently of members’ political agendas. Shareholders are presented as holding a supervisory role that is ostensibly limited to financial contributions by way of replenishments, where individual states’ policy preferences purportedly have no place. As hard as IFIs try to mute political influences, it is a common belief among critics that multilateral financial institutions are, in fact, political.1 The World Bank and the IMF have both been heavily attacked as instruments of American foreign policy, part of the “Washington Consensus.”2 Based on their Articles of Agreement, it was not the founders’ overt intention to create political institutions. For example, the following two sections from the Articles of Agreement of the WB illustrate the desire of the drafters to avoid influencing or being influenced by the political interests of members: The Bank shall make arrangements to ensure that the proceeds of any loans are used for the purposes for which the loan was granted with due attention to considerations of economy and efficiency and without regard to political or other non-economic influences or considerations. [Art. III, sec. 5b] The Bank and its officers shall not interfere in the political affairs of any member; nor shall they be influenced in their decisions by the political character of the member or members concerned. Only economic considerations shall be relevant to their decisions, and these considerations shall be weighed impartially in order to achieve the purposes stated in Article I. [Art. IV, sec. 10]

The RDBs, with structures resembling the WB, adopted this philosophy as well. I use the term structure in same way it was used by Krasner (1981): derived from Waltz (1979) and based on the “structural model” proposed 1

2

See Stiglitz, 2002; Thacker, 1999; Stone, 2002, 2011; Weaver, 2008; Babb, 2009; Vreeland, 2003, 2007; and Gould, 2006. See Stiglitz, 2002; and Gilpin, 1987.

6.1 Political Interests and RDBs

197

by Keohane and Nye (1977). In this context, the RDBs’ Articles of Agreement and (commercial) bank-like characteristics help them espouse a politics-free image,3 which is further bolstered by the nonpublic nature of their deliberations about policy. While the RDBs eschew overt political identity, similar to the political susceptibility of national central banks,4 RDBs are in fact subject to influences of powerful stakeholders that can manipulate policy to achieve political interests. Whereas the member states delegate the daily running of the institutions to the professional staff, the influence of powerful shareholders is evident, particularly in extreme political situations. The hegemon(s) also exert their power on issues such as which countries can become members, which countries are eligible for loans, and whether certain countries should be suspended from borrowing. For example, the fact that Cuba is not a member of the IDB is directly connected to American hegemonic influence.5 Though other members – both borrowing and donor – have often discussed the possibility of Cuba’s membership at the IDB and though high-ranking managers express their exasperation that Cuba is not a member, this issue was resolved according to American preferences.6 Other examples of the politics of membership include North Korea’s noninclusion in the AsDB and Afghanistan’s suspension from that bank in 1979. Major donors are also active in making sure the financial statuses of the institutions are not compromised. They are concerned that the development banks keep a balanced budget, so that donor committed callable capital is not actually called upon. For example, in 1989, during rocky negotiations for the seventh replenishment of the IDB, the United States threatened to withdraw its subscription account from the bank when the US Treasury could not reconcile its differences with the bank’s management and with its borrowing members. The serious disagreements were primarily a result of the growing debt crisis in Latin America, and the debate about whether financial assistance should be offered only in combination with the adoption of policy changes designed to trigger fiscal improvements. 3

4 5

6

For more on the structure, see analysis in Chapter 2. Note that the EBRD differs from the other RDBs in that it does purport, in its Articles of Agreement, to consider domestic political affairs of countries it lends to. See, for example, Cukierman and Webb, 1995; Cukierman, 1994; Maxfield, 1997. IDB officials interviewed for this book emphasized that many Latin American countries would support Cuban membership. They stressed that Cuba has an interest in being part of an institution that services the entire region, possibly benefitting from access to capital. Based on interviews conducted with IDB staff between 1999 and 2002.

198

Political and Economic Constraints

Instances of hegemonic exercise of authority are the most observable but not the exclusive examples of hegemonic power; the influence of hegemon(s) percolates under the surface.7 These informal hegemonic interactions defy quantification. They are manifested in closed-door discussions, behind-the-scenes lobbying, and even “water-cooler” hallway negotiations. Hegemonic influence is also expressed in the staff that occupies top positions in the RDBs: whether administrators are veterans of investment banks or experienced political activists impacts the way they interact with board members, as well as the extent of navigation in the “political space” they make available to shareholder countries. For example, IDB administrators are attuned to the political agendas that motivate shareholder states. They use this information to navigate between political pressures and the policies they wish to advance. Ultimately, they are successful in incrementally making changes in the bank’s strategy.8 Conversely, the EBRD’s staff of bankers seems to follow a rigid agenda that is based on banking preferences. Although the overall purpose of the EBRD is more political than that of the other RDBs, it conducts itself very much like a commercial bank in its daily operations. In Chapter 5, hegemonic influence was analyzed quantitatively using the affinity measure. This analysis examined whether borrowing member countries that vote along with the hegemon(s) in the UNGA are likely to receive more loans. The findings for the IDB, for example, pointed to a change in policy when the Cold War ended: during the Cold War, voting opposite the United States predicted more loans; post–Cold War, borrowing members who voted along with the United States at the UNGA were more likely to receive higher loan amounts. Findings for the AsDB suggest a similar trend with respect to Japan’s hegemonic role (voting opposite Japan predicts more loans during the Cold War, while voting alongside Japan in the UNGA predict more loan amounts after the Cold War). These findings, however, are not sufficient to establish the place – and power – of the hegemon in the institution. For a more accurate portrayal of the role of the hegemons we need to assess the special provisions awarded to hegemons in the Articles of Agreement, the hegemons’ ability 7

8

Based on interviews with Barbara Upton (2000), IDB staff (2000, 2001, 2007), and OECD Development Center researchers who closely work with RDB managers (2001–2002). For example, a number of IDB staff at its Washington, DC, headquarters stipulate that they succeeded in navigating the shareholders and administrators of the development bank to approve the creation of a “Civil Society” department. As one staff put it “we started with one person and a desk, moved to a corner in an office, then to an office, and now we have a department” (interview: IDB, Washington, DC, April 2000).

6.1 Political Interests and RDBs

199

table 6.1 Hegemony at the four RDBs

Bank IDB AsDB AfDB EBRD

Hegemony

AAA Rating

Overt Political Agenda

Collaboration with Private Sector

Yes Yes No Yes

No No No Yes

Some Some None Yes

Single Bipolar None Multiple

to influence policy through an existing (or evolving) institutional culture and past practices, and the impact of shares (voting power) on the hegemon(s) ability to wield power. The results for the relationship between the affinity measure and loans examine the correlation between political support for the hegemon(s) at the UNGA and the likelihood of being rewarded by the hegemon(s) via loans made by the RDBs (the RDBs being a policy conduit); it does not assess the degree to which hegemons influence the operations of the RDBs more generally – in affecting their overall preferences, their institutional structure, and their position along the bank/development agency continuum. Hegemon(s) ability to influence policy, then, constitutes potential or latent power: while it does not clearly translate into overt rewarding of allies with loans, it can impact policies of the development banks. Whether the borrowing country is ideologically aligned with the United States seems to matter less than the overall impact of the development bank’s presence on the borrowing region. Table 6.1 summarizes the hegemonic structures of the four RDBs in the context of credit rating and relation to the private sector, as well as the existence of a deliberate political agenda. The next four sections examine the relations of each RDB with its hegemon(s).

6.1.1

The IDB

The United States is the sole hegemon at the IDB. The IDB’s Articles of Agreement (created in 1961) set a limit of about 30 percent of shares for the United States (while the limit for regional members’ aggregate votes is always over 50 percent to assure borrowing members hold majority shares).

200

Political and Economic Constraints

Although not a majority shareholder, American voting shares far exceed those of any other member country.9 American hegemony, especially during the IDB’s formative years, was not only manifest via its voting shares set in the Articles of Agreement, but also through US financial contributions. In the original capital subscription to the IDB’s ordinary capital resources, the United States provided $150 million paid-in capital and $200 million in callable capital, out of a total subscription of $800 million. Since many of the other members’ capital contributions were in gold or convertible capital, the United States essentially provided 74 percent of the IDB’s usable capital.10 The American financial share only grew after the first capital increase in 1964 and “by the end of the 1960s the United States provided 92 percent of the IDB’s usable ordinary resources” (Krasner, 1981, p. 308). Setting this standard early on, the United States continued its central financial role in the IDB’s Ordinary Capital Resources (OCR) and in its concessional lending window, the FSO. Since the United States also provided the lion’s share of callable capital that accorded the IDB the ability to borrow on the market, it was America’s central role in the development bank that supported its high credit rating. For their part, credit agencies assign ratings based on the assessment of the countries committing funds to the institutions, prioritizing financial solvency in the form of funds that are readily available. Until the late 1990s this model worked well for the IDB, enhancing its profile and allowing it to borrow cheaply on the market, thereby increasing its net flows to its borrowing member countries. However, as we have seen in the Chapter 5, in more recent years – with the sharp increase in financial resources available to Latin American countries – the role of the IDB as a public institution that is uniquely positioned to offer development lending has been subject to competition from private financial flows. It is precisely at this juncture that the institutional structure of a single hegemon became more crucial. With US support, the IDB has been able to make some changes in the way it engages with its borrowers, including shifting priorities and creating additional funds that collaborate with the private sector.11 For the United States, this is a worthwhile strategy as its 9 10 11

See Table 3.1 and Appendix 2. Krasner, 1981, p. 308. In recent years, a central role of the European Office of the IDB has been to mobilize resources into investment funds. This included substantial efforts to reserve funds for “strategic partnership with the private sector and civil society” (Office in Europe Activity Report, 2011, p. 5).

6.1 Political Interests and RDBs

201

primary political interest vis-à-vis the IDB is that the development bank remains relevant in an increasingly competitive international financial environment. And in 2008, when the global financial crisis hit Latin America, the IDB was able to step up its lending program and provide a safety net for some of the poorest borrowing member countries.12 IDB officials cite the advantages of headquarters located close to the hegemonic power structures. Coordination and discussions with US officials are facilitated by the proximity of the bank’s headquarters to the decision-making institutions of its most powerful shareholder. Furthermore, this geographic proximity facilitates the formation of informal networks between IDB staff and US decision makers, providing additional problem-solving channels. At the Fund for Special Operations (FSO), the concessional lending arm of the IDB, the United States maintains preferred status – veto power – granted in the Articles of Agreement (Article IV section 9[b]): the stipulations that all decisions concerning the operations of the Fund be approved by three-quarters of the votes, translates into American veto (holding 30 percent of the votes). The United States has refrained, for the most part, from using its power to veto decisions about loans, but having the option to do so impacts the decision-making process of the management and the board. By avoiding using its influence overtly via voting, the United States preempts the possible negative perception that would be generated by such action. Shareholders’ preference to resolve potential veto situations before a vote occurs demonstrates the IDB’s delicate legitimacy with its constituencies in Latin America. Moreover, the mere existence of US veto power at the FSO encourages staff and management to account for this power by avoiding proposals that the United States would veto. Special US status is manifested, according to IDB officials, in negotiations before voting, including at the IDB’s nonconcessional lending window; it is rarely the case that a matter disputed by the United States, or of which the United States is not supportive, will actually make it to a vote. Requests for loans that might seem controversial (whether in terms of the country of destination, nature of project, or amount) are discussed first off the record until an understanding is reached. Thus, although the United States does not have official veto power at the IDB, its status at the FSO, its large percentage of shares, and

12

See Financial Times report on the Future of the MDBs, Sept. 24, 2012; and based on interview with Carlos Jarque, Director of IDB Europe, April 2012.

202

Political and Economic Constraints

its vice-presidential position, are some of the factors that make American interests influential at the IDB.13 An example of US influence over the IDB’s OCR occurred in 1987–1989 during replenishment talks. The Reagan administration, represented by Treasury Secretary Baker, insisted that the IDB implement the “Baker Plan” as a condition for financial replenishment. In this plan, the United States predicated debt relief on liberalizing policy reforms. This was the first time in the history of the IDB that such demands – tying financing with policy – were made in public. And, ultimately, IDB management and borrowing members acquiesced to these conditions.14 At a crucial juncture – replenishment negotiations – the United States was able to exercise its hegemonic power and force the IDB to conform to American parameters for overall lending policy. The negotiations lasted four years and were a fight for the soul of the IDB. The finance ministers of key regional members (Mexico, Brazil, and Argentina) walked out on James Baker (US Secretary of Treasury) during these grueling meetings. They demanded a change in US approach and a replacement of the American vice president of the bank. After an eighteen-month stalemate, George H. Bush was elected president and an agreement was reached following a marathon all-night negotiation session in Amsterdam. It was the IDB’s president, Enrique Iglesias, who managed to find a way around the impasse – keeping the United States at bay while implementing the necessary reforms for the bank to operate more effectively with continued US support. Most significantly, this standoff with the United States reiterates major donors’ role in the governance structure of the development bank: the United States weighed in heavily when its interest in the general policy of the IDB was at stake. When Brazil, Mexico, and Argentina took a stand that was at odds with that of the United States, they failed. Iglesias was instrumental not only in his ability to deal with the Americans, but even more so in the charisma he carried in the eyes of

13

14

Obtaining data on loans that were requested and denied is impossible for all RDBs. Thus, official loan request data is skewed in that it already takes into account the likelihood of approval. However, this shouldn’t be surprising and is, generally, the practice in every financial institution (including private ones). Because RDBs are supposed to be lenders of last resort, the assumption of this study is that borrowing countries that have a need for credit and did not get any loans on a given year would have applied for loans, even absent a record of such requests. See Babb, 2009, especially chapter 5; House Appropriation Subcommittee, Appropriations for 1989.

6.1 Political Interests and RDBs

203

his borrowing constituents, who backed the agreement he reached with the United States. This example also demonstrates that despite the assured majority of the borrowing members and the limit on US shares (set to 31.11 percent), American demands were met because sustained support of the bank by the United States is crucial. In practice, American de facto influence is greater that its 30 percent of the shares. This framework is unique to the IDB when compared to the other RDBs, and it sets the tone for the development bank’s operations, and how member countries, staff, and management interact with the hegemon.15 First, this overarching power structure within the IDB encourages internal negotiations and agreements before decisions reach the voting stage in board meetings.16 American power at the development bank is manifested in potential use – on a daily basis, decision-making is left to the discretion of the administrative staff. The parameters for the IDB’s activities are set by the hegemon, but the United States does not operate as an intrusive hegemon looking to reward with loans only those countries that support it unconditionally. Interviews with IDB officials reinforce this description of the decision-making process at the bank.17 Within the working framework provided by the United States, the IDB’s staff retains significant control over the accounting and finance departments. They provide the reports on the overviews of loan proposals and the assessment of loan feasibility. The IDB has made recent efforts to further decentralize, delegating more responsibilities to its regional offices.18 In its current reform agenda, the IDB stresses the central role of the country office in identifying and supervising projects for financing. This is reflected in increased resources for field offices so that staff there can assume greater responsibilities.

15

16 17 18

It should be noted that Canada, although a member of the IDB, does not undertake a hegemonic role (it is the one other regional donor country on the IDB’s board). American hegemonic status is rooted in Cold War policies and its global superpower status. Similar to other regional regimes, Canada does not pursue a power position (e.g., NAFTA). Canada’s stature as a regional donor is different than Japan’s. While Canada does not appear to compete with US hegemonic influence, Japan’s role in the AsDB is much more active since it considers itself a regional hegemon. See Babb, 2009; Krasner, 1981. February 2000–June 2002. “Over the past several years, there have been a number of reports on Bank operation which have stressed the importance of the Country Offices as an asset of the institution, and which have recommended greater efforts to decentralize resources and authority to the country offices” (www.iadb.org/policies, 2011).

204

Political and Economic Constraints

table 6.2 US abstentions and negative notes on loans approved by the IDB 1972 1973 1974 1975 1976 1977 1978 1979 1 Number of US Opposition Votes or Abstentions Total Number of 48 Projects Approved by IDB

1

0

9

10

14

9

8

50

51

62

66

74

63

49

Source: Krasner (1981, p. 314).

As a result, there is minimal opportunity for disputes when the board meets: assessment reports by the regional offices are heavily weighted and regarded as expert recommendations. Overall then, delegation and decentralization at the IDB occurs within the framework of American hegemony, since US support was necessary for these reforms to occur. Within the parameters of the Articles of Agreement and broad US interests, borrowing members have found ways to advance their preferences and make the development bank “their own.” The IDB’s singlehegemon structure, in this case, provides the framework for handling policy disputes. With only one hegemonic donor, there is only one policy vector – between the hegemon and donors – whereas an institutional structure that includes more than one hegemon adds more dimensions to the donor-borrower relationship. Borrowing members of the IDB need not worry about competing hegemonic interests. That, coupled with the dominant work culture that emphasizes informal and open discussions and interpersonal relations (mostly Latin American nationals manage the bank), contributes to a decision-making process where confrontation is generally an exception.19 Even in the IDB’s early years, from 1972 to 1979, as demonstrated by Table 6.2, the United States was opposed to or abstained from votes on the board in only 52 project cases (or about 11 percent), out of a total of 463 projects approved. Over the years, as institutional learning developed, the norm has been consensus with the hegemon, with differences mostly settled prior to voting. On occasion, when there are legal and legislative impediments to supporting a loan, the United States abstains. This allows loans to get 19

Based on interviews with seven IDB officials 1999–2002; see also Copelovitch, 2010.

6.1 Political Interests and RDBs

205

approved without jeopardizing American domestic interests (the abstention for the purpose of the IDB is largely symbolic; opposing a vote on a loan is rare and reserved only for extraordinary situations). For example, the controversial 2003 loan ($75 million) to complete the construction of a natural gas pipeline from the Camisea gas fields in the Amazon, over the Andes mountains, to a liquefaction plant on the coast of Lima, Peru was a case where US interagency disagreement – over environmental consequences – evoked heightened scrutiny. In this instance, the State Department asked for the extreme measure of US opposition to the loan. This request was a product of severe pressure from environmental NGOs as well as opposition to the project by the environmental staff at USAID. Ultimately, a decision was made that the National Security Council (NSC) would form a working group. The group included representatives from all interested US agencies (State Department, USAID, the Council for Environmental Quality, NSC, Treasury Department, and a few others). Finding ways to include loan provisions for transparency, participation by nonfinancial parties, and international environmental organizations, the working group instructed the US representative to abstain in the loan vote, allowing it to pass.20 This example further illustrates the nature of American hegemonic involvement with the IDB: when there is intra-agency conflict over the desired US position on a policy, the resulting American vote of abstention is symbolic because it was evident that there would be a majority in favor of the loan. America’s use of its hegemonic status at the IDB is, therefore, influenced by competing domestic interests in the United States as well as concern for how the use of power would be perceived by the IDB’s borrowing members. The United States concedes to the institution and its shareholders with the exception of extreme situations (e.g., the 1989 crisis). This demonstrates the limits on how the United States, in this instance, exercises its hegemonic power, and its strategic use of political capital. In this context, IDB officials stress the importance of the president of the bank in solidifying consensus. For example, some claim that Enrique Iglesias, the bank’s president from 1987 to 2005, who was well liked and respected in Washington circles, allowed for a smooth operation of the institution since he diplomatically settled potential conflicts with the hegemon before they reached an explosive stage. Iglesias’ leadership abilities and reputation also provided the IDB with a level of trust on the 20

Sanford, 2005, pp. 10–11.

206

Political and Economic Constraints

part of the US Treasury. Thus, in a system where formal votes on controversial proposals are rarely taken, the importance of the IDB President’s control over agenda-setting – what is discussed and put for a vote – is instrumental in ensuring that any differences between the institution and its most powerful shareholder are addressed informally. Finally, the IDB’s own self-studies that raised concerns about the degree to which lending since the 1990s effectively contributed to development would suggest that the reforms undertaken to satisfy the United States in the late 1980s were not necessarily conducive to the mission of the IDB as seen by its managers and staff.21 In this case, US exercise of power was an instance of “negative power,” where pressure was applied because the IDB was not acting in US interests. According to Wade’s assessment of American hegemonic influence at the World Bank (2002), “. . . to remain within the zone of effectiveness – the zone where the Bank is reliably responsive to US foreign policy aims and US domestic constituency groups but is still seen as having enough independence to induce weaker states to listen seriously to it as an independent development advisor – the US often limits its direct intervention to negative power, ensuring that the Bank does not do or say things contrary to US objectives, as distinct from instructing it more proactively.”22 Thus, hegemonic influence is manifest in preventing policies that are politically or financially unacceptable to the hegemon, not in managing ongoing activities.

6.1.2

The AsDB

The United States and Japan share a hegemonic status as the two largest (and equal) shareholders of the AsDB. The interaction between the two hegemons, and between them and the other member states creates a power dynamic on the board of the AsDB that is substantially different from that of the IDB. In the AsDB, the United States and Japan often compete to exert their position of leadership. The “behind the scenes” negotiations that take place are, therefore, more complex than at the IDB because both hegemons need to concur before proposals proceed to a vote.23

21

22 23

As Barbara Upton argues (interview August 23, 2000, Washington, DC), the IDB’s success in the early 1990s was at least partly due to the economic and political developments experienced by Latin American countries in the preceding twenty-five years and unrelated to the reforms undertaken by the bank in the late 1980s. Wade, 2002, p. 204. Based on interviews with former General Counsel to the AsDB, 2000.

6.1 Political Interests and RDBs

207

Exacerbating the bipolar nature of the hegemonic leadership of the AsDB, the two leading shareholders (13.20 percent of shares/votes for each) follow divergent leadership styles and cultures. In contrast to the IDB, the AsDB guarantees the hegemons’ involvement not only on the board, but also with the top positions in the administration.24 Where IDB members elect a president who is always from a borrowing country in the region, the AsDB’s Articles of Agreement provide that the president is a national of a regional country, but it does not stipulate that it should also be a borrowing member (Japan is a regional member and not a borrower). In practice, these top administrative posts have always been held by Japan and the United States, leaving little room for borrowers’ input.25 This institutional design results in an RDB that is clearly much more “donor centered.”26 The Japanese president and American vice president, make the administrative position of the two hegemons secure. Because donor rather than borrowing countries hold the top administrative positions, disagreements on policy are more commonly matters within the bank’s administration and are not limited to interaction among board members. Despite their post–WWII alliance, Japan and the United States compete for regional influence in Asia. Via the AsDB, they use their leverage to affect the bank’s policies as well as its institutional culture – the way it is run and how it operates. The friction between the two hegemons is a result of their respective geopolitical interests in Asia. For the United States, a stronghold in Asia was a central interest during the Cold War and remained so in its aftermath, spurred by Asia’s growing economic success. Japan, after recovering from the devastation of WWII, was interested in reinstating its central role in Asia. The IDB’s and AsDB’s distinct origins lead to major differences between their respective policy-making apparatus. As discussed in Chapter 2, the AsDB was created in 1967 not out of developing countries’ persistent lobbying, but as a result of donors’ interest. The continued importance of the AsDB to major donors has produced an institution that is well funded and supported, receives high ratings from credit

24

25

26

Although the IDB reserves the position of vice president for an American representative, the presidential position is always held by a professional from a borrowing, nonhegemonic country. This assures at the very least a perception of an institution that belongs to the borrowers as much as to the donors. Specifically, the AsDB has always had a Japanese president and an American vice president. “The formal organizational structure of the ADB reflects the major role played by its donor members” (Krasner, 1981, p. 317).

208

Political and Economic Constraints

agencies, and runs with relative efficiency and effectiveness.27 However, developing (borrowing) countries are less integrated into the AsDB’s decision-making apparatus than are the borrowing Latin American countries in the case of the IDB.28 Further, competition between the two hegemons of the AsDB and Japan’s deep interest in reinforcing its position in Asia assure centralization and hegemonic supervision. Although the AsDB’s headquarters are located far from its two influential hegemons, the administrative involvement of the hegemons – including holding top positions at the bank – substantially impacts the development bank’s conduct and culture. The divide between regional and nonregional members of the AsDB also differs from that of the IDB (and the AfDB). While “regional” members of the RDBs are also its borrowers, the AsDB counts nonborrowing members among its regional members: with Japan, Australia, and New Zealand categorized as regional members although they do not borrow from the bank, developing countries that borrow from the AsDB hold only a minority of the shares (45.819 percent).29 Compounding this competition, the institutional culture of the AsDB is quite different than that of the IDB. By all accounts, “professionalism” in the form of strict adherence to bureaucratic procedures is dominant in guiding the practices of staff and managers alike.30 The AsDB’s “topdown” formation (rather than a “bottom-up” as in the IDB) has had an impact on its institutional culture and the way in which donor and borrowing members interact.31 The AsDB also lacks the kind of active involvement by borrowing countries that is part of an organizational cultural developed over the years at the IDB and the AfDB. The top–bottom establishment set the tone for a noncohesive borrowers’ 27

28 29

30

31

This is with reference to the operation of the AsDB and not necessarily the results of the projects it funds. Interviews with AsDB and IDB staff and former officials, 1999, 2002, 2012. This is in sharp contrast to the IDB’s division of power, where developing countries have a block of majority votes. This seems to be the influence of a Japanese work and institutional ethic, laced with secrecy. A personal example of this difference was manifested in the way my requests for data on loans was processed by each of these institutions: while the IDB immediately responded to my requests and gave me access to its intranet, where I had permission to sift through all documentation of loans, the AsDB was more reserved, eventually sending me a hard-copy of all individual loans by year and country, rather than giving me access to its databases. Krasner (1981) underscores the stark difference between the AsDB and the IDB noting that least developed countries (LDCs) that are members of the AsDB have not had a blocking vote on the development banks’ operations (p. 317).

6.1 Political Interests and RDBs

209

block. Never forming a coalition to lobby for a development bank in the initial phase, the AsDB’s borrowing members’ have not formed voting or lobbying blocks around common policy preferences. Thus, although neither the United States nor Japan hold a veto power (or an official stature similar to that of the United States at the IDB), the combination of the voting power allocation and motivation for creating of the development bank solidified an institutional structure in which developing countries have had little tangible influence over the institution’s policy.32 It is not surprising then that debates over policy, according to accounts from numerous current and former officials, are between the competing hegemons rather than between the donors and borrowers or between borrowers. While Japan and the United States have enjoyed a political alliance since the end of WWII, they have been economic competitors in the last few decades. As Japan’s economy grew and stabilized, it sought and gained regional influence. For Japan, the creation of the AsDB in the 1960s was an opportunity to solidify its economic hegemony in Asia, while for the United States the development bank promised political opportunities during the Cold War era. The two hegemons, then, have at times collided over policy preferences and goals. A former AsDB General Counsel, Barbara Upton, and others33 surmise that disputes have largely been behind-the-scenes because, as Babb (2009) explains, there is “congruence of US and Japanese interests” (p. 29). Further, Kappagoda (1995) argues that Japan has always made an effort to accommodate the United States. This institutional structure often leads managers of the AsDB to rely on financially astute operating procedures, which can be seen as a balancing mechanism, to offset political intervention.34 This institutional characteristic 32

33

34

Furthermore, with Japan and the US holding the top two administrative positions at the bank, borrowing members do not have representative leadership in the top administrative ranks of the AsDB in the same way that borrowing member countries are represented at the IDB and the AfDB (where presidents are always nationals of a borrowing member country). Interviews with RDB and OECD officials suggest that borrowing members of the AsDB, in contrast to borrowing members of the IDB and AfDB, have reduced sense of ownership of the institution. Interviews conducted with former AsDB General Counsel and Barbara Upton (1999–2000), and with OECD staff who work closely with the RDBs 2001–2002. Exclusion of political considerations are set in the AsDB’s Articles of Agreement, Article 36(2): “The Bank, its President, Vice-President(s), officers and staff shall not interfere in the political affairs of any member, nor shall they be influenced in their decisions by the political character of the member concerned. Only economic considerations shall be relevant to their decisions. Such considerations shall be weighed impartially in order to achieve and carry out the purpose and functions of the Bank.”

210

Political and Economic Constraints

is consistent with the quantitative finding in Chapter 5, indicating that during the Cold War, voting alongside Japan in the UN General Assembly is associated with less loans for borrowing members (while after the Cold War it is associated with more loans). Essentially, the division in hegemonic influence at the AsDB, especially during the Cold War, encourages a business-like environment supported by both hegemons, where policy is perceived as financially sound. Political motivation, until 1991, is limited to Cold War alliances, but within this general parameter, financial considerations dominate.35 Post–Cold War, Japan appears to have become more dominant and its own political concerns set the overall agenda. But like the preceding decades, power sharing with the US makes financially sound decisions (within these parameters) more palatable (less “political”) to both hegemons. The AsDB’s lending, then, appears to be motivated by banking concerns more than by needbased considerations.36 Finally, Japanese support for like-minded countries after the Cold War ended is a recent strategic choice given China’s rise. In fact, for both hegemons, China’s quest for regional (and global) influence poses a challenge. China still borrows from the AsDB, however, in recent years, China has been actively changing its role in IFIs. Whether China will seek to augment its involvement with the AsDB remains unclear at this point. Given the AsDB’s structure, it may not be in China’s interest to seek more power within the institution, but rather to influence it from the outside. Overall, China seems to have taken a measured approach with regard to its involvement in the global financial and political architecture. While it substantially increased its bilateral role in developing countries, China appears to prefer a policy model that is driven by calculated interests and avoids interfering in other countries’ domestic preferences. For example, since 2010 the Development Assistance Committee of the OECD has made substantial efforts to engage China, including allowing its participation in DAC aid review proceedings. China’s stance thus far has been cautious, refusing to make any formal commitments.37

35

36

37

The RDBs are concerned with how they are perceived by member countries, potential investors (private sector), rating agencies, and the public(s) that are affected by the loans, including interest groups and civil society in borrowing members. Overall, this perception contributes to the assessment of the rating of the RDBs as well as their contribution as development agencies. Interviews in 2001–2003 confirm this, and see Krasner (1981, p. 319) for similar findings for the AsDB’s early years. Interviews with OECD and DAC officials, July 2010 and April 2012.

6.1 Political Interests and RDBs 6.1.3

211

The AfDB

While the AfDB is the most independent RDB (from a hegemonic control perspective), it is also the smallest, in terms of financial resources.38 Like the AsDB, the AfDB was created in 1964 on the basis of the WB’s institutional model (see Chapter 2). However, as discussed in previous chapters, creating the AfDB was purely an African initiative. With a predominantly African administration (including the highest-ranking positions such as president and vice president), the AfDB has resisted the involvement of donors or hegemons. The statistical findings in Chapter 5 confirmed that affinity with the United States is, in fact, a negative predictor for loans (i.e. voting along with the United States at the UNGA predicts smaller loan amounts).39 Its independent stance has cost the AfDB on a number of fronts. First, donor countries, having fewer political stakes in the institution, have not been generous with their checkbooks or with their diplomacy. As a result, the AfDB has a substantially smaller budget than the other RDBs (see Table 3.1). With less hegemonic involvement, the AfDB has been poorly endowed, compounding the challenge of attracting private sources of financing.40 The importance of credit ratings and the central role of credit rating agencies contributed to this difficulty: Both Standard & Poor’s and Moody’s have periodically lowered the AfDB’s credit rating (to AA+ or AA). Rating agencies’ most important consideration in evaluating MDBs is the creditworthiness of their members, making the involvement of nonregional shareholders necessary for a high credit score (the credit rating of the regional African shareholders is lower than that of nonregional donors, leading to a lower credit score for the institution). In lowering the credit rating for the AfDB, rating agencies indicate that major donors do not fully back the institution. Chapter 5 demonstrated that for the AfDB, FDI is not associated with loans.41 Although there is no guarantee that exposure to private

38 39

40

41

Krasner, 1981, p. 322. The reasons for examining the United States and France as central hegemons are France’s colonial heritage and continued influence in the African continent, and the United States’s prominent role in creating the institution: The United States is the largest nonregional shareholder at the AfDB while France exerts substantial political and cultural influence in large parts of Africa and sees itself as a central player on the continent. The positive financial outlook of the IDB and AsDB has helped them recruit, support, and maintain the institutional integrity of the development banks, and encouraged publicprivate partnerships. There is a positive association with FDI for the IDB, AsDB, and EBRD.

212

Political and Economic Constraints

companies via procurement contracts would lead to further investment, the AfDB’s low budget and limited membership over an extended period, put the development bank at a disadvantage.42 Finally, with its administration comprised of local representatives from borrowing members, governing with little external oversight, it has been targeted with charges of corruption and inefficiency. Whether warranted or not, such critiques have had an impact on the AfDB’s reputation in the eyes of donors and investors alike. Thus, although the AfDB is the property of its owners and borrowers – African countries – it suffers from a negative image and shortfall of resources. If nonborrowing donors had higher stakes and held larger shares in the institution, the AfDB’s budget would likely be larger and nonregional donors’ hegemonic role would have had a positive impact on its overall perception by rating-agencies, shareholders, and investors.43 More than just a stumbling block for raising finances, the absence of hierarchy at the AfDB leaves member countries vying for influence without setting a clear agenda. This leads to inefficiency and, as some critics argue, the cumbersome nature of the AfDB, including corruption. Many of these problems stem from the power structure of the development bank, where the largest shareholders are also borrowing members, and where vote shares are spread thinly across many borrowers. For example, Nigeria, which holds 9.7 percent of the votes of the AfDB (in 2012) – more than any other single country – is also a borrowing member. It is followed by the United States (6.9 percent), Japan (5.7 percent), Egypt (5.6 percent), and South Africa (5.1 percent). In May 2012, borrowing members of the AfDB held 58.65 percent of the votes (shares) of the development bank – the highest percentage of borrowing country control, in terms of voting shares, of any of the RDBs. Since there is no clear leadership position, the large distribution of shares across member countries can be an obstacle to governance. As it was first 42

43

As previously stated, a positive relationship between FDI and Loans, point to policy preferences based on financial considerations. While this puts an RDB closer to the “bank” end of the spectrum, it is also indicative of the existence – and changing nature – of FDI. Africa receives a substantially smaller amount of FDI, which could explain the lack of association between loans and FDI. We could, however, also observe that AfDB loans are not guided by the existence or rise of FDI – a positive aspect given development concerns, but a potentially negative one from a financial perspective. Overall, the dichotomy of “Bank” and “Development Agency” is embodied in the association between FDI and loans. Based on interviews with OECD officials and researchers who have worked closely with AfDB managers and staff and studied the institution extensively.

6.1 Political Interests and RDBs

213

conceived, the founding African nations were steadfast in their preference to exclude non-African countries from membership. Moreover, it was relatively equitable in its distribution of votes among African countries, with a maximum ratio of 5:1 in votes (while the maximum subscription ratio was 30:1).44 Thus, through the 1980s, AfDB decisions were made through a simple majority vote, as power distribution was relatively equal (in 1982, when nonregional members were allowed to join the bank due to a shortage in financing, this balance shifted). There is a direct connection between voting shares, overall shareholder composition, and the development bank’s ability to raise capital. Capital commitments in the AfDB’s initial years present an opportunity to measure the comparative degree to which securing total capital amounts and giving nonregional donors more control are supportive or detrimental to a development bank’s operations and identity. The AfDB embodies the struggle developing countries face in seeking development aid: the reluctance of donors to transfer substantial funds to a development institution that limits donor power.45 First, with overall small capital volume ($300 million in the initial capitalization), fifty percent was paid-in while the other fifty-percent designated as callable capital.46 In addition to its limited access to liquid amounts compared to other RDBs, the small sum of the callable portion of the capital commitment prevented the AfDB from generating additional funds through international financial markets (since the committed funds were insufficient as collateral for bond issues).47 Financial challenges were exacerbated when, member countries, many of which were struggling financially, failed to meet even the paid-in portion of their capital subscriptions, leaving the AfDB with only sixty percent of its paid-in portion in the late 1960s. 44

45

46

47

In comparison, in the 1970s the ratio of largest to smallest number of votes at the IDB was 588:1 and at the AsDB it was 33:1 (Krasner, 1981, p. 322). Ibid.; “The African Development Bank has achieved more autonomy than either of its counterparts. But its ability to recruit resources has been limited . . . Absent any effective control over daily operations of the Fund, developing countries were reluctant to make any serious commitments. In the late 1970s this led African states to admit industrialized countries to the Bank itself” (Krasner, 1981, p. 326). This is a high percent of paid-in capital, while the total capital amount is small (compared to other MDBs where callable capital is typically between 80 and 90 percent of total capital). As previously explained, the RDBs’ financial resources are generated, in addition to members’ contributions and commitments, via investments and financial markets, and specifically by issuing bonds. This is one of the reasons the RDBs are concerned with their credit ratings – these ratings have a high impact on the RDBs ability to generate income from bonds.

214

Political and Economic Constraints

The impact of voting distribution at the AfDB has therefore been twofold: (1) there are fewer funds and they are less reliable, and (2) lower callable capital amounts hinder financial growth through investment. In the 1980s, the AfDB began opening its membership to nonregional members in order to increase its funds, ensure financial stability, and secure a high credit rating (although all of these have been volatile over the years). As demonstrated in Chapter 5, whether or not there is a dominant hegemon, development banks are subject to similar financial pressures when they disburse loans, making their overall policies relatively similar. For the AfDB this finding leads to the conclusion that the high cost associated with preserving the ownership of the development bank in the hands of its borrowing members does not necessarily yield developmentcentered dividends: if the policy outputs are similar, then the participation of a hegemon(s) that could increase access to critical financing is desirable. Nevertheless, there are nontangible benefits to borrowing countries’ ownership of the development bank, and under different circumstances (for example, if there were better governance, more funding, or institutional reconfiguration), the AfDB could represent the most effectual form of IFI governance. 6.1.4

The EBRD

The most recent addition to this group of RDBs, the EBRD, has multiple hegemonic patrons. Both the Western European countries that set out to create it, and the United States, whose absence from the board would have meant its political marginalization (given American superpower status at the end of the Cold War), have high stakes in the operations of the EBRD.48 To that extent, the major European players – France, the United Kingdom, and Germany – alongside the United States – exhibit dominance and control not only in the vision and direction of the institution, but also in its daily activities.49 These four countries are the major 48

49

Just as the Cold War ended and the United States was emerging as a single dominant global superpower, the European architects of the EBRD were mindful of the necessity for the United States to be represented on the board of the bank. From the point of view of the United States, given the timing of the creation of the bank and the political changes taking place in Russia, it would be inconceivable not to hold membership in a financial institution with substantial potential to influence Russian political and economic reforms. “The EBRD was originally conceived by the French as a Europe-only body that would counterbalance the Americans and increase Europe’s influence in Eastern Europe. The Germans agreed to the establishment of the Bank only if the United States could join. After the Soviet Union said that it also wanted to join the Bank, it became clear to the

6.1 Political Interests and RDBs

215

contributors and conduits of policy-making. As EBRD officials note in multiple conversations, the overtly political, Western-based agenda of the bank was a deliberate decision of the original founders, who aimed at finding a novel, and perhaps improved, recipe for a public development institution.50 It became possible to explore new formulas for the creation of such an institution in the aftermath of the controversy surrounding the World Bank’s structural adjustment programs in the late 1980s.51 Additionally, the end of the Cold War marked a freedom to set the threshold for development aid at a higher (or different) level – both economically and politically – since the fundamental need to maintain a sphere of influence was no longer paramount. For donor countries, these two factors were restrictive and liberating at the same time: restrictive, because the operations of IFIs would now be the focus of NGOs and advocacy groups who seek public awareness, placing them under heightened scrutiny; and liberating, because a relief from Cold War pressures allowed developed countries to concentrate their attention on attaining favorable political and economic benefits from their involvement with these institutions, with developing countries having no other direct and accessible sources of patronage.52 The influence of the founding donors is evident throughout the EBRD’s headquarters and in its activities. For starters, the mission to help Eastern European countries in their transition to market economies demonstrates the donors’ political agenda. The goal of promoting market economies is a Western-based idea and a post–Cold War preference. Additionally, donors demonstrate understandable self-interest, both political (fostering democratization) and financial (expanding markets and decentralizing). Second, the bank’s top administrators, including its president, come from Western

50 51

52

Europeans that US participation was needed in order to strengthen the Bank’s capacity for dealing with the USSR” (Sanford, 2005, p. 20). While this was the view at the time of conception of the EBRD, during the Bush administration, US interest in the bank was even stronger after the collapse of the USSR and the anticipation of Russia’s strong relationship with the EBRD. Interviews at EBRD headquarters, February 2002. The structural adjustment policies of the World Bank and the IMF were heavily criticized in the late 1980s and 1990s by NGOs and policy makers. These policies were thought to have directly contributed to the debt crisis of developing countries, causing more strife and underdevelopment instead of contributing to advancement. During the Cold War, the primary interest of the US and the Soviet Union was retaining and expanding their spheres of influence regardless of the regime type or economic policies of their respective beneficiaries.

216

Political and Economic Constraints

European countries or the United States, furthering the EBRD’s donorinfluenced culture. A subtler look at the “working culture” of the RDBs reveals a significant change in approach between the earlier-generation RDBs and the newest one. Whereas the IDB, AsDB, and AfDB are mostly staffed with economists and policy-makers with development-work backgrounds along with practitioners who worked in developing countries, at the EBRD staffing is more reminiscent of an investment bank. Located in the City of London, Europe’s financial center (as opposed to the IDB which is located in Washington, the American political hub), the EBRD prides itself on hiring experienced bankers. While there are top-positioned economists in the EBRD’s administrative hierarchy, an interview with a senior economist at the bank’s headquarters underscores a matter-offact, bank-like, approach that is concerned mostly with making sound financial decisions.53 Although nationals of Eastern European countries hold posts based on quotas, many of them, according to accounts by various EBRD officials, come from the banking world. In addition, the EBRD works much more closely than the other RDBs on joint collaborative projects with private institutions and makes loans available without requiring government guarantees. While the older institutions have been making a transition to more private-sector-based investment and lending practices, their movement has been slow and cumbersome in comparison to the EBRD’s institutionalized modus operandi. The EBRD, then, represents a new model of multilateral development bank, where despite its multiple hegemonic players and its straightforward political agenda, there seems to be more delegation from the donors to the institution. This may result from a hands-off philosophy, rather than from the fact that at the EBRD, the interests and approaches of the institution (staff) and the hegemons are more closely aligned than in the other regional development banks: the bank-like mentality of the staff goes hand-in-hand with the donors’ drive to promote market economies. To that extent, the EBRD’s power structure is markedly different than the unipolar, bipolar, or nonhegemonic structures of the other RDBs. Its multiple hegemonic character suggests that the major donors engage in balancing and that negotiations to avoid stalemates are more complex as 53

Interview at the EBRD headquarters in London, January 2002. See also 2012 Standard & Poor’s report on EBRD financial solvency (in Supranationals Special Edition 2012, especially pp. 74–75).

6.1 Political Interests and RDBs

217

they involve more actors. For example, in 1995 the total number of shares held by borrowing members was 12.34 percent. The shares held by nonborrowing, European countries constituted 63 percent and those held by nonborrowing nonregionals were 24.65 percent (this has not changed much in the past few years). Of the borrowing members, the only country with a significant number of shares is Romania (4.05 percent). The rest of the borrowing members hold very little influence – mostly less than 1 percent of the shares each. Of the 25 regional nonborrowers, France, Germany, Italy, and the United Kingdom are the only noteworthy, substantial shareholders (with 8.62 percent each in 1995).54 In 2012, with some new shareholders joining the EBRD, the shares of France, Germany, Italy, and the United Kingdom were only slightly decreased (to 8.52 percent) – with all four preserving their equal shares. As for the nonregional donors, of the existing 9 in 1995, the United States and Japan have the most votes with 10.12 percent and 8.62 percent of the shares respectively, with only a slight decrease by 2012 – the United States to 10 percent, and Japan to 8.52 percent.55 Thus, with five major shareholders (four of them analyzed in Chapter 5 – United Kingdom, France, Germany, United States), each holding relatively equal shares, there is no dominating power (at least in terms of voting power). Further, the diffusion of voting power between other donors, regional and nonregional, makes solving disputes, if there were any, difficult.56 In practice, consensus between these important board members is sought prior to voting. This is simplified by the EBRD’s high degree of delegation granted to the administration of the organization. In fact, even board members do not have access to information about interest rates on loans (interest rates vary depending on the loan’s destination and project). Since various countries/projects have different interest rates (margins), this 54

55

56

Once the United States decided that EBRD membership was a national interest, the “Bush Administration sought and obtained for the United States the largest single country voting share (10%) in order to give the United States a significant voice – though the EU countries together have 56% of the vote – and to more firmly link the United States with Europe (Sanford, 2005, p. 21). I did not include Japan in the hegemonic analysis in Chapter 5 because Western European powers and the United States were the architects of the EBRD and, according to interviews with bank officials, have the most political interest in the institution. Japan’s donor status, if anything, contributes to a preference for sound banking practices rather than political influence, and its interests are mostly associated with financial access and procurement. As one EBRD chief economist states: “There has not been a case that went to the Board and was rejected. Otherwise, it would not have been sent to the board” (Interview, January 2002).

218

Political and Economic Constraints

information is determined, calculated, and approved only at the banking level. The board is responsible for general policy guidelines, and is not involved in the day-to-day activities.57 The institution is run like a private investment bank, where the board oversees the general activities and meets for replenishment discussions, but has very limited input when it comes to assessing projects and approving loans.58 The dominant issues addressed by the board, according to EBRD officials, are corporate governance, environmental policy, and the institution’s Articles. As for the risks undertaken by particular loans, the board has very limited involvement. Although members of the board must approve every project, the bankers’ units gather the information and process the initial selection of projects slated for approval.59 The multipolar power structure of the EBRD is also evident in the reputation of major shareholders within the bank. For example, staff and borrowing members do not hold the United States in high regard.60 As one EBRD banker explained, it is clear that the United States has an agenda to have a preference for medium-sized projects that are supported by the NYSE (New York Stock Exchange) and thus objects to smaller microloans to the private sector as well as to larger, government guaranteed loans. Conversely, staff and borrowers generally view the EU members more positively. The other nonregional members usually keep away from policy

57

58

59

60

“No member of the Board is aware of the price of a loan” (Interview with chief EBRD banker, January 2002). In interviews held January 2002, various officials at the EBRD stated that the board rarely involves itself in decisions on specifics, and that the staff of the bank is able to present proposals to the board taking into account the members’ preferences, thus avoiding dispute and prolonged negotiations. Although boards of the other RDBs are also not involved in daily operations, interviews with OECD researchers familiar with all four RDBs (2001–2002) suggest that the EBRD stands out as the board least involved in specific decisions about policy. While the EBRD’s president is always European, its vice president is always American, and, it seems, usually comes from the banking and business community rather than the political community. For example, VPs have included Ron Freeman (Citibank), Charles Frank (GE Capital), and Noreen Doyle (Merrill Lynch and Banker’s Trust). They are not appointed by the US government, but merely approved. Various accounts of EBRD officials state that the bank is not important for Washington, and that it is de facto a “European Bank” with European countries in control. However, by the same accounts, officials concede that there is overall US interest in the banks’ operations, particularly as they pertain to Russia. As stated earlier, American interest in the EBRD may have changed in recent years – with the apparent redirection of the EBRD to expand its operations to North Africa and the Middle East – to countries undergoing democratic transitions outside of Europe.

6.2 Why Be a Donor?

219

decisions unless they have a direct interest, such as Korea’s opposition to steel projects in Romania that might pose competition to its own market. Thus, the EBRD’s mission, despite the shareholders’ political game, is the promotion of business – not just among regional shareholders and borrowers, but also among nonregional shareholders. By “selling” the bank through various marketing strategies, the EBRD operates like commercial, investment bank. In doing so, the EBRD promotes private bidding from a variety of private donors (in addition to shareholders), enlarging the circle of those involved in the development process. In sum, while a multiple hegemonic structure could lead to policy stalemates and competing coalitions, the EBRD membership and voting structure produce efficient outputs in accordance with the banking ethos of the development institution. Substantial delegation allows managers and staff to make loan policies, basing them on board approved guidelines. Since the EBRD’s general preference for promoting market economies rests in the nexus of consensus between shareholders and staff, the hegemonic configuration is of little consequence to decision-making.

6.2

why be a donor?

In addition to carrying out their mission, RDBs have to navigate the domestic political fluctuations of their shareholders. Nonborrowing shareholders (states) have to answer to domestic procedures and regulations that guide financial allocations for aid. Consequently, the substate domestic political process precedes shareholders’ formation of policy preferences vis-a-vis the RDBs. These preferences range from the general characteristics of the institutions’ policies (such as environmental concerns, labor practices) to the outcome of board approvals and loan disbursements. As individual member countries alter their policies toward development and global economic relations (including through domestic legislation that can legally bind them), their positions on various aspects of RDB lending change accordingly. This political environment poses a challenge to the RDBs. While it is commonplace to regard member countries as cohesive units when assessing them in the context of IFIs, it is essential to keep in mind the often complex domestic policy-making procedures that generate MDB guidelines. For managers and staff of the RDBs, the domestic political debates of their shareholders can be both advantageous and problematic for advancing the development agenda they advocate. On the one hand,

220

Political and Economic Constraints

understanding the intricacies of the decision-making process of major donors such as the United States helps knowledgeable staff (and representatives of borrowing members) manipulate the system to their benefit. On the other hand, a cumbersome domestic process for donor-hegemons may also negatively impact the development banks’ ability to adapt to new geopolitical and economic environments, and potentially motivates the RDBs to adhere to mainstream lending practices that are not challenged by powerful special interests in donor countries. The United States for example, via the Working Group on Multilateral Assistance (WGMA) has held routine review meetings concerning the RDBs, included in the group’s assessment of all MDBs and their loan proposals. “Every Thursday, staff from Treasury, State and Commerce Departments and USAID meet to review all loans and grants that are scheduled for consideration by the banks’ executive boards during the next two weeks” (Sanford, 2005, p. 4). Treasury chairs this meeting and “it examines the technical and policy dimensions of all prospective MDB loans and the applicability of legislative mandates requiring the United States to oppose certain kinds of loans or loan in certain kinds of circumstances” (ibid.).61 For forty years, these meetings have been a staple of the United States government’s concerted effort to align its MDB policy with its own legislative requirements. And as Sanford further notes, “A few issues may go to the Cabinet secretaries or the White House for resolution, but that is rare” (ibid.). In these deliberations, Treasury usually has the upper hand. First, the documents are extremely long and detailed and few staff members manage to study them carefully.62 Second, Treasury typically informs representatives from other agencies how they intend to vote (and why) on each loan. And while other staff members at times express their agencies’ concerns, their role usually amounts to communicating Treasury’s position to their superiors. It is rare that Treasury would change its position as a consequence of other agencies’ viewpoints.63

61

62

63

This is also based on an interview of Jonathan Sanford at the Congressional Research Service (CRS), August 2005. For example, as Sanford notes, the “WGMA agenda for its June 24, 2005 meeting included 87 loans or grants, 17 where economic or policy implications were the principal concern and 70 where legislative mandates might be applicable.” All 87 loans were scheduled for consideration by MDB boards by July 8 of that year. Overall, the WGMA examines more than 900 loan and grant documents per year (Sanford, 2005). See also Sanford, 1982. Sanford, 2005; Babb, 2009.

6.2 Why Be a Donor?

221

The next two sections explore the motivation behind hegemon(s) and other donors’ interest in RDB membership. In a global economy where aid can be administered via numerous channels, these sections explain the advantages donor countries find in multilateral lending via MDBs.

6.2.1

Political Motives of Major Donors

In the 1960s the United States was the sole “net” donor of the IDB and held 42.05 percent of the voting power (Tussie, 1995, p. 3). At that time the IDB displayed preference to state-led programs, in support of importsubstitution-industrialization (ISI).64 It was also during a time when Cold War considerations were high on the US Administration’s foreign policy agenda. This was one of the primary reasons the United States backed the creation of the AfDB and the AsDB following that of the IDB (see Chapter 3). However, the political considerations of a hegemon do not preclude an insistence on sound financial practices, which may impact the institutional policies in such a way that they end up falling closer to “bank” on the development agency/bank spectrum.65 Most nonborrowing donors (but especially hegemons) view IFI membership as an opportunity to (potentially) exert influence over the borrowing members of the institution.66 The ability to approve and disburse loans to countries that seek credit gives the RDBs leverage. For donor shareholders this influence is an opportunity for political gain. However, as long as the major political objectives are fulfilled, donors direct their attention to concerns about the financial aspects of the RDBs’ operations. Therefore, the fact that they delegate to bankers who favor sound financial policy does not mean donors are disengaged from or indifferent to the development banks’ operations and outputs. In fact, managers and staff usually follow donor demands in keeping the institutions operative in this way by making sure loans are disbursed, defaults do not occur, and donor commitments are not called upon.

64

65

66

The popular ISI strategy for development in the 1960s and part of the 1970s encouraged states to develop and protect infant industry by restricting imports. This required government investment and spending and deemed too costly by the mid-1970s. The financial soundness of the RDBs is one of the central political concerns of major donors. As we have seen in Chapter 5, this may not happen in practice. But hegemons often believe that they can use IFIs for influence, or they use this argument to convince domestic constituencies that IFI membership is in their interest.

222

Political and Economic Constraints

More than streamlining the governance of the development banks, maintaining sound financial practices enables donor states to justify domestically their monetary support of the institutions, arguing the advantages of “multilateralism” and the efficiency and effectiveness of the multilateral bank setting. This is an important consideration because the finite domestic resources earmarked for aid are divided between various multilateral institutions, as well as bilateral aid agencies. When RDBs generate a positive image, and the returns on investments are aligned with donors’ economic interests (in addition to meeting their political motivations), donors can more easily “sell” their support for the institution to domestic constituents. Donor countries, then, weigh the available alternatives for aid provision to developing countries. Taking into account potential political benefits (such as influence over policy of an institution or a borrower, or their own reputation) as well as expected economic benefits (such as securing an investment environment for their private sector), Western countries on the boards of the RDBs attempt to minimize the cost of providing financial aid to developing countries by maximizing benefits to their own political and economic objectives. A number of channels of aid are considered part of donors’ Official Development Assistance (ODA). While each lending/aid source is unique (WB, RDBs, and individual states), once disbursed, loans and aid overlap in their geographical destination. In contrast to bilateral aid scenarios, the RDBs provide donors with an aid disbursal infrastructure that includes information-gathering reports, institutionalized mechanisms by which to administer transfers, and servicing of the loan-making process. These are all elements that are attractive to donors because they can view the institutions as banks that deliver the kind of services and support needed for their investments to succeed, in addition to reducing transaction costs and providing experts in country offices.67 RDBs, then, offer a specialized aid system comprised of a vast bureaucratic professional network. Bilateral aid agencies that are a part of individual countries’ foreign policy apparatus cannot be as cost-effective. The daily activities of the RDBs are dispersed across regional offices, where the staff is comprised mainly of locals.68 67

68

RDBs have regional office in every borrowing country. See Chapter 2 for a discussion of principal-agent theory, explaining the logic behind principals’ preferences in delegating authority to an agency. See Milner (2006) (in Darren Hawkins et al.) and Milner and Tingley (2010) for a discussion of bilateral versus multilateral aid. The regional offices are central to establishing a reputation of “closeness” to the borrowing members and of catering to their needs. However, in some instances (especially in

6.2 Why Be a Donor?

223

Further, policy makers often consider RDBs advantageous as expert multilateral lenders due to their regional specialization, a feature that sets them apart from the World Bank.69 Still, RDBs’ leverage to satisfy donors’ interests is not without controversy. For example, the most contentious standoff between the United States government and the IDB was when the Reagan administration reached an almost insurmountable impasse with the development bank during the Seventh Replenishment negotiations (see Section 6.1.1). 6.2.2

Other Donors

Most donor members at the RDBs are not hegemons. Their political gain from RDB membership is threefold: First, they want to be part of the “club” – there is a desire not to be left out of an organization once it is created. Board membership awards donors with a forum to interact and lowers transaction costs for aid. Furthermore, membership affords access to information about development projects and RDB development policies. The opportunity to be on the board and have access to that information is important for many countries, even if influence over decisionmaking is limited. This is especially the case for those countries that have business interests in the borrowing region of the RDB.70 Second, the financial commitment integral to membership contributes toward meeting the 0.7 percent of GDP ODA target.71 Politically, getting closer to the 0.7 percent of GDP is a benefit when it comes to national reputation: by meeting or approaching to the 0.7 percent target contribution, countries signal their intent to honor the commitment they make in their IO memberships – in this case, the UN and OECD that set these aid standards. RDB managers have been strategic in making this case to donor members, urging them to increase their financial contribution to the RDBs thereby fulfilling their international aid obligation. The IDB, for example,

69 70

71

Africa) the regional offices are obsolete since their lines of communication with the central headquarters, other offices, and experts are poor, and the movement of officials between offices is very difficult due to lack of infrastructure and direct flights. This was one of the central points raised by the 2000 IFIAC Report. See Martin (1992), Abbott and Snidal (1998), and Mansfield and Pevehouse (2006) for more on the reasons states join International Organizations (IOs). While the former two discuss coordination and comment interests, the latter focus on domestic interests in forging the way for states to join IOs. In 1970, the 0.7% ODA target was first set (by UN and OECD/DAC) and has been repeatedly re-endorsed at the highest level at international aid and development conferences, most recently in the 2005 Millennium Development Goals.

224

Political and Economic Constraints

has been establishing funds that make contributing attractive for donors: “when their commitments count towards the Millennium Development Goals, and at the same time do not increase government deficit, it is worthwhile for members to maintain or increase their contributions to the bank” (Carlos Jarque, interview at IDB European Office, April 10, 2012). Further, Mr. Jarque stressed that Latin America presents an opportunity for donors – where contributing to special funds yields returns. In Haiti, for example, more than 70 percent of schools are private. Development, then, cannot only be through grants, and a combination of 20 percent grant and 80 percent reimbursable loans would not be available on the market from private financing.72 Brian Atwood, Chairman of the Development Assistance Committee (DAC) that oversees bilateral aid at the OECD confirms this. He states that when the DAC puts pressure on its members to increase their bilateral aid to meet the GDP benchmark, it often finds itself competing with the multilateral development banks that found ways to incorporate special funds into their portfolios, making the multilateral institutions more attractive from an investment standpoint.73 Third, RDB membership is attractive to donors because opportunities for influence are not directly linked to voting power on the board of the institution. Power in the RDBs is not necessarily measured by actual votes, but is often engendered in the deliberations and negotiations that take place prior to the votes.74 These negotiations may afford nonregional smaller donors (as well as borrowers) more leverage than their actual shares reflect when issues that are important to the hegemon(s) in a bilateral or other comparable setting are linked to the decision-making at the RDBs.75 Furthermore, in the context of information access, for hegemons and other shareholder members alike, relying on the staff of the RDBs for research and reports substantially diminishes the significance of the difference in power based on voting shares. Finally, potential economic gains draw nonborrowing shareholders to the RDBs. Membership status provides donor countries procurement rights, meaning that only companies based in member countries can

72 73 74

75

Jarque, interview at IDB European Office, April 10, 2012. Atwood, interview at DAC Headquarters, OECD, April 12, 2012. A point emphasized by Leo Harari, the IDB’s special representative to Europe (interviews November 2001–March 2002). This is a situation that is similar to a “pivotal party” in coalitions.

6.2 Why Be a Donor?

225

compete for bids. This is a compelling financial incentive, particularly when callable capital remains just that – callable. Therefore, with a contribution of the paid-in portion of their commitments, donors gain greater access to markets for their private sectors.76 What’s more, access to procurement contracts for approved projects opens the door to private companies for further direct business in the RDBs’ borrowing countries, compounding the economic gains of donor participation beyond the borders of RDB lending. Creating business opportunities for their domestic companies was especially of interest to donors in the 1960s and 1970s when the bulk of projects were in the infrastructure sector. Projects of this kind are typically completed more quickly and fully paid for, including significant profit margins. Although difficult to measure, this is one explanation for the RDBs’ struggle to lend more to social sectors that are not “bankable.” The method of project lending favored by IFIs for the majority of their operating years is designed to be attractive for donors particularly in terms of the procurement rights available to them. Conversely, long-term projects with small profit margins on investments are less attractive for the private sector, partly because there are fewer procurement opportunities.77 At the same time, as we have seen in Chapters 4 and 5, increased FDI volumes to developing countries imply that the RDBs now have private competition for those projects that are potentially more “bankable.” Nonregional donor presence helps ensure that the development banks’ policies are financially responsible.78 Furthermore, nonregional donors help credit ratings. Both the RDBs and the donors have a strong interest in sustaining high credit ratings, and rating agencies, whose assessment has substantial implications for the RDBs’ financial health, place great value on the involvement of non-regional donors. Typically, the identity of nonregional donors signals relative financial solvency, the degree of financial risk, and thus the creditworthiness of the institution. With that in mind, a secure foundation for regional banks’ operations is made possible

76

77

78

This explanation for attracting donors was reiterated by numerous current and past RDB officials (interviews 2000–2002, 2005, 2012). Infrastructure projects are more profitable because costs are typically associated with labor-hours and equipment, much of which is available for multiple uses and thus its cost is discounted. These projects are potentially very lucrative. This motivation on the part of donor shareholders implies factors that go beyond considering the risk level of loan: they consider the potential profit for their domestic private sector. See, for example, Sherk, 1999, p. 1.

226

Political and Economic Constraints

when RDBs maintain a posture based on the experience and expertise that is readily available in donor countries, coupled with a certain “distance” that invites “objective” and arguably more financially sound decisions.

6.3

power to borrowers?

One of the initial reasons for the creation of the RDBs, from the viewpoint of the borrowing countries, was to establish institutions in which developing countries are not as marginal as they are at the World Bank.79 RDBs governance structure has given borrowing member countries unique stakes in the development banks: borrowers have a sense of ownership, responsibility, and a level of commitment to the banks greater than that of the WB.80 Thus, even if RDBs’ structure and strategy resemble the WB, and the preferences of dominant nonregional members take precedence, the process that carries out these objectives positively impacts borrower engagement and satisfaction. For example, as Tussie (1995) contends: “The IDB can make the donor’s agenda compatible with national and regional sensitivities; it can then evaluate the region’s agenda and make it presentable for funding” (p. 1). By their own accounts, developing countries wanted an active, central role in the decision-making processes related to the development assistance headed their way. Addressing this concern, the RDBs, with varying degrees, allowed borrowing member countries to play important roles in the institution designed to aid them. But even with similar initial parameters of institutional design, the RDBs evolved in different directions – not just in terms of the actual development outputs and the role played by developing members, but also in terms of institutional standing and reputation. Conceptually, it is reasonable to assume that borrowing members would represent their needs in the most development-effective way. Indeed, this view has been corroborated in the IFIAC Report, and is advanced by many of the RDB officials interviewed for this study.81 The argument developing countries in Latin America were making in the 1940s and 1950s to promote the establishment of a regional bank was that 79

80

81

This is the case for the IDB and AfDB. And, in the case of the EBRD, a frustration with the classic regional development bank model as well. Interviews with IDB, AsDB officials 2000; Jean-Claude Berthelemy (OECD Development Centre, Africa economist); interviews 2001–2002. See Dell, 1972; IFIAC Report, 2000; Culpeper, 1997; Tussie, 1995; Kappagoda, 1995; and English and Mule, 1995.

6.3 Power to Borrowers?

227

by including borrowing countries as members with decision-making power, the bank would better serve the region’s development needs. Findings in Chapter 5 and the earlier discussion in this chapter demonstrate that the Articles of Agreement and the financial parameters within which the RDBs operate influence the extent of borrowing members’ ability to exert power. For example, the analysis of the AfDB showed that even when an institution is predominantly controlled by regional borrowers, concern for maintaining funding and guarding credit ratings motivate a conservative lending policy (and one that often puts considerations of “bankability” above those of the borrowers’ development needs). All this suggests that the mere inclusion of borrowing members on the boards of the RDBs, even when borrowers hold the majority of the voting shares, does not necessarily promote development oriented policies. If borrowing members’ authority in the development bank does not guarantee development-driven policies, why was it important to structure them in this way? There are several explanations for the motivation behind the inclusion of borrowing members as RDB shareholders. First, and most obvious, at the RDBs’ creation and during their early years, borrowing and donor members truly believed that borrowers’ greater participation would result in development-driven policies better-suited for their respective regions without compromising donors’ interests (especially during the Cold War era). Second, the RDBs benefit by including borrowing members more actively in the governing and implementation processes – they legitimate the institutions’ policies as much as donors’ memberships contribute to strong credit ratings.82 Third, perception of the banks’ operations by donor governments and their own decisionmakers is important:83 The inclusion of borrowers as potentially influential board members allows donors, if the need arises, to make the argument that policies advanced by the institution were a result of borrowing members’ preferences. Given all this, to what extent have the governments of borrowing member countries been part of the process? As discussed, much of the daily work is delegated to the staff, while hegemons oversee the entire decision-making process and overall operations of the RDBs. Since

82

83

Critics, however, claim that borrowing board members (that is, developing countries) only legitimize donors’ decision-making in undertaking activities that are, ultimately, not very different than those of the Bretton Woods institutions governed by the “Washington Consensus.” See Upton, 2000.

228

Political and Economic Constraints

internal decision-making and procedural arrangements determine the RDBs’ outputs, and, therefore, their place on the development agency/ bank spectrum, it is important to understand how decisions are made and the extent to which borrowing members have influence. The decision making process can be broken down to two important components: (1) The players that participate in the process. (2) The rules and guidelines to which these players must adhere. These two elements essentially define the world of possibilities available to the decision makers, borrowing members among them. The rules and guidelines, as articulated in the RDBs’ Articles of Agreement, provide a framework, while the players insert their interests, both political and economic. Since the Articles of Agreement provide only general guidelines that are prone to interpretation, the participating players, that is, states, can often exact decisive influence on the institution’s ultimate policy decisions. Moreover, the RDBs’ staff operates as internal agents; exercising certain powers that were granted them through delegation (see Chapter 3). This is particularly true of the development banks’ administrative leadership, especially the president. However, since most of the RDBs’ operations are distributed to field offices that provide specific local information, and then go through various substantive and financial approval stages in the field and headquarters, final loan decisions are based on the work of many staff members, thereby increasing the opportunity for borrowers’ input. As for the Articles of Agreement, that of the AfDB, for example, states the following: “The purpose of the Bank shall be to contribute to the economic development and social progress of its members – individually and jointly” (chapter 1, Article 1).84 In the explanatory notes to this Article, the founders state that these guiding definitions and purposes are interpreted according to the example of the IBRD (World Bank) and IDB (ibid., p. 96). This article is intentionally vague in its definition of the terms described (e.g., economic development and social progress), leaving the actual decision-making to the staff and members of the institution. Additionally, the added explanatory notes emphasize the close connection between the AfDB and the institution it was modeled after: its structure

84

United Nations Economic Commission for Africa, Agreement Establishing the African Development Bank, 1964, p. 96.

6.3 Power to Borrowers?

229

was not created in a vacuum, but rather based on the existing MDBs (IBRD/WB and IDB).85 Furthermore, the Standard and Poor’s report of the financial standing of the IDB states: “The bank’s franchise value rests upon its role as a stable source of long-term funds for developing nations at narrow spreads over its cost of funding, enhanced by its willingness to lend during times of financial stress and its expertise in project lending and public administration. It may also benefit from regional member countries’ feeling that this is ‘their’ institution” (Standard & Poor’s, 2002, p. 3).86 Whether accurate or not, borrowing members’ perception of their ability to influence institutional policy is a political asset to the RDBs. As is the case for any bank, let alone for public institutions such as the RDBs, shareholder and public confidence is essential: it helps secure financial resources and aids in creating partnerships for projects and programs. Thus, the RDBs carefully maintain an appearance that borrowers are incorporated as partners-inpolicymaking. This serves shareholder countries by validating the necessity of the RDBs based on a unique structure that is inclusive of borrowing members. Donors certainly have an interest in sustaining the appearance of inclusion, but the extent to which borrowing members set the agenda for policy is difficult to discern and changes over time. Furthermore, governments of borrowing members have changed over time, and leadership interests often reflect the interests of donors more than particular domestic interests, if those conflict. Latin American countries, for example, have undergone significant changes since the IDB began lending to the region. According to officials interviewed for this study, the IDB is highly visible in Latin America where local constituents trust that regional members govern it. Leo Harari, a former head of the IDB’s Special Office in Europe, noted that the bank oversaw four major policy-related changes over the years that made an impact on the way developing members interacted on the IDB board as well as how their position was perceived by the general population of those borrowers. First, it was a development bank working in a region of mostly military regimes. Second, it oversaw the transition to democracy in Latin America, and although not directly responsible for this change, the democratic developments were a partial product of economic factors to which the 85

86

References of similarity to the IBRD and the IDB appear in explanations to most of the Articles in the agreement establishing the AfDB. Emphasis added.

230

Political and Economic Constraints

bank contributed. Third, the IDB became an integral part of the “Washington Consensus,” under the close scrutiny of the United States Treasury and other financial institutions.87 And fourth, most recently, it has become subject to IMF supervision (the IMF approves the criteria for loan-making mostly due to economic crises in the 1990s in which the IMF had to step in as a lender of last resort; thus, by seizing control over terms of loans made by MDBs other than itself, the IMF is trying to manage its possible future interventions).88 All this points to a limited role for borrowing members. In fact, all of these stages reiterate the centrality of the bank to donor interests, and in particular the United States during the Cold War (working with military regimes; transition to democracy) and post–Cold War (Washington Consensus and IMF supervision). Borrowing members’ role in the RDBs has also been affected by changes in the global political and economic climate. Despite this, the IDB has managed to retain a reputation of a development bank guided by borrowing members’ preferences in its policy output, to a large extent due to the high profile of top managers (and the president) who are Latin American national and very well connected in the region.89 For example, the first loan ever made by the IDB was for water sewage in the mountains of Peru. This loan was socially responsible and progressive at the time – targeting the improvement of the quality of life of impoverished peoples. The IDB has since managed, to a certain extent, to retain its image as a progressive development institution, and one that is firmly “Latin American.” In contrast, the fact that African countries have been leading the AfDB’s policy decisions has not necessarily made it a better institution in the eyes of Africans. As discussed in previous chapters, nepotism and corruption – whether real or perceived – pose a problem for the AfDB, and not just from the perspective of donors. Moreover, reduced donor appeal affects the amount of capital commitments and contributions, the level of technical assistance and expertise, and the overall support and efficiency that can be derived from closer collaboration with developed countries. It is noteworthy, then, that the RDBs position on the “development agency” end of the development agency/bank spectrum does not directly

87 88 89

For more on this see Stiglitz, 2002. Based on numerous interviews with Mr. Harari in 2001 and 2002. These changes include Latin American regime transitions in the 1980s and 1990s, OPEC oil crisis in the 1970s, economic recession in the 1980s, the end of the Cold War, and the most recent 2008 economic recession.

6.3 Power to Borrowers?

231

correspond to regional members’ voting shares or influence over the institution: at times, more control by the regional members can be counterproductive to the goals of the institutions.90 Generating a perception of “ownership” does not necessarily equal “control.” At the IDB, for example, regional members are commonly viewed as the bank’s owners despite the United States’s powerful status on the board and in voting shares. In contrast, where borrowing members of the AfDB are both owners and have control of the institution, the struggle to balance political and financial factors that influence the development bank’s policies remains challenging: although borrowers dominate decision-making, they face the same financial constraints as other development banks. Thus, even with regional control and ownership, concerns about funding, credit ratings, and loan repayments remain central to the AfDB’s board and management. Of the four RDBs, the AsDB and the EBRD are the ones with the least active involvement of borrowing members, as analyzed earlier. The AsDB, working in a region that has experienced impressive growth and economic development, manages to maintain its good reputation partly because of its conservative financial policies, which reinforce its standing as professional rather than politicized. In this way, the development bank has retained the trust and support of borrowing and donor countries alike.91 Standard and Poor’s summarizes the rationale for the AsDB’s high creditrating in the following way: “Solid financial profile; conservative financial policies; and strong shareholder support” (Standard & Poor’s, 2002, p. 13). The EBRD, which makes loans in a region that is experiencing a different development process than other regions, encourages its borrowers to adhere to the bank’s guidelines and to work directly with the bank’s staff, rather than via country representatives.92

90

91

92

For example, the power of borrowers does not automatically translate into a more constructive development record. “The AsDB has enjoyed a reputation for unusually conservative financial policies among the MDBs, including a conservative gearing ratio, low leverage, and high liquidity targets. Underlying the Bank’s financial performance has been the region’s economic performance, which has been better than that of the other developing regions” (Kappagoda, 1995, p. 121). Since the connection between the AsDB and Asia’s overall economic performance is subject to debate (along the lines of chicken-and-egg), this statement can only point to the observation that the AsDB has been performing well financially, and that this success has coincided with Asia’s overall economic strength. For example, a project proposal can be directly filed through the Internet – something that is not possible yet at the other RDBs.

232

Political and Economic Constraints

To sum up, developing countries, as decision-makers, can have both a positive and a negative impact on the institution: their central role may contribute to the institution’s reputation as a truly local organization, one that addresses the real concerns of developing countries and empowers the borrowers to take their future into their own hands. On the other hand, regional members’ insistence on control can deter donors from actively supporting the institution by limiting their capital subscriptions to it or by withholding committed funds, thereby effectively constraining the RDBs’ lending activities. Overall, while participation of borrowers in the RDBs governance is generally beneficial – both in terms of perception and in actual policy outputs, borrowers’ control of the banks more likely hinders the RDBs’ ability to be as effective (development institutions) as they could be because: (1) borrowing members’ enhanced decision-making role on the board of an RDB does not necessarily translate into policies that are more development-oriented and less “bankable,” because the institution faces common financial imperatives, whether policy is led by donors or by borrowers; and, (2) the RDBs face difficulty raising funds as a result of donor marginalization.

6.4

wealthy states, poor states, and rdb structures

Under their current mandate and structure, RDBs have the potential to be important contributors to development. Unlike private banks, the RDBs are not primarily (and officially) concerned about profits or diminishing returns. This should allow them to make loans to projects that are not expected to reap quick financial returns. It also eases lending to public institutions in poor countries and funding for projects in less lucrative sectors (e.g., health and education).93 Furthermore, unlike the WB, which determines repayment periods for loans based on the recipient country’s financial indicators, the RDBs institute a project-based repayment program. In the case of the AsDB, repayment periods range from ten to thirty years, not including grace periods that can last up to ten additional years. The IDB and the AfDB employ similar policies, while the EBRD directs more of its funding to the private sector with much shorter maturities (at the end of 2002 less than 93

“Since MLIs are not profit-maximizing and are generally well capitalized, deciding to maintain ample liquidity is easier for them than for commercial institutions. The fact that it is accomplished at the cost of lowering returns on assets and equity is not a major concern to their shareholders” (Standard & Poor’s, 2002, p. 5).

6.4 Wealthy States, Poor States, and RDB Structures

233

2.2 billion of its 6.8 billion Euros in outstanding loans was sovereignguaranteed). The interest rates on the loans made by the RDBs also vary; most are lower than commercial banks, but not as low as the concessional funds or, obviously, grants. It seems to be easier for RDBs (particularly those with a concessional lending window) to use this concessional arm to cover defaults on OCR loans. This has been done before (e.g., during the 1997 Asian Financial Crisis) and can conceivably be repeated.94 In their efforts to keep both donor countries and borrowers satisfied, RDBs attempt to bridge effective investment and the development needs of borrowers. But these two roles have often been at odds. For example, in February 2005, the IDB published a report on “The New Lending Framework: Assessment Report in Recommendations.” In it, the development bank attempts to outline a plan for future lending. This came on the heels of decreasing net financial flows from the IDB to borrowing members and an admission that with growing resources available to its borrowers, the IDB had not been optimal in meeting its mission goals.95 The RDBs are important to the interests of the hegemons as much as they are important to the borrowing members. And although the organizational structure preserves the power of major donors and hegemons in some of these institutions, the uniqueness of the RDBs is in the subtlety of the dialogue between donors and borrowers – one that cannot be quantified, but that certainly exists and influences the way these institutions operate. Further, the RDBs reflect the variations in hegemonic influence that stem from hegemonic configuration: the unipolar hegemony of the United States in the IDB, the bipolar hegemony of the United States and Japan in the AsDB, the multipolar hegemony of the EBRD, and the nonexistent hegemony in the AfDB. The role of the hegemon(s) is modified by the institutional design, delegation, and a culture that develops over time. In conjunction with the institutional staff attributes, the configuration of hegemonic influence on the boards of the RDBs can greatly impact the policies they follow as well as their ability to change. While Chapter 5 presents statistical results of the activities of each of these institutions, this chapter explains in detail why each of these institutions might (or might not) have different outcomes. These differences

94

95

By doing so, the RDBs emphasize the centrality of the OCR lending window to their policy (outputs): Repayments on OCR loans are necessary for the RDBs to operate under this design, and the nonconcessional lending window is sometimes used to support OCR lending (prevent default) rather than real investment. Committee of the Board of Governors, IDB, 2005. Document: (AG-1/02, CA-450(3/03).

234

Political and Economic Constraints

arise, in part, because of the varying donor-borrowing configurations, hegemonic involvement, and the ability of all constituents to identify joint interests. It is evident that while some observers of the IFIs stress developing countries’ powerlessness in the face of pressure by donors, these institutions (and borrowing members) also benefit from the involvement of rich countries, and, I argue, would not be able to function effectively without them.96 In fact, representatives from the US Treasury affirm that the United States has a clear agenda for the MDBs, advanced by a special office at the Treasury that is charged with supervising and formulating the US agenda for these institutions. Poverty alleviation, an interest the US agency promotes, is often overlooked by critics. Not only does development expand the markets for US goods, but it also helps create a safer international environment at a time when extreme poverty and conflict in developing nations is a growing concern for Western countries in their efforts to enhance security. Moreover, global issues such as the environment, epidemics, and financial crises are all exacerbated by poverty and underdevelopment – and even more so in war zones and during recession. Thus, development assistance is not simply or even primarily an altruistic endeavor for the United States: it has been consistently considered a national security preference. However, this ideal nexus of poverty alleviation/ development, where both donors/hegemons and borrowing member countries can maximize their joint political interests is seldom met because of the financial constraints imposed by the institutional structure of the development banks. Whether developing countries are truly integrated into the activities of the RDBs and empowered to make decisions that impact their development, or whether they only provide a “stamp of approval” for decisions made by donors, differs from one institution to another, as well as over time in each of the development banks. Furthermore, a central point made in this chapter is that even when integrated, the decisions borrowing members make as shareholders may not be substantially different from

96

Mingst, 1990, concludes about the AfDB: “Attention to arenas and levels of political conflict within an organization permits us to cross the analytical foci dominant in the field of international organization. In dispelling the doctrine of economic neutrality so pervasive in the formal institutional structure of multilateral development banks, we rediscover the utility of examining structures and analyzing organizational behavior. We discover what roles the organizations could play in creatively solving substantive problems. We explore the extent to which these organizations only reflect hegemonic interests or could actually undermine hegemonic power” (p. 6).

6.4 Wealthy States, Poor States, and RDB Structures

235

those of donor members, given the institutional constraints involved (e.g., the AfDB). In sum, ensuring necessary donor support, RDBs must meet certain financial and political criteria. This means that they need to maneuver between the two ends of the spectrum – from the financial soundness of a “bank” to the development agenda of a “development institution.” They pursue both avenues simultaneously, but in different ways. The AsDB does so by preserving its conservative, donor-controlled reputation, and the IDB, by presenting a façade that elevates regional members, while the United States retains its power. The AfDB has not been very successful in finding a way to generate both donor and regional support, although there are signs that things are changing (including its recent credit-rating upgrade).97 And, finally, the EBRD’s different agenda and much more specific niche allows it to sustain its operations successfully (although not without critique), particularly given that poverty alleviation is not actually part of its mission. These institutions form regimes in which states interact, cooperate, and even pursue their selfish political interests, overcoming the “anarchic” tendency of international relations. But, crucially, once a part of these institutions, defection becomes too costly for both donors and borrowers. Thus, the multiple forces that bolster the RDBs increase the likelihood of their ongoing operations, even when the banks fall short of meeting their objectives or if the merit of their contribution to development becomes a “moving target” due to changes in the global financial and credit markets.

97

“The Board of Directors approved a comprehensive set of institutional reforms in 2006 and 2007. Implementation of these reforms, which continued in 2009, has: (i) strengthened the link between institutional priorities and resource allocation; (ii) enhanced institutional budget flexibility through increased fungibility and devolved authority; (iii) established a new accountability and performance framework by linking deliverables to Key Performance Indicators (KPIs); and (iv) introduced a consolidated, multi-year programming and budgeting framework. The institutional reforms cover enhanced corporate services delivery, budget reforms, decentralization, and improved operations business processes . . .” (AfDB Annual Report, 2009, p. 51).

conclusion Future Outlook

If only you can be completely for, or completely against! But to be against, you’ve got to have other hopes to offer to mankind. (Simone de Beauvoir, The Mandarins, p. 323)

As the gap widens between rich and poor, both within countries and on a global scale, the need to reduce it becomes increasingly urgent. In response, international institutions designed to address this problem have proliferated. For states, establishing these international institutions required a creative and delicate balance between the desire to tackle inequality (whether sincere or strategic) and the desire to see “returns” (both in terms of financial solvency and of development “results”). Regional Development Banks (RDBs), a subset of these institutions, demonstrate the challenges of this balancing act. These challenges are a product of the development banks’ main function: making loans. The RDBs face the dual pressures of, on the one hand, ensuring that their financial portfolio is sound by averting loan defaults, and, on the other, reducing poverty and contributing to development. A central puzzle guiding this book has been that these two goals are often conflicting – in trying to balance their banking constraints with the objectives set in their Articles of Agreement, RDBs’ policies often fall short of stated goals. This book’s analysis of the RDBs advances the fundamental question of whether international financial institutions can function as agents of economic development while saddled with banking constraints. The book’s findings demonstrate that RDBs function more like banks. To move closer to the “development agency” end of the spectrum, donor member countries need to alter the RDBs’ mode of operation. This requires 236

Conclusion: Future Outlook

237

a normative shift that would downplay the centrality of the credit ratings of the banks (and their financial efficiency), and instead emphasize the rewards (financial and otherwise) of successful development. In their current form, RDBs attract donors with a financially efficient model that offers them an outlet for development aid at a relatively low cost (and with substantial benefits to their private sectors). A paradigmatic shift would necessitate donors’ willingness to incur greater financial costs, at least in the short term, while placing a higher value on nontangible benefits such as reputation and security. This would entail acknowledging the realistic public cost of development and poverty alleviation by potentially increasing the paid-in portions of committed funds. Conversely, if RDBs’ structure and operational mode remain unchanged, we must alter the way we evaluate them and the standards to which we hold them. RDBs are meant to narrow the development gap by providing credit for development projects in their respective borrowing member countries. Their design as “development banks” is centered on the premise that by extending credit to projects that are deemed necessary for development but would not get comparable funding without the support of these public banks, RDBs provide a necessary public good that furthers development goals. In contrast to grants – which might create lasting dependency on economic assistance, require regular fund replenishment, and contribute to the moral hazard problem – loans provide the necessary credit while supposedly encouraging responsible spending (because the borrowers know they must eventually repay the loans). Therefore, RDBs are supposed to function as unique financiers whose interest rates on loans and loan maturities could not be matched by private creditors. Although this element of their distinctive role is not always transparent, has changed over the years, and differs from one bank to the next, the central added value shared by all RDBs has consistently been their commitment to offer credit where it would otherwise not be available. After all, as a public initiative, development banks are designed to support the public good of development, not to generate profits. Although RDBs are not in the business of profit maximizing, to function as intended they rely on borrowers to repay loans. In an ideal scenario, it is supposed to work perfectly: a sovereign loan leads to development that contributes to economic growth, which makes it possible for the borrowing government to repay the loan when it matures. RDBs’ annual budgets depend on the revenue generated by repaid loans along with member countries’ paid-in capital and returns on investments. These budgetary components are intertwined because their amount and

238

Conclusion: Future Outlook

value is directly associated with RDBs’ credit ratings, which in turn is based on the financial parameters of the institutions. For example, a loan default could prompt rating agencies to downgrade an RDB’s rating. Such a downgrade would negatively impact yields on investments and would potentially jeopardize paid-in contributions. Or, a shift in paid-in contributions could trigger a change in credit rating, which might also affect investment revenues, and so forth. Thus, in prioritizing high credit ratings, the RDBs focus on the likelihood that loans made will be repaid and on assembling donor members (and hegemons) with advanced and reliable economies, since rating agencies also take into account donor credit ratings when issuing the RDBs’ ratings. Like other IFIs, RDBs rely on financial contributions from donor countries. The countries that typically provide the bulk of the financing do not borrow from the banks and membership in the RDBs is contingent on donors fulfilling their own political and economic interests. Setting global incentives for development contributions – such as the UN/OECD-recommended ODA target of 0.7% of GDP – motivates countries to consider financing the RDBs as a way to meet these contribution targets. But since other ODA options exist, including bilateral transfers that afford countries full control of development assistance tailored to their preferences, RDBs must offer an attractive “product” that appeals to donor countries and sustains their support. Consequently, an institutional model that requires only a small portion of the committed capital to be “paid in,” plus the benefit of procurement rights, and a bolstered reputation for the donors, are all packaged by RDBs to ensure donor support. For this model to work, RDBs are pressured to meet AAA credit rating standards, which prevents them from pursuing an agenda that is driven by the most urgent development needs of recipient countries. The research findings reported in this book demonstrate the difficulty in balancing these banking demands with development objectives. The amount of foreign direct investment (FDI) countries receive reflects the changing global financial environment impacting borrowing countries. Testing the association between RDB loans and FDI helps identify the banks’ added value to the countries receiving their credit and locates the RDBs on the Bank/Development Agency continuum. This book’s analysis demonstrates that loans are associated with FDI at the IDB, AsDB, and EBRD. Hence, borrowing countries of these banks that benefit from higher FDI amounts are also more likely to receive higher loan amounts. This finding places these RDBs closer to the “Bank” end of the spectrum because attracting more FDI suggests an increased likelihood for

Conclusion: Future Outlook

239

credit availability from private sources, and countries with higher FDI typically have the more stable and robust economies of their regions. Likewise, the association between loans and social indicators of a country’s level of development (health and education) is indicative of RDB inclination to support countries with more developed social sectors. None of the RDBs had a definitive preference for lending to countries with lower school enrolments and higher mortality rates, even though this preference would have been expected of a development institution. AsDB loans were slightly associated with poorer health scores, but no association was found when it came to education. The EBRD also had mixed results with less lending to countries with higher education scores, but also fewer loans to countries with poorer health. And, both the IDB and the AfDB loans decrease as a borrowing country’s score on health is lower, while no association is recorded between loans and education – placing these two RDBs even further toward the “Bank” end of the spectrum. In terms of political factors, strategic lending is manifested in hegemon(s)’ relative influence over RDB policies to favor allies in loan allocation. The overall statistical and qualitative findings paint a complex picture of hegemonic influence on the banks’ policies. Specifically, while each RDB has different hegemon(s), a different hegemonic structure, and a different relationship with hegemons and donors, their collective lending policies generally preference sound banking standards. This book’s findings substantiate the hegemonic role of the United States at the IDB. However, despite the US’s singular power, the IDB maintains an identity as a “Latin American Bank,” and, with few exceptions, shareholders delegate policy to the bank’s management and staff. At the AsDB, while statistical results point to Japan’s central role as a hegemon, interviews confirm that both Japan and the US, with equal voting shares, compete for influence – checking that one hegemon is not advancing policies that the other hegemon opposes. For AsDB management this means ensuring that both hegemons support program proposals. To maintain this hegemonic balance, the AsDB resorts to a bankcentered approach that typically averts tension between the hegemons. Similarly, while statistical results demonstrate Germany’s central role at the EBRD, further research reveals that the bank’s multiple hegemonic structure compels its managers to find policy equilibrium acceptable to all hegemons. Therefore, with two hegemons (the AsDB) and multiple hegemons (the EBRD), these development banks’ lending policy are, for the most part, led by strict institutional bank-centered guidelines that constitute a common denominator for hegemons, shareholders, and staff.

240

Conclusion: Future Outlook

The institutional culture, which derives from the circumstances that contributed to the RDBs’ foundation, also impacts RDB policies and reform options, and emerges as central to the way in which each of the banks conducts itself. Both the IDB and the AfDB were an initiative of borrowing member countries. However, their relationship with donor members differs greatly. While African countries resisted donor influence and hegemonic presence, the IDB accepted the role of the United States as a sole hegemon and has balanced US influence with the bank’s Latin American character. The AsDB and the EBRD, on the other hand, were initiated and designed by donors who continue to play a central role in policy making at both these institutions, holding a majority of the shares and votes on the Board of Governors. This book suggests that countries that get more loans are not the poorest in their regions; rather, they are the ones that attract more private investment and whose overall economy, health, and education sectors are stronger relative to their respective regions. This book also establishes that hegemons are essential for assuring funding and high credit ratings, and that the absence of hegemon(s) does not guarantee lending that is driven solely by development concerns. While aid may still be needed in all of the countries serviced by the RDBs, there is greater probability that RDB loans do not meet the “last resort” criteria in borrowing countries that attract more private finance and are faring better on social indicators. Thus, RDB lending is strategic, preferring countries that would be less likely to default. The similar findings for all four RDBs with regard to strategic lending, despite differences in hegemonic configurations, point to the flaws embedded in the design of development agencies that are structured like banks. My findings suggest, then, that institutional structures that prioritize sound banking and high credit ratings over development concerns influence the banks’ policies more than their hegemonic configurations. In general, this book demonstrates that development banks – designed to make loans, foster development, alleviate poverty, and remain financially solvent – are held to impossible standards by their founders (and, often, critics). Researchers and practitioners typically count these institutions as part of the global aid apparatus designed to assist developing countries; however, by making loans and contributing to the mounting debt experienced by many poor countries, these institutions are sometimes part of the problem rather than the solution. RDBs’ preference for sovereign loans to countries in the mid- to high range of development in the regions they lend to underscores the banks’ concern about the

Conclusion: Future Outlook

241

consequences of borrowers’ debt burden. More developed countries offer safer investments for RDBs, as they are less likely to default on loans or require a bailout to repay their debt. But this strategy does not provide optimal development aid and poverty alleviation to the regions these banks serve because it does not consistently address the most pressing development concerns of borrowers that have more difficulty in securing credit. By establishing IFIs (including RDBs) donor countries concede that market failure necessitates public agencies to help prevent further poverty and underdevelopment. Consequently, the moral hazard problem versus the reality of market failure are embodied in the controversy over IFIs as well as in their internal conflicting identities of development agencies and banks. On the one hand, market economies contribute to growing inequality, poverty, and underdevelopment – this is where they fail and where public agencies are arguably the only remedy. On the other hand, a reliance on public agencies can produce complacency and encourage suboptimal behavior on the part of borrowers. The RDBs are at a crossroads: should they continue to sponsor projects that now have higher prospects of being financed by other means, or should they shift their focus to the most socially and economically disadvantaged countries, where results might not materialize immediately, and where no other institution is likely to invest? This book’s findings provide a platform for further research that would help expose the RDBs’ shortcomings and delineate possible avenues for their reform. First, how can the banks and those who study them differentiate between projects that could otherwise be financed by the private sector and those that need public financing? Second, can we determine the relative importance of loans to borrowers and the actual impact of specific RDB-funded projects on development and poverty alleviation? Moreover, are RDBs better suited than their (larger) global counterparts and (smaller) subregional and domestic development agencies to administer development programs? Third, which organizations or corporations carry out the projects and programs funded by RDBs? Are these public financial institutions in effect facilitating private sector (mostly donor-country based) contracts in developing countries in a way that benefits private actors that carry out the projects at the expense of real development investment? Examining these questions would: (1) offer essential context when RDBs are considering reform; (2) substantially improve the appraisal of IFIs that administer development; and (3) illuminate the relationship between public organizations and private enterprise in the field of

242

Conclusion: Future Outlook

development – an area of research that has received little attention, but that bears greatly on all facets of globalization. In addition, other institutions that establish financial relations between countries and provide credit for the purpose of development warrant further investigation, especially in light of the expansion of development players in this field. Most notably, the New Development Bank (NDB), formerly known as the BRICS bank, was established by Brazil, Russia, India, China, and South Africa as an alternative to the Western-dominated IFIs. However, similarly designed as a credit providing development institution, it remains to be seen whether the NDB will successfully meet long-term development goals while maintaining a balanced financial portfolio – precisely where the RDBs have struggled.1 At a time when new development banks are formed while the old ones persist, this book lays out the challenges faced by the “development bank” structure. RDBs have been successful over the years in terms of maintaining ongoing shareholder support. But if we judge RDB performance based on why they were ostensibly created, they have not performed to their potential. The public monies that support RDBs are intended to be a last resort outlet and not a replacement for private financing. The two central obstacles to the RDBs’ ability to secure a last resort lender status are (1) the institutional structure that imposes banking criteria, driven by the need to meet rating agencies’ standards, and (2) a constantly shifting global financial and credit market that makes the RDBs’ added value a moving target, especially since the end of the Cold War. Despite their shortcomings, I believe there is an important role for RDBs in the development aid infrastructure. Even if development banks fail to live up to the goals articulated in their founding Articles of Agreement, their existence serves an important purpose in a volatile global economic climate, where financial crises rapidly spread. Developing countries are especially vulnerable during financial crises since they lack the domestic economic infrastructure to halt escalation and provide a safety net. Public credit agencies are, then, crucial for developing countries in times of fiscal turmoil (as the AsDB was during the 1997 Asian Financial Crisis). If RDBs “compete” with private financiers during stable economic climates, they become essential – lenders of last resort – during financial crises and recessions because that is when it is easiest for RDBs to operate

1

Similarly, the newly established Asian Infrastructure Investment Bank (AIIB), lead by China, may find itself subject to the same criticisms.

Conclusion: Future Outlook

243

free of their banking constraints and when competition with private financers is less likely. To reach their potential during more stable economic times, RDBs need to be reformed. The dichotomy of “bank” and “development agency” is reflected in the 2000 IFIAC report that recommended changing the RDBs’ name to “regional development agencies.” This task force identified the problem with the “bank” feature of the RDBs and its impact on the institutions’ ability to affect sustainable development. Changing the name alone will not change how these institutions operate. What would it mean to transform the development banks into development agencies? If the banking features of the RDBs inhibit them from fulfilling their development mission, then accomplishing development objectives would necessitate structural changes that include amending the centrality of the “bank” characteristic. For example, development and poverty alleviation – the purpose of the RDBs – are long-term processes. When institutions pursue these goals, there is no assurance that the strategy used will yield positive results (especially in the short run) and returns cannot be guaranteed. This presents a dilemma for institutions that aspire to be rated for their financial solvency and are therefore judged on short-term banking parameters. RDBs’ ability to prioritize development concerns over financial solvency is, then, rooted in politics: when powerful member states steer RDBs in that development direction, as a result of interests that drive member states to take financial risks because political rewards make it worthwhile, then the institutions can be navigated to meet, or at least genuinely pursue, their explicit long-term goals. A hegemon, then, has the potential to positively influence the RDBs while the absence of hegemon(s) jeopardizes RDB assets – compromising their ability to provide the necessary long-term financing to their borrowers. If RDBs were relieved from the obligation to maintain high credit ratings to secure financing (donor countries would commit resources with the expectation of higher risk), then it is possible that the IDB and the AfDB, given their institutional cultures that are very much “regional” with a strong sense that the banks belong to the borrowers, would be more likely to support borrowers’ need-based development proposals. To be sure, the AfDB is closer to a development agency than the other RDBs, but it struggles to ensure financial commitments and therefore it strategically lends. The IDB is closer to a bank, but with the support and backing of a single hegemon that often delegates, it could potentially capitalize on hegemonic political interests to push for better development policies.

244

Conclusion: Future Outlook

It might prioritize development over financial concerns as borrowers can more easily “negotiate” with one hegemon. Furthermore, a single hegemon has the authority to facilitate compromises in favor of development and incur the possible costs of such a development driven policy. In light of global changes, to provide an added value rather than compete with existing resources, RDBs should also create a mechanism that allows for flexibility in assessing proposals and put them to the “lender of last resort” test. Especially in a volatile global economy, the safety net of public development institutions is critical, but the role of RDBs as participants in a solution for credit shortage to developing countries has to take precedence, while ensuring that the institutions are not complicit in compounding borrowers’ debt burden. Furthermore, emphasizing the nontangible value of meeting ODA benchmarks, such as improving donor reputation and creating business opportunities for donors’ and hegemons’ private sectors, could alter the cost-benefit assessment of the financial commitments donors make to RDBs and the “returns” that they expect. This would facilitate a shift in priorities when making policy – credit that is driven by development needs rather than by financial parameters. To that extent, donors (and the RDBs) would have to accommodate the possibility of loan defaults or refinancing. They have to see themselves first and foremost as providers of a public good worth investing in for the long-term. This would require a shift in RDBs’ emphasis to nonfinancial returns. In Dickens’s novel, Pip’s great expectations were shattered when he lost everything and had to start over. During the heyday of British colonialism, when nobility was revered and industrialists became wealthy, rampant poverty stood in stark contrast to the riches of the lucky. The twentieth and twenty-first centuries global divide between rich and poor countries mirrors the complex moral quandaries that exposed the dark side of the industrial revolution. The noble goal of becoming a “developed” state is filled with Dickensian dilemmas about the sources of aid, the power of benefactors, and the responsibility of states that have succeeded to help the less fortunate. And while RDBs maneuver the maze of relationships between wealthy and poor countries, struggling to define their role in the shifting development landscape, we have downgraded our expectations from great to simply satisfactory.

appendix 1 Life Expectancy and HDI Rank

Life expectancy index, 1999 One of the three indices on which the human development index is built. HDI Rank 9 4 24 7 2 11 3 22 20 13 1 21 5 23 8 16 25 30 17 14 19

Country Japan Sweden Hong Kong, China (SAR) Iceland Australia Switzerland Canada Israel Italy France Norway Spain Belgium Greece Netherlands Austria Cyprus Malta Germany United Kingdom New Zealand

Life Expectancy Index, 1999 0.93 0.91 0.91 0.90 0.90 0.90 0.89 0.89 0.89 0.89 0.89 0.89 0.89 0.89 0.88 0.88 0.88 0.88 0.88 0.87 0.87 (continued)

245

246

Appendix 1: Life Expectancy and HDI Rank

(continued) HDI Rank 10 26 12 6 31 18 41 15 43 32 28 29 39 78 45 27 33 37 49 52 54 46 34 40 38 35 85 76 60 65 72 61 51 56 81 47 79

Country Finland Singapore Luxembourg United States Barbados Ireland Costa Rica Denmark Kuwait Brunei Darussalam Portugal Slovenia Chile Jamaica United Arab Emirates Korea, Rep. of Czech Republic Uruguay Trinidad and Tobago Panama Belize Croatia Argentina Bahrain Poland Slovakia Albania Georgia Macedonia, TFYR Lebanon Armenia Venezuela Mexico Malaysia Sri Lanka Lithuania Azerbaijan

Life Expectancy Index, 1999 0.87 0.87 0.87 0.86 0.86 0.86 0.85 0.85 0.85 0.85 0.84 0.84 0.84 0.84 0.83 0.83 0.83 0.82 0.82 0.81 0.81 0.81 0.80 0.80 0.80 0.80 0.80 0.80 0.80 0.80 0.80 0.79 0.79 0.79 0.78 0.78 0.77 (continued)

Appendix 1: Life Expectancy and HDI Rank

247

(continued) HDI Rank 68 63 36 97 62 57 71 64 44 59 87 88 50 66 80 89 58 84 95 82 91 48 100 42 70 96 67 99 53 90 73 106 74 101 69 92

Country Saudi Arabia Mauritius Hungary Syrian Arab Republic Colombia Bulgaria Oman Suriname Estonia Libyan Arab Jamahiriya China Jordan Latvia Thailand Paraguay Tunisia Romania Ecuador El Salvador Turkey Cape Verde Qatar Algeria Bahamas Philippines Samoa (Western) Fiji Uzbekistan Belarus Iran, Islamic Rep. of Peru Nicaragua Ukraine Vietnam Brazil Kyrgyzstan

Source: Human Development Report (2002).

Life Expectancy Index, 1999 0.77 0.77 0.77 0.76 0.76 0.76 0.76 0.76 0.76 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.74 0.74 0.74 0.74 0.74 0.74 0.73 0.73 0.73 0.73 0.73 0.73 0.72 0.72 0.72 0.71 0.71 0.71

appendix 2 RDB Shareholders

IDB members/shareholders (as of Dec. 31, 1999) Country Argentina Bahamas Barbados Belize Bolivia Brazil Chile Colombia Costa Rica Dominican Republic Ecuador El Salvador Guatemala Guyana Haiti Honduras Jamaica Mexico Nicaragua Panama Paraguay Peru Suriname

Voting Powers 900289 17533 10902 9313 72393 900289 247298 247298 36256 48355 48355 36256 48355 13528 36256 36256 48355 578767 36256 36256 36256 120580 7477

Percentage 10.760 0.210 0.130 0.111 0.865 10.760 2.956 2.956 0.433 0.578 0.578 0.433 0.578 0.162 0.433 0.433 0.578 6.917 0.433 0.433 0.433 1.441 0.089 (continued)

248

Appendix 2: RDB Shareholders

249

(continued) Country

Voting Powers

Percentage

Trinidad and Tobago Uruguay Venezuela Total

36256 96642 482402 4188179

0.433 1.155 5.766 50.057

NONREGIONALS Austria Belgium Canada Croatia Denmark Finland France Germany Israel Italy Japan Netherlands Norway Portugal Slovenia Spain Sweden Switzerland United Kingdom United States Total Unallocated Grand total

13447 27573 335022 4153 12644 13447 158773 158773 13261 158773 418777 28342 12644 4609 2569 158773 24240 39482 80686 2512664 4178652 2233 8366831

0.161 0.330 4.004 0.050 0.151 0.161 1.898 1.898 0.158 0.898 5.005 0.339 0.151 0.055 0.031 1.898 0.290 0.472 0.964 30.031 15.908 100

250

Appendix 2: RDB Shareholders AfDB members/shareholders (as of Dec. 31, 1999) As at 31 December, 1998

Borrowing Code

Country

ALG ANG BEN BOT BFO BUI CAO CAP CEN CHA COM CON ZAI IVO DJI EGY EQG ERI ETH GAB BAM GHA GUI GNB KEN LES LBR LIB MAG MAW MLI MAA MAS MOR MZM

Algeria Angola Benin Botswana Burkina Faso Burundi Cameroon Cape Verde Central Afr. Rep. Chad Comoros Congo Congo, Dem. Rep. Ivory Coast Djibouti Egypt Equatorial Guinea Eritrea Ethiopia Gabon Gambia Ghana Guinea Guinea Bissau Kenya Lesotho Liberia Libya Madagascar Malawi Mali Mauritania Mauritius Morocco Mozambique

Voting Powers 63844 20314 4119 38773 8669 5797 14855 2240 1675 2438 1025 9329 23365 80988 1895 92665 1039 3487 29249 25234 3017 37273 8833 1225 26445 3361 5282 58430 12281 7097 8526 5519 12225 60105 11955

Percentage 4.007 1.275 0.258 2.433 0.544 0.364 0.932 0.141 0.105 0.153 0.064 0.585 1.466 5.082 0.119 5.815 0.065 0.219 1.836 1.584 0.189 2.339 0.554 0.077 1.660 0.211 0.331 3.667 0.771 0.445 0.535 0.346 0.767 3.772 0.750 (continued)

Appendix 2: RDB Shareholders

251

(continued) As at 31 December, 1998 Borrowing Code

Country

Voting Powers

NAM NIR NIG RWA STP SEN SEY SIE SOM SAF SUD SWA TAZ TOG TUN UGA ZAM ZIM TOT

Namibia Niger Nigeria Rwanda Sao Tome & Prin. Senegal Seychelles Sierra Leone Somalia South Africa Sudan Swaziland Tanzania Togo Tunisia Uganda Zambia Zimbabwe Total

6713 6151 154968 2611 2227 18423 1809 5362 2769 16228 11300 6593 14033 4077 25721 10061 23864 42289 1047773

0.421 0.386 9.725 0.164 0.140 1.156 0.114 0.336 0.174 1.018 0.709 0.414 0.881 0.256 1.614 0.631 1.498 2.654 65.754

Nonborrowing ARG AUS BEL BRA CAN CHN DEN FIN FRN GER IND ITA JPN ROK KUW

Argentina Austria Belgium Brazil Canada China Denmark Finland France Germany India Italy Japan Korea Kuwait

4617 6617 9241 6617 51025 15625 16161 7185 51025 55953 3625 33121 74329 6617 6617

0.290 0.415 0.580 0.415 3.202 0.981 1.014 0.451 3.202 3.511 0.227 2.079 4.665 0.415 0.415

Percentage

(continued)

Appendix 2: RDB Shareholders

252 (continued)

As at 31 December, 1998 Borrowing Code

Country

NTH NOR POR SAU SPN SWD SWZ UK US

Voting Powers

Netherlands Norway Portugal Saudi Arabia Spain Sweden Switzerland United Kingdom United States Total

Percentage

10857 16161 3649 3225 8497 21361 20305 23189 90077

0.681 1.014 0.229 0.202 0.533 1.341 1.274 1.455 5.653 100

AsDB members/shareholders (as of Dec. 31, 1997) Country REGIONAL Afghanistan Australia Azerbaijan Bangladesh Bhutan Cambodia China Cook Islands Fiji Hong Kong, China India Indonesia Japan Kazakhstan Kiribati

Year of Membership

1966 1966 1999 1973 1982 1966 1986 1976 1970 1969 1966 1966 1966 1994 1974

Subscribed Capita (% of total)

0.034 5.871 0.451 1.036 0.006 0.05 6.539 0.003 0.069 0.553 6.424 5.526 15.836 0.818 0.004

Voting Power (% of total)

0.355 5.025 0.689 1.157 0.333 0.368 5.559 0.33 0.383 0.77 5.467 4.749 12.997 0.983 0.331 (continued)

Appendix 2: RDB Shareholders

253

(continued) Country Korea, Republic of Kyrgyz, Republic Lao, People’s Democratic Republic Malaysia Maldives Marshall Islands Micronesia Mongolia Myanmar Nauru Nepal New Zealand Pakistan Papua New Guinea Philippines Samoa Singapore Solomon Islands Sri Lanka Taipei, China (Taiwan) Tajikistan Thailand Timor-Leste Tonga Turkmenistan Tuvalu Uzbekistan Vanuatu Vietnam Subtotal Regional NONREGIONAL Austria Belgium Canada Denmark Finland

Year of Membership

Subscribed Capita (% of total)

Voting Power (% of total)

1966 1994 1966

5.112 0.303 0.014

4.417 0.571 0.339

1966 1978 1990 1990 1991 1973 1991 1966 1966 1966 1971 1966 1966 1966 1973 1966 1966 1998 1966 2002 1972 2000 1993 1995 1981 1966

2.763 0.004 0.003 0.004 0.015 0.553 0.004 0.149 1.558 2.21 0.095 2.418 0.003 0.345 0.007 0.588 1.105 0.291 1.382 0.01 0.004 0.257 0.001 0.684 0.007 0.346 63.458

2.538 0.331 0.33 0.331 0.34 0.77 0.331 0.447 1.575 2.096 0.404 2.262 0.331 0.604 0.333 0.799 1.212 0.56 1.433 0.336 0.331 0.533 0.329 0.875 0.333 0.605 65.192

1966 1966 1966 1966 1966

0.345 0.345 5.308 0.345 0.345

0.604 0.604 4.574 0.604 0.604 (continued)

Appendix 2: RDB Shareholders

254 (continued)

Year of Membership

Country France Germany Italy The Netherlands Norway Portugal Spain Sweden Switzerland Turkey United Kingdom United States Subtotal Nonregional Total

1970 1966 1966 1966 1966 2002 1986 1966 1967 1991 1966 1966

Subscribed Capita (% of total)

Voting Power (% of total)

2.362 4.39 1.834 1.041 0.345 0.345 0.345 0.345 0.592 0.345 2.072 15.836 36.542 100

2.217 3.84 1.795 1.161 0.604 0.604 0.604 0.604 0.802 0.604 1.986 12.997 34.808 100

EBRD members/shareholders (as of Dec. 31, 2002) Member Albania Armenia Australia Austria Azerbaijan Belarus Belgium Bosnia and Herzegovina Bulgaria Canada Croatia Cyprus Czech Republic Denmark Egypt

Date of membership 18 Dec 1991 7 Dec 1992 29 May 1990 29 May 1990 25 Sep 1992 10 Jun 1992 29 May 1990 17 Jun 1996 29 May 1990 29 May 1990 15 Apr 1993 29 May 1990 1 Jan 1993 29 May 1990 29 May 1990

Capital subscription, EUR million 20.00 10.00 200.00 456.00 20.00 40.00 456.00 33.80 158.00 680.00 72.92 20.00 170.66 240.00 20.00 (continued)

Appendix 2: RDB Shareholders

255

(continued) Member Estonia Finland France Georgia Germany Greece Hungary Iceland Ireland Israel Italy Japan Kazakhstan Korea, Republic of Kyrgyz Republic Latvia Liechtenstein Lithuania Luxembourg Former Yugoslav Republic of Macedonia Malta Mexico Moldova Mongolia Morocco Netherlands New Zealand Norway Poland Portugal Romania Russia Serbia and Montenegro (formerly Federal Republic of Yugoslavia) Slovak Republic Slovenia

Date of membership

Capital subscription, EUR million

28 Feb 1992 29 May 1990 29 May 1990 4 Sep 1992 29 May 1990 29 May 1990 29 May 1990 29 May 1990 29 May 1990 29 May 1990 29 May 1990 29 May 1990 27 Jul 1992 29 May 1990 5 Jun 1992 18 Mar 1992 29 May 1990 5 Mar 1992 29 May 1990 21 Apr 1993

20.00 250.00 1,703.50 20.00 1,703.50 130.00 158.00 20.00 60.00 130.00 1,703.50 1,703.50 46.00 200.00 20.00 20.00 4.00 20.00 40.00 13.82

29 May 1990 29 May 1990 5 May 1992 9 Oct 2000 29 May 1990 29 May 1990 29 May 1990 29 May 1990 29 May 1990 29 May 1990 29 May 1990 9 Apr 1992 19 Jan 2001

2.00 30.00 20.00 2.00 10.00 496.00 10.00 250.00 256.00 84.00 96.00 800.00 93.50

1 Jan 1993 23 Dec 1992

85.34 41.96 (continued)

256

Appendix 2: RDB Shareholders

(continued) Member Spain Sweden Switzerland Tajikistan Turkey Turkmenistan Ukraine United Kingdom United States of America Uzbekistan Federal Republic of Yugoslavia: see Serbia and Montenegro European Community European Investment Bank

Date of membership

Capital subscription, EUR million

29 May 1990 29 May 1990 29 May 1990 16 Oct 1992 29 May 1990 1 Jun 1992 13 Apr 1992 29 May 1990 29 May 1990 30 Apr 1992

680.00 456.00 456.00 20.00 230.00 2.00 160.00 1,703.50 2,000.00 42.00

29 May 1990 29 May 1990

600.00 600.00

appendix 3 AsDB Professional Staff

AsDB Professional Staff: 1981–1991 1981

Regional DMCsa India Philippines Korea Malaysia Other DMCs Australia and New Zealand Japan Nonregional US Canada Others not needed Total DMCs Others a

1986

1991

A

B

A

B

A

B

300 215 41 31 23 20 100 46

42 32 9 2 4 3 14 3

376 260 48 40 30 24 118 53

51 34 10 3 5 2 14 7

395 279 48 41 26 24 140 55

57 39 13 3 2 4 17 9

39 155 48 29 78 455 215 240

7 21 8 3 10 63 32 31

63 219 64 38 117 595 260 335

10 25 8 4 13 76 34 42

61 203 68 34 101 598 279 319

9 25 11 5 9 82 39 43

Developing Member Countries. Notes: Column A: All professional staff. Column B: Manager and above. Source: From The Asian Development Bank by Nihal Kappagoda. Copyright © 1995 by Lynne Rienner Publishers, Inc. Used with permission of the publisher.

257

Bibliography

Abbott, Kenneth W. and Duncan Snidal. 1998. Why States Act through Formal International Organizations. The Journal of Conflict Resolution 42(1): 3–32. African Development Bank. 1999. Compendium of Statistics on Bank Group Operations. Ivory Coast: African Development Bank. 1998. Annual Report 1998. Ivory Coast: African Development Bank. 1999. ADB Bulletin. Ivory Coast: African Development Bank (Vol. 2, No. 3). 1999. African Development Report: Infrastructure Development in Africa. Ivory Coast: African Development Bank 1999. In Brief. Ivory Coast: African Development Bank. 1999. The Vision of the African Development Bank. Ivory Coast: African Development Bank. 2000. African Development Report. Ivory Coast: African Development Bank. Alesina, Alberto and David Dollar. 2000. Who Gives Foreign Aid to Whom and Why? Journal of Economic Growth 5: 33–63. Anton Strezhnev and Erik Voeten, 2012–08, “United Nations General Assembly Voting Data,” http://hdl.handle.net/1902.1/12379 UNF:5: fWzDBiI+iY41v5zJF3JnoQ==Erik Voeten [Distributor] V4 [Version] Aoki, Masahiko and Hyunk-Ki Kim eds. 1995. Corporate Governance in Transitional Economies. Washington, DC: World Bank. Arena, Marco, Carmen Reinhart, and Francisco Vázquez. 2006. The Lending Channel in Emerging Economies: Are Foreign Banks Different? NBER Working Paper No. 12340. Asian Development Bank. Annual Report (various years, 1970–2011). Manila: Asian Development Bank. Babb, Sarah. 2009. Behind the Development Banks: Washington Politics, World Poverty, and the Wealth of Nations. Chicago, IL: University of Chicago Press. Baldwin, David A. 1965. The International Bank in Political Perspective. World Politics 18(1): 68–81. 1969. Foreign Aid, Intervention, and Influence. World Politics 21(3): 425–447. ed. 1993. Neorealism and Neoliberalism: The Contemporary Debate. New York: Columbia University Press.

259

260

Bibliography

Barnett, Michael L. and Martha Finnemore. 1999. The Politics, Power, and Pathologies of International Organization. International Organization 53(4): 699–732. Bartels, L. M. 1991. Instrumental and “Quasi-Instrumental” Variables. American Journal of Political Science 35: 777–800. Beck, Nathaniel L. and Jonathan N. Katz. 1995. What to Do (and Not to Do) with Time-Series—Cross-Section Data. American Political Science Review 89(3): 634–947. Beim, David O. and Charles W. Calomiris 2001. Emerging Financial Markets. Boston: McGraw-Hill. Ben-Artzi, Ruth. 2005. Great Expectations: The Politics of RDBs. PhD Dissertation, Columbia University, New York, NY. 2016. Peer Pressure as Soft Power: The Case of the Development Assistance Committee. Working Paper. Berthelemy, Jean-Claude ed. 1995. Wither African Economies? Paris: OECD. Bird, Graham, and Dane Rowlands. 2003. Political Economy Influences within the Life- Cycle of IMF programmes. World Economy 26(9): 1255–1278. Bordo, Michael D. and Barry Eichengreen. 1993. A Retrospective on the Bretton Woods System: Lessons from International Monetary Reform. Chicago: University of Chicago Press. Brambor, Thomas, William Roberts Clark, and Matt Golder. 2005. Understanding Interaction Models: Improving Empirical Analyses. Political Analysis 13: 1–20. Buell, Frederick. 1994. National Culture and the New Global System. Baltimore: Johns Hopkins University Press. Bueno de Mesquita, Bruce and Alastair Smith. 2007. Foreign Aid and Policy Concessions. Journal of Conflict Resolution 51(2): 251. Buite, Willem and Steven Fries. 2002. What Should the Multilateral Development Banks Do? EBRD Working Paper No. 74. London: EBRD. Calomiris, Charles. 1998. The IMF’s Imprudent Role as a Lender of Last Resort. The Cato Journal 17(3): 275–294. 2000. When Will Economics Guide IMF and World Bank Reforms? The Cato Journal 20(1): 85–103. 2002. A Globalist Manifesto for Public Policy. London: The Institute of Economic Affairs. 2003. Lesson from Argentina and Brazil. The Cato Journal 23(1): 33–45. Cameron, Colin A. and Pravin K. Trivedi. 2005. Microeconometrics: Methods and Applications. New York: Cambridge University Press. Cardoso, Fernando Henrique, and Enzo Faletto. 1979. Dependency and Development in Latin America. Berkeley: University of California Press. Chandavarkar, Anand G. 1996. Central Banking in Developing Countries. New York: St. Martin’s Press. Chwieroth, Jeffrey M. 2008. Organizational Change “From Within”: Exploring the World Bank’s Early Lending Policies. Review of International Political Economy 15(4): 481–505. Copelovitch, Mark S. 2010. The International Monetary Fund in the Global Economy. New York: Cambridge University Press.

Bibliography

261

Cox Gary W. and Mathew D. McCubbins. 1986. Electoral Politics as a Redistributive Game. The Journal of Politics 48: 370–389. Cukierman, Alex. 1994. Central Bank Independence and Monetary Control. The Economic Journal 104: 1437–1448. Cukierman, Alex, and Steven B. Webb. 1995. Political Influence on the Central Bank. The World Bank Economic Review 9(3): 397–423. Culpeper, Roy. 1997. Titans or Behemoths? Boulder, CO: Lynne Rienner Publishers, Inc. David, Wilfred L. 1985. The IMF Policy Paradigm: The Macroeconomics of Stabilization, Structural Adjustment, and Economic Development. New York: Praeger Publishers. Dell, Sidney. 1972. The Inter-American Development Bank: A Study in Development Financing. New York: Praeger Publishers. DiMaggio, Paul. 1997. Culture and Cognition. Annual Review of Sociology 23: 263–287. Dixit, A., and John Londregan. 1996. The Determinants of Success of Special Interests in Redistributive Politics. Journal of Politics 58: 1132–1155. Downs, George W., David M. Rocke, and Peter N. Barsoom. 1996. Is the Good News about Compliance Good News about Cooperation? International Organization 50:3. Dreher, Axel, and Roland Vaubel. 2004a. Do IMF and IBRD Cause Moral Hazard and Political Business Cycles? Evidence from Panel Data. Open Economies Review 15(1): 5–22. 2004b. The Causes and Consequences of IMF Conditionality. Emerging Markets Finance and Trade 40(3): 26–54. Dunning, Thad. 2004. Conditioning the Effects of Aid: Cold War Politics, Donor Credibility, and Democracy in Africa. International Organization 58(3): 409–423. Easterly, William. 2001. The Elusive Quest for Growth. Cambridge, MA: MIT Press. 2006. The White Man’s Burden. New York: The Penguin Press. EBRD. 1991. A Changing Europe – Annual Report. London: EBRD. 1999. The EBRD: Its Role and Activities. London: EBRD (doc. 4042). 1999. Financing with the EBRD. London: EBRD (doc. 4067). Annual Report 1991–2011 Eccles and Gwin. 1999. Supporting Effective Aid: A Framework for Future Concessional Funding of Multilateral Development Banks. Washington, DC: Overseas Development Council. Economist, January 31, 2004. Bridge to Nowhere: A Bridge in the Middle of Nowhere Has Become a Symbol of Bad Development. Edwards, Martin S. 2006. “Signaling Credibility? The IMF and Catalytic Finance.” Journal of International Relations and Development 9(1): 27–52. Eijffinger, Sylvester C. W. 1996. The Political Economy of Central Bank Independence. Princeton, NJ: Princeton University Press. English, Philip E. and Harris M. Mule. 1995. The African Development Bank. Boulder, CO: Lynne Rienner Publishers, Inc.

262

Bibliography

Evans, Peter B. 1995. Embedded Autonomy: States and Industrial Transformation. Princeton, NJ: Princeton University Press. Fearon, James D. 1998. Bargaining, Enforcement, and International Cooperation. International Organization 52:2. Feinberg, Richard E. 1988. The Changing Relationship between the World Bank and the International Monetary Fund. International Organization 42(3): 545–560. Financial Times, September 24, 2012. The Future of the MDBs. Finnemore, Martha. 1996. National Interests in International Society. Ithaca, NY: Cornell University Press. Finnemore, Martha and Katheryn Sikkink. 1998. International Norm Dynamics and Political Change. International Organization 52(4): 887–918. Frieden, Jeffry A. and David A. Lake. 1995. International Political Economy: Perspectives on Global Power and Change. New York: St. Martin’s Press. 1991. Debt, Development, and Democracy: Modern Political Economy and Latin America, 1965–1985. Princeton, NJ: Princeton University Press. Friedman, Milton. 1982. Capitalism and Freedom. Chicago: University of Chicago Press. Fry, Maxwell J. 1996. Central Banking in Developing Countries: Objectives, Activities and Independence. London: Routledge. Gartzke, Eric. http://dss.ucsd.edu/~egartzke/htmlpages/data.html Gilpin, Robert. 1981. War and Change in International Politics. New York: Cambridge University Press. 1987. The Political Economy of International Relations. Princeton, NJ: Princeton University Press. Goldberg, Linda B., Gerard Dages, and Daniel Kinney. 2000. Foreign and Domestic Bank Participants in Emerging Markets: Lessons from Mexico and Argentina. NBER Working Paper No. 7714. Gould, Erica. 2006. Money Talks. Stanford: Stanford University Press. Gowa, Joanne. 1984. Hegemons, IOs, and Markets: The Case of the Substitution Account. International Organization 38(4): 661–683. Grieco, Joseph. 1990. Cooperation Among Nations. Ithaca: Cornell University Press. Griffin, Kieth. 1989. Alternative Strategies for Economic Development. London: The Macmillan Press Ltd. Gwin, Catherine and Joan Nelson eds. 1997. Perspectives on Aid and Development. Policy Essay No. 22. Washington, DC: Overseas Development Council. Haggard, Stephan et al. 1993. The Politics of Finance in Developing Countries. Ithaca: Cornell University Press. Haggard, Stephan and Robert R. Kaufman. 1992. The Politics of Economic Adjustment: International Constraints, Distributive Conflicts, and the State. Princeton, NJ: Princeton University Press. Hawkins, Darren et al. eds. 2006. Delegation and Agency in International Organizations. UK: Cambridge University Press. Head, John W. 2004. For Richer or Poorer: Assessing the Criticisms Directed at the Multilateral Development Banks. Kansas Law Review 52: 241.

Bibliography

263

Helleiner, G. K. ed. 1996. The International Monetary and Financial System: Developing-Country Perspectives. London: Macmillan Press Ltd. Hirschman, Albert O. 1985 (1st edition 1971). A Bias for Hope: Essays on Development and Latin America. London: Westview Press. IBRD. 1954. Annual Report. Inter-American Development Bank. Annual Report (various years, 1963–2011) Washington, DC: Inter-American Development Bank. Report of the European Office of the IaDB, 2012, 2013. Jeanneau, Serge and Marian Micu. 2002. Determinants of International Bank Lending to Emerging Market Countries. BIS Working Paper No. 112. Kam, C. D. and Robert J. Franzese, Jr. 2007. Modeling and interpreting interactive hypotheses in regression analysis. Ann Arbor: University of Michigan Press. Kanbur, Ravi and Todd Sandler. 1999. The Future of Development Assistance: Common Pools and International Public Goods. Policy Essay No. 25. Washington, DC: Overseas Development Council. Kappagoda, Nihal. 1995. The Asian Development Bank. Boulder, CO: Lynne Rienner Publishers, Inc. Keele, Luke and Nathan J. Kelly. 2005. Dynamic Models for Dynamic Theories: The Ins and Outs of Lagged Dependent Variables. Political Analysis. Keohane, Robert O. 1984. After Hegemony. Princeton, NJ: Princeton University Press. Keohane, Robert O. and Joseph S. Nye. 1977. Power and Interdependence: World Politics in Transition. Boston, MA: Little, Brown. Keohane, Robert O. and Lisa L. Martin. 1995. The Promise of Institutional Theory. International Security 20(1): 39–51. Kilby, Christopher. 2006. Donor Influence in Multilateral Development Banks: The Case of the Asian Development Bank. Review of International Organizations 1(2):173–195 Kindleberger, C. P. 1981. Dominance and Leadership in the International Economy: Exploitation, Public Goods, and Free Rides. International Studies Quarterly 25(2): 242–254. Koremenos, Barbara, Charles Lipson, and Duncan Snidal. 2001. The Rational Design of International Institutions. International Organization 55(4): 761–799. Krasner, Stephen D. ed. 1983. International Regimes. Ithaca, NY: Cornell University Press. 1981. Power Structures and Regional Development Banks. International Organization 35(2): 303–328. Krueger, Ann O. 1993. Political Economy of Policy Reform in Developing Countries. Cambridge, MA: MIT Press. 1998. Wither the World Bank and the IMF? Journal of Economic Literature 36(4): 1983–2020. 1999. The World Bank Group in the International Economy. Draft (Treasury).

264

Bibliography

Krugman, Paul. 1993. Regionalism versus Multilateralism: Analytical Notes. In New Dimensions in Regional Integration, edited by Jaime de Melo and Arvind Panagariya, 58–79. New York: Cambridge University Press. 1994. The Myth of the Asian Miracle. Foreign Affairs. November/December 1994. Lawrence, Robert Z. 1996. Regionalism, Multilateralism, and Deeper Integration. Washington, DC: Brookings Institution. Lerrick, Adam. The World Bank’s Wrong Portfolio. The International Economy. January/February 2000. Lim, Daniel Yew Mao and James Raymond Vreeland. 2013. Regional Organizations and International Politics: Japanese Influence over the Asian Development Bank and the UN Security Council. World Politics 65:34–72. Mansfield, Edward D. and Helen V. Milner. 1997. The Political Economy of Regionalism. New York: Cambridge University Press. Mansfield, Edward D. and Jon C. Pevehouse. 2006. Democratization and International Organizations. International Organization 60(1): 137–167. March, James and Johan P. Olsen. 1989. Rediscovering Institutions: The Organizational Basis of Politics. New York: Free Press. Martin, Lisa L. 1992. Coercive Cooperation: Explaining Multilateral Economic Sanctions. Princeton, NJ: Princeton University Press. Martin, Lisa and Beth Simmons. 1998. Theories and Empirical Studies of International Institutions. International Organization 52(4): 729–757. Maxfield, Sylvia. 1997. Gatekeepers of Growth: The International Political Economy of Central Banking in Developing Countries. Princeton, NJ: Princeton University Press. McKinlay, R. D. and R. Little. 1977. A Foreign Policy Model of US Bilateral Aid Allocation. World Politics 30(1): 58–86. Mearsheimer, John J. 1995. The False Promise of International Institutions. International Security 19(3): 5–49. Meier, Gerald M. 1995. Leading Issues in Economic Development. New York: Oxford University Press. Meyer, John W. and Brian Rowan. 1977. Institutionalized Organizations: Formal Structure as Myth and Ceremony. American Journal of Sociology 83(2): 340–363. Mill, J. S. 1994. Principles of Political Economy. Oxford: Oxford University Press. Milner, Helen V. 1997. Interests, Institutions, and Information. Princeton, NJ: Princeton University Press. 2006. Why Multilateralism? Foreign Aid and Domestic Principal-Agent Problems. In Delegation and Agency in International Organizations, edited by Darren Hawkins et al., 107–139 New York: Cambridge University Press.. Milner, Helen V. and Dustin Tingley. 2010. The Political Economy of US Foreign Aid: American Legislators and the Domestic Politics of Aid. Economics & Politics 22(2): 200–232. Mingst, Karen A. 1990. Politics and the African Development Bank. Lexington, KY: The University Press of Kentucky. Mosley, Layna. 2003. Global Capital and National Governments. Cambridge: Cambridge University Press.

Bibliography

265

Mosley, Paul. 1988. On Persuading a Leopard to Change His Spots: Optimal Strategies for Donors and Recipients of Conditional Development Aid. In Toward a Political Economy of Development, edited by Robert H. Bates, 47–79. Berkeley, CA: University of California Press. Mosley, Paul, Jane Harrigan, and John Toye. 1991. Aid and Power: The World Bank and Policy-Based Lending. London: Routledge. Moyo, Dambisa. 2009. Dead Aid. New York: Farrar, Straus, and Giroux. Nelson, Rebecca M. 2012. Multilateral Development Banks: Overview and Issues for Congress. CRS Report for Congress 7–5700/R41170. Neumayer, Eric. 2003a. The Determinants of Aid Allocation by Regional Multilateral Development Banks and United Nations Agencies. International Studies Quarterly 47: 101–122. 2003b. The Pattern of Aid Giving: The Impact of Good Governance on Development Assistance. New York: Routledge. Nielson, D. L. and M. J. Tierney. 2003. Delegation to International Organizations: Agency Theory and World Bank Environment Reform. International Organization 57(2): 241–276. 2005. Theory, Data, and Hypothesis Testing: World Bank Environmental Reform Redux. International Organization 59(3): 785–800. OECD Development Center, Policy Brief No. 17, 1999. O’Halloran, Sharyn and David Epstein. 1999. Delegating Powers: A Transaction Cost Politics Approach to Policy Making under Separate Powers. New York: Cambridge University Press. Oman, Charles. 1996. The Policy Challenges of Globalisation and Regionalisation. Dev. Centre Policy Brief No. 11. Paris: OECD. Oye, Kenneth ed. 1986. Cooperation under Anarchy. Princeton, NJ: Princeton University Press. Pollack, Mark A. 1997. Delegation, Agency, and Agenda Setting in the EC. International Organization 51(1): 99–134. Posen, Adam. 1994. Central Bank Independence and Disinflationary Credibility: A Missing Link. Washington, DC: Brookings Institution. Przeworski, Adam. 1991. Democracy and the Market: Political and Economic Reforms in Eastern Europe and Latin America. Cambridge University Press. 1995. Sustainable Democracy. New York: Cambridge University Press. Radelet, Steve and Jeffrey Sachs. 1997. Asia’s Reemergence. Foreign Affairs. November. Riker, William. 1980. Implications from the Disequilibrium of Majority Rule for the Study of Institutions. American Political Science Review 74: 432–447. Rodrik, Dani. 1999. The New Global Economy and Developing Countries: Making Openness Work. Washington, DC: Overseas Development Council. 2007. One Economics Many Recipes. Princeton, NJ: Princeton University Press. Rudra, Nita. 2008. Globalization and the Race to the Bottom in Developing Countries: Who Really Gets Hurt? New York: Cambridge University Press. Rueschemeyer, Stephens and Evelyne Huber Stephens. 1992. Capitalist Development and Democracy. Chicago: University of Chicago Press.

266

Bibliography

Ruggie, John G. ed. 1993. Multilateralism Matters. New York: Columbia University Press. 1998. What Makes the World Hang Together. International Organization 52 (3): 855–886. Sachs, Jeffrey D. 2005. The End of Poverty: Economic Possibilities for Our Time. New York: Penguin. Sachs, Jeffrey D., Kwesi Botchwey, Maciej Cuchra, and Sara Sievers. 1999. Implementing Debt Relief for the HIPCs. Columbia University, Earth Institute Report. Sanford, Jonathan E. 2005. International Financial Institutions: Funding US Participation. CRS Report for Congress. RS22134. Sanford, Jonathan E. and Martin A. Weiss. 2004. The International Monetary Fund: Organization, Function, and Role in the International Economy. CRS Report for Congress. RL32364. Satyanath, Shanker. 2006. Globalization, Politics, and Financial Turmoil. New York: Cambridge University Press. Schelling, Thomas C. 1966. Arms and Influence. New Haven: Yale University Press. Sen, Amartya ed. 1984. Resources, Values, and Development. Oxford: Blackwell Sherk, Donald. 1999. The Asian Development Bank. Unpublished. Simmons, B. A. and D. J. Hopkins. (2005). The Constraining Power of International Treaties: Theory and Methods. American Political Science Review 99(4): 623–631. Smith, Adam. 2005. The Wealth of Nations. London: Penguin Books. Snidal, Duncan. 1985. The Limits of Hegemonic Stability Theory. International Organization 39(4): 579–614. Sovey, Allison J and Donald P. Green. 2011. Instrumental Variables Estimation in Political Science: A Readers’ Guide. American Journal of Political Science 55 (1): 188–200. Standard & Poor’s. 1999. Supranationals. New York: Standard & Poor’s. 2002. Sovereigns. New York: Standard & Poor’s. 2003. Supranationals. New York: Standard & Poor’s. 2009. Supranationals. New York: Standard & Poor’s. Stein, A. A. (1990). Why Nations Cooperate: Circumstance and Choice in International Relations. Ithaca: Cornell University Press. Stiglitz, Joseph E. 2002. Globalization and Its Discontents. New York: W. W. Norton. Stone, Randall W. 2002. Lending Credibility: The International Monetary Fund and the Post Communist Transition. Princeton, NJ: Princeton University Press. 2011. Controlling Institutions: International Organizations and the Global Economy. New York: Cambridge University Press. Strand, Jonathan R. 2003a. Measuring Voting Power in an International Institution: The United States and the Inter-American Development Bank. Economics of Governance 4(1): 19–36. 2003b. Power relations in an embedded institution: the European Bank for Reconstruction and Development. European Integration 25(2): 115–129.

Bibliography

267

2010. Regional Development Banks: Lending with a Regional Flavor. Strandow, Daniel, Michael Findley, Daniel Nielson, and Joshua Powell. 2011. The UCDP-AidData codebook on Geo-Referencing Foreign Aid. Version 1.1. Uppsala Conflict Data Program. Uppsala, Sweden: Uppsala University. Summers, Lawrence H. and Vinod Thomas. 1993. Recent Lessons of Development. The World Bank Research Observer 8(2): 241–254. Sunkel, O. ed. 1993. Development from Within: Toward a Neostructuralist Approach for Latin America. Boulder, CO: Lynne Rienner Publishers, Inc. Thacker, Strom C. 1999. The High Politics of IMF Lending. World Politics 52(1): 38–75. Thant, Myo, Min Tang, and Hiroshi Kakazu eds. 1994. Growth Triangles in Asia: A New Approach to Regional Economic Cooperation. Oxford: Oxford University Press for Asian Development Bank. Tierney, Michael J., Daniel L. Nielson, Darren G. Hawkins, J. Timmons Roberts, Michael G. Findley, Ryan M. Powers, Bradley Parks, Sven E. Wilson, and Robert L. Hicks. 2011. More Dollars than Sense: Refining Our Knowledge of Development Finance Using Aid Data. World Development 39(11): 1891–1906. Tussie, Diana. 1995. The Inter-American Development Bank. Boulder, CO: Lynne Rienner Publishers, Inc. United Nations Economic Commission for Africa. 1964. Agreement Establishing the African Development Bank. Upton, Barbara. 2000. The Multilateral Development Banks. Westport, CT: Praeger Publishers. Vaubel, Roland. 1986. A public Choice Approach to International Organization. Public Choice 51(1): 39–57. 1991. The Political Economy of the International Monetary Fund: A Public Choice Analysis. In The Political Economy of International Organizations, edited by Roland Vaubel and Thomas Willett, 204–244. Boulder, CO: Westview Press. 1996. Bureaucracy at the World Bank and the IMF: A Comparison of Evidence. The World Economy 19(2): 195–210. 2006. Principal-Agent Problems in International Organizations. The Review of International Organizations 1(2): 125–138. Vreeland, James Raymond. 2003. The IMF and Economic Development. New York: Cambridge University Press. 2007. The International Monetary Fund: Politics of Conditional Lending. New York: Routledge. Wade, Robert. 1996. Japan, the World Bank, and the Art of Paradigm Maintenance: The East Asian Miracle in Political Perspective. New Left Review 217: 3–36. 2002. US Hegemony and the World Bank: The Fight over People and Ideas. Review of International Political Economy 9(2): 201–229. Waltz, Kenneth. 1959. Man, the State, and War. New York: Columbia University Press. 1979. Theory of International Politics. Reading, MA: Addison-Wesley.

268

Bibliography

Weaver, C. 2008. Hypocrisy Trap: The World Bank and the Poverty of Reform. Princeton, NJ: Princeton University Press. Weber, Steven. 1994. Origins of the European Bank for Reconstruction and Development. International Organization 48(1): 1–38. Weiss, Martin A. 2005a. The World Bank: Changing Leadership and Issues for the United States Congress. CRS Report for Congress. RS22029. 2005b. The World Bank: International Development Association’s 14th Replenishment (2006–2008). CRS Report for Congress. RS22162. White, John. 1974. The Politics of Foreign Aid. New York.: St Martin’s Press. The White House, May 19, 2011. Office of the Press Secretary. “Remarks by the President on the Middle East and North Africa.” State Department: Washington, DC. Whitman, Marina V. N. 1999. New World, New Rules: The Changing Role of the American Corporation. Cambridge: Harvard Business School Press. Winters, M. S. (2010). Choosing to Target: What Types of Countries Get Different Types of World Bank Projects. World Politics 62(3): 422–458. World Bank, Articles of Agreement (1944). World Bank, World Development Indicators (2012). http://www.afdb.org (African Development Bank). http://www.asiandevbank.org (Asian Development Bank). http://www.ebrd.com (European Bank for Reconstruction and Development). http://www.iadb.org (Inter-American Development Bank).

Index

1997 financial crisis, 83 Abbott and Snidal, 53, 223 Adam Lerrick, xvii, 24, 71 additionality, 20, 131, 144 Affinity Measure, 116, 117, 182 Afghanistan, 60, 110, 113, 138, 197, 252 African Development Fund, 81, 115 agriculture sector AfDB, 102 Aid foreign policy view, 54 interest based, 53 need based, 53 AidData, 267 Alesina and Dollar, 43, 53 American foreign policy, 34, 65, 162, 196 American hegemony, 90, 200, 204 American influence, 2 Aoki and Kim, 43 Arena, Reinhart, and Vazquez, 152, 154 Argentina, ix, 8, 37, 45, 73, 113, 140, 146, 202, 246, 248, 251, 260, 262 AsDB dual-hegemonic structure, 168 Asian Financial Crisis, 8, 29, 101, 124, 125, 147, 233, 242 Asian Tigers, 138, 186 authorized loans, 122, 124 Axelrod, 42 Babb, 8, 54, 58, 64, 83, 124, 196, 202, 203, 209, 220, 259 Baker, 124, 202 Baker Plan, 202

Baldwin, xvi, 15, 47, 259 Bangladesh, 113, 140, 142, 143, 252 bank function RDBs, 8, 56 bank/development agency continuum, 44, 56, 75, 199 scale, 19, 36 spectrum, 160, 163 bankable, 51, 96, 100, 102, 105, 115, 123, 193, 225, 232 “bankable” projects, 100, 105, 115 Barnett and Finnemore, 24, 51 Bartels, 155, 260 Beck and Katz, 153 Beim and Calomoris, 46 Bird and Rowlands, 118 borrowing member countries voting shares, 71, 75, 83, 84, 86, 90, 92, 94, 97, 110, 132, 154, 166, 198, 200, 201, 209, 226, 227, 234, 237, 240 Brambor, Clark, and Golder, 157 Brazil, 29, 36, 38, 45, 73, 113, 202, 242, 247, 248, 251, 260 Brazilian National Development Bank, 146 Brian Atwood, 224 bridge the gap, 100 Bueno de Mesquita and Smith, 43 callable capital, 22, 47, 70, 71, 72, 85, 89, 197, 200, 213, 225 Cameron and Trivedi, 160

269

270

Index

capital flows to developing countries, 12, 33, 99, 114, 117 capital increase, 59, 71, 79, 200 Cardoso and Faletto, 28 Carlos Jarque, xv, 59, 76, 201, 224 China AsDB, 36, 37, 45, 60, 93, 113, 210, 242, 245, 247, 251, 252, 253 Cold War RDB creation, 64 Cold War and the IDB, 163 Collective Principal, 121 concessional, 19, 81, 82, 89, 98, 99, 100, 107, 114, 115, 154, 200, 201, 233 Copelovitch, 34, 121, 154, 155, 204, 260 Corporate governance, 42 Cox and McCubbins, 191 credit agencies, 77, 83, 200, 208, 242 credit crisis, 8 credit rating, 16, 22, 69, 71, 72, 77, 78, 84, 191, 199, 200, 214, 238 credit ratings, 18, 63, 69, 83, 95, 107, 126, 211, 213, 225, 227, 231, 237, 238, 240, 243 Cuba, 65, 88, 110, 163, 191, 197 IDB, 60 Cukierman, 197, 261 Cukierman and Webb, 197 Culpeper, 8, 12, 23, 65, 67, 68, 92, 99, 103, 107, 226, 261 Cummings, 138 DAC, xiii, 4, 18, 32, 116, 210, 223, 224 debt crisis, 29, 101, 104, 197, 215 debt relief, 30, 31, 78, 202 debt servicing, 30, 110, 127, 143 debt, accumulating, 125 Dell, 100, 122, 123, 124, 226, 261 democracy, 61, 116, 121, 129, 151, 153, 162, 168, 174, 181, 188, 229 democracy and loans during and after the Cold War, 189 democratization, 6, 68, 85, 86, 93, 111, 121, 215 development function RDBs, 8 disbursed loans, 2, 69, 122, 154 Dixit and Londregan, 191 Donald Kaberuka, 92, 96

Dreher and Vaubel, 154 Dunning, 43, 163, 261 Easterly, xvii, 8, 24, 27, 28, 29, 31, 58, 104, 105, 106, 111, 119, 120, 125, 138, 261 economic development, 6, 41, 48, 56, 67, 78, 83, 117, 133, 143, 161, 228, 231, 236 education FDI and the EBRD, 187 long-term, 120 Edwards, 118, 261 emerging markets, 26, 45, 57, 119, 135 endogeneity, 118, 154, 155 English and Mule, 66, 80, 102, 226 Enrique Iglesias, 96, 202, 205 Epstein and O’Halloran, 42 European Council, 68 European Investment Bank, 68, 256 Evans, 11, 262 export-led growth, 29 FDI Africa, 133 Asia, 135 Central and Eastern Europe, 135 importance for economic development, ix, x, 40, 56, 57, 61, 99, 102, 109, 110, 112, 116, 117, 118, 119, 122, 125, 132, 133, 136, 150, 155, 160, 161, 162, 166, 170, 176, 182, 211, 225, 238, 239 Latin America, 134 to Asia, 186 Fearon, 42, 262 Finnemore, 24, 51, 260, 262 first loan IDB, Peru, 230 François Mitterrand, 68, 84 Fund for Special Operations, 201 Gartzke, 57, 116, 262 GDP ODA target, 223 get the loans out, 105, 112 Gilpin, 42, 43, 44, 196, 262 global financial crisis, 201 global financial markets, 10, 57 globalization, 7, 35, 36, 37, 38, 39, 242 Goldberg, Dages, and Kinney, 152 Gould, 53, 65, 112, 196, 262 Gowa, 42, 43, 262

Index grants in place of loans, 127 Grieco, 42, 262 gross loan amounts versus net amounts, 122 growth and development, 27, 119, 120, 135 Haggard and Kaufman, 11 Haiti, ix, 36, 73, 140, 224, 248 Hawkins et al., 42, 48, 264 Head, 130, 262 health and education, 29, 30, 61, 117, 138, 161, 232, 239 hegemonic influence AsDB, 169 hegemonic involvement AfDB, 174 Hegemonic Stability Theory, 42, 43, 45, 61, 266 hegemons, 6, 43, 44, 45, 46, 50, 55, 57, 58, 61, 77, 78, 88, 110, 116, 117, 120, 152, 153, 168, 170, 181, 182, 189, 191, 192, 195, 198, 199, 206, 207, 208, 209, 210, 211, 216, 220, 221, 223, 224, 227, 233, 234, 238, 239, 240, 244 HIPCs, 30, 77, 90 Horst Kohler, 95 Human Development Index, xi, 119, 120 IDB lending and the Cold War, 191 IDB, effect of US Affinity on loan amounts at different levels of infant mortality rates, 188 IDB’s Europe Office, 59 IDB’s offices in Europe and in Tokyo, 75 IFIAC report, 5, 11, 13, 14, 19, 22, 24, 25, 30, 38, 77, 78, 107, 126, 146, 147, 223, 226, 243 IFIs goals, 20 reform, 5 IMF, 3, 4, 5, 10, 11, 14, 15, 21, 22, 23, 30, 33, 34, 39, 42, 54, 57, 58, 59, 65, 66, 95, 96, 104, 105, 112, 118, 121, 125, 155, 196, 215, 230, 260, 261, 263, 267 institution definition, 17 institutional culture, 4, 24, 199, 207, 208, 240

271

institutionalization of development, 9 Institutions outcomes, 18 interest rates, 80, 89, 90, 131 international cooperation, 42, 43 International Financial Institution Advisory Commission (IFIAC), 11 Jacques Attali, xvi, 93, 95 Jacques de Larosiere, 95 Japan, x, 4, 23, 45, 57, 59, 60, 67, 70, 77, 82, 83, 84, 85, 88, 90, 91, 92, 94, 97, 98, 101, 117, 120, 126, 127, 153, 166, 168, 169, 170, 191, 198, 206, 207, 208, 209, 210, 217, 233, 239, 245, 249, 251, 252, 255, 257, 267 Jaques Attali, 86 Jean Lemierre, 96 Jeanneau and Mice, 152 John Stuart Mill, 26 Kam and Franzese, 154 Kappagoda, 67, 89, 93, 101, 114, 209, 226, 231, 257, 263 Keele and Kelly, 153 Keohane, 42, 45, 263 Keohane and Nye, 197 Keynes, 22 Kilby, xvii, 166, 263 Kindleberger, 43, 44, 263 Krasner, 8, 43, 44, 83, 87, 196, 200, 203, 204, 207, 208, 210, 211, 213, 263 Krugman, 29, 138, 264 lender-of-last-resort, 112, 118 lenders of last resort, 8, 20, 56, 202, 242 Leo Harari, 100, 224, 229 loan amounts FDI and Infant Mortality, 184 loan disbursements versus loan commitments, 121 Mansfield and Pevehouse, 223 Marginal effects, 157 market economy, 68 market failure, 8, 9, 20, 26, 32, 241 Martin, 42, 52, 223, 260, 261, 262, 263, 264, 266, 268 maturities, 80, 89, 90, 99, 232, 237 Maxfield, 197, 264 McKinlay and Little, 53, 54

272

Index

Mexico, 3, 37, 45, 73, 76, 85, 96, 113, 202, 246, 248, 255, 262 microlending, 86 Millennium Development Goals, 223, 224 Milner and Tingley, 222 Milton Friedman, 8, 26 Mingst, 76, 78, 80, 234, 264 National Security Council, 205 Neumayer, 8, 117, 265 Newly Industrialized Economies, 101 NGOs, xiii, 4, 12, 18, 36, 205, 215 Nielson and Tierney, 42, 111 Nigeria, 45, 70, 78, 113, 212, 251 nonconcessional, 89 North Korea, 60, 110, 197 Obama, 86 OECD, xiii, xv, xvi, 3, 13, 18, 19, 32, 35, 40, 47, 68, 76, 116, 146, 198, 209, 210, 223, 224, 238, 260, 265 Official Development Assistance, 222 OLS models, 157 Ordinary Capital Resources IDB, 71, 89, 200 organizational culture, 42, 50, 51, 52, 58, 61 Oye, 42, 265 paid-in capital, 22, 47, 69, 70, 71, 72, 79, 115, 200, 213, 225, 237, 238 Participatory Governance, 47, 48 political goals, 6, 85, 93, 129, 191, 193 Political Proximity Hypothesis, 59, 60 Pollack, 42, 48, 49, 265 portfolio and equity investment, 118 poverty alleviation, 2, 3, 9, 11, 15, 18, 19, 28, 30, 31, 58, 71, 85, 101, 102, 105, 118, 125, 152, 161, 193, 234, 235, 237, 241, 243 poverty alleviation and development, 2, 9, 19, 58, 71, 102, 125, 152, 193 principal-agent relations, 4 Principal-Agent Theory, 42, 48 Privatization, 86 procurement, 47, 59, 212, 217, 224, 225, 238 Przeworski, 11, 265 Radelet and Sachs, 29, 138 ratio of donors to borrowers, 69, 88

RDBs governance, 46 largest shareholders, 44 reduce transation costs, 222 RDBs’ headquarters, 90 RDBs’ mission, 9, 155 Rodrik, 117, 265 Russia, 37, 86, 111, 131, 136, 148, 214, 218, 242, 255 Sachs, 8, 30, 31, 265, 266 Sanford, 205, 215, 217, 220, 266 Schelling, 15, 266 Schneider, 47 Seventh Replenishment IDB, 124, 223 Sherk, 67, 225, 266 Sierra Leone, ix, 120, 127, 143, 251 Simmons and Hopkins, 154 Singapore, ix, 140, 143, 146, 246, 253 Sir Suma Chakrabarti, 96 Slovenia, ix, 85, 144, 246, 249, 255 Snidal, 42, 53, 259, 263, 266 Southeast Asia, 83, 101 sovereign loans, 18, 89, 240 Sovey and Green, 155 Standard & Poor’s, 18, 71, 72, 77, 78, 83, 84, 125, 211, 216, 266 Standard and Poor’s, 70, 229, 231, 232 State Department, 205, 268 state-led policies, 65 Stein, 42, 266 Stiglitz, 8, 58, 196, 230, 266 Stone, 24, 34, 111, 154, 196, 266 Strand, 8, 266 strategic lending, 15, 18, 19, 24, 57, 105, 110, 118, 140, 239, 240 Strezhnev and Voeten, 57 structural adjustment, 29, 102, 215 Sunkel, 100, 267 sustainable development, xiv, 6, 100, 243 Thacker, 24, 33, 34, 53, 54, 57, 58, 59, 60, 154, 196, 267 theories of aid, 53 Thomas Mirow, 96 trade openness, 116, 119, 166, 170 Tunisia, ix, 113, 127, 143, 144, 247, 251 Tussie, 64, 65, 73, 100, 101, 123, 221, 226, 267

Index UN Commission for Africa, 66 UN Economic Commission for Asia and the Far East, 67 Upton, xvii, 12, 198, 206, 209, 227, 267 US interest, 162, 215, 218 US Treasury, 19, 34, 40, 197, 206, 230, 234 USAID, 205, 220 Uzbekistan, ix, 85, 113, 144, 247, 253, 256 Vaubel, 48, 49, 58, 261, 267 Venezuela, 45, 73, 113, 163, 246, 249 Vietnam war, 83 Voeten, 116, 156, 259 vote shares change, 59

273

voting shares, 2 Vreeland, 24, 53, 196, 264, 267 Wade, 111, 206, 267 Waltz, 196, 267 Washington Consensus, 58, 149, 176, 196, 227, 230 wealth gap, 27 Weaver, 34, 58, 111, 196, 268 White, 52, 53, 64, 100, 220, 261, 268 Working Group on Multilateral Assistance, 220 World Bank, 3, 4, 5, 8, 10, 11, 14, 15, 18, 21, 22, 23, 25, 28, 30, 31, 34, 36, 38, 39, 45, 58, 64, 65, 69, 75, 78, 87, 101, 102, 103, 105, 108, 109, 111, 126, 146, 196, 206, 215, 223, 226, 228, 259, 260, 262, 263, 264, 265, 268 and RDBs, 16